Automatic segmentation of tumor plays a vital role in diagnosis and surgical planning. This paper deals
with techniques which providing solution for detecting hepatic tumor in Computed Tomography (CT)
images. The main aim of this work is to analyze performance of tumor detection techniques like Knowledge
Based Constraints, Graph Cut Method and Gradient Vector Flow active contour. These three techniquesare computed using sensitivity, specificity and accuracy. From the evaluated result, knowledge based constraints method is better than other graph cut method and gradient vector flow active contour.
Robust breathing signal extraction from cone beam CT projections based on ada...Wookjin Choi
This document summarizes a research paper that proposes a novel method for extracting breathing signals from cone beam CT projections without using external markers. The method uses an adaptive filtering technique to enhance weak oscillating structures in the Amsterdam Shroud image generated from the projections. A two-step optimization approach is then used to reveal the large-scale regularity of the breathing signals. Evaluation on 5 patient data sets found the new algorithm outperformed existing methods by extracting less noisy signals with errors of only -0.07±1.58 breaths per minute compared to reference signals. While results are promising, the study had a small data set and image quality remains limited.
ENHANCING SEGMENTATION APPROACHES FROM GAUSSIAN MIXTURE MODEL AND EXPECTED MA...Christo Ananth
Automatic liver tumor segmentation would bigly influence liver treatment organizing strategy and follow-up assessment, as a result of organization and joining of full picture information. Right now, develop a totally programmed technique for liver tumor division in CT picture. Introductory liver division comprises of applying a functioning form strategy. In the wake of separating liver applying Super pixel division Algorithm for portioning liver tumor proficiently. In the proposed work, we will investigate these procedures so as to improve division of various segments of the CT pictures. The exploratory outcomes indicated that the proposed strategy was exact for liver tumor division.
Neutrosophic sets and fuzzy c means clustering for improving ct liver image s...Aboul Ella Hassanien
The document proposes a hybrid method using neutrosophic sets and fuzzy c-means clustering to improve liver segmentation in CT images. It transforms the image into neutrosophic domains of truth, indeterminacy, and falsity. Thresholds are adapted using fuzzy c-means to binarize the domains. Experimental results on 30 abdominal CT images found 88% accuracy by Jaccard index and 94% by Dice coefficient, outperforming other methods. The approach effectively handles noise and uncertainty to produce clear liver boundaries.
Image processing in lung cancer screening and treatmentWookjin Choi
The document discusses image processing techniques for lung cancer screening and treatment. It covers topics like lung segmentation, nodule detection, computer-aided diagnosis, image-guided radiotherapy, and quantitative assessment of tumor response. Lung segmentation is used to isolate the lungs from other organs in CT images. Nodule detection algorithms then aim to find potential cancerous nodules. Computer-aided diagnosis systems analyze extracted features of nodules to determine if they are malignant or benign. Image-guided radiotherapy utilizes 4D CT and gating to account for tumor motion during treatment. Quantitative metrics like standardized uptake value are used to assess tumor response in PET imaging.
Computer aided detection of pulmonary nodules using genetic programmingWookjin Choi
This document describes a method for detecting pulmonary nodules in CT scans using genetic programming. It first segments the lung regions from CT images and extracts nodule candidates. Features are then extracted from the candidates. Genetic programming is used to classify candidates as nodules or non-nodules by optimizing combinations of features. The method was tested on a publicly available lung image database, achieving a true positive rate of over 90% and low false positive rate.
computer aided detection of pulmonary nodules in ct scansWookjin Choi
The document discusses computer aided detection of pulmonary nodules in CT scans. It introduces lung cancer as a major health problem and describes how detecting nodules early can improve survival rates. It then provides an overview of pulmonary nodule detection CAD systems, describing their general structure and evaluating various approaches in the literature. Key contributions are genetic programming and shape-based classifiers and a hierarchical block analysis method that achieved high performance on a publicly available lung image database.
automatic detection of pulmonary nodules in lung ct imagesWookjin Choi
The document discusses lung cancer detection using CT scans and pulmonary nodule detection systems. It describes how CT scans are used to detect lung nodules early and increase survival rates. It then discusses the challenges of evaluating large CT data sets and the use of pulmonary nodule detection CAD systems to assist radiologists. The document goes on to describe a proposed CAD system that includes lung segmentation, nodule candidate detection using multi-thresholding and feature extraction, and a genetic programming based classifier to analyze features and detect nodules. Experimental results on a publicly available lung image database show the system achieved over 80% accuracy on test data for nodule detection.
Robust breathing signal extraction from cone beam CT projections based on ada...Wookjin Choi
This document summarizes a research paper that proposes a novel method for extracting breathing signals from cone beam CT projections without using external markers. The method uses an adaptive filtering technique to enhance weak oscillating structures in the Amsterdam Shroud image generated from the projections. A two-step optimization approach is then used to reveal the large-scale regularity of the breathing signals. Evaluation on 5 patient data sets found the new algorithm outperformed existing methods by extracting less noisy signals with errors of only -0.07±1.58 breaths per minute compared to reference signals. While results are promising, the study had a small data set and image quality remains limited.
ENHANCING SEGMENTATION APPROACHES FROM GAUSSIAN MIXTURE MODEL AND EXPECTED MA...Christo Ananth
Automatic liver tumor segmentation would bigly influence liver treatment organizing strategy and follow-up assessment, as a result of organization and joining of full picture information. Right now, develop a totally programmed technique for liver tumor division in CT picture. Introductory liver division comprises of applying a functioning form strategy. In the wake of separating liver applying Super pixel division Algorithm for portioning liver tumor proficiently. In the proposed work, we will investigate these procedures so as to improve division of various segments of the CT pictures. The exploratory outcomes indicated that the proposed strategy was exact for liver tumor division.
Neutrosophic sets and fuzzy c means clustering for improving ct liver image s...Aboul Ella Hassanien
The document proposes a hybrid method using neutrosophic sets and fuzzy c-means clustering to improve liver segmentation in CT images. It transforms the image into neutrosophic domains of truth, indeterminacy, and falsity. Thresholds are adapted using fuzzy c-means to binarize the domains. Experimental results on 30 abdominal CT images found 88% accuracy by Jaccard index and 94% by Dice coefficient, outperforming other methods. The approach effectively handles noise and uncertainty to produce clear liver boundaries.
Image processing in lung cancer screening and treatmentWookjin Choi
The document discusses image processing techniques for lung cancer screening and treatment. It covers topics like lung segmentation, nodule detection, computer-aided diagnosis, image-guided radiotherapy, and quantitative assessment of tumor response. Lung segmentation is used to isolate the lungs from other organs in CT images. Nodule detection algorithms then aim to find potential cancerous nodules. Computer-aided diagnosis systems analyze extracted features of nodules to determine if they are malignant or benign. Image-guided radiotherapy utilizes 4D CT and gating to account for tumor motion during treatment. Quantitative metrics like standardized uptake value are used to assess tumor response in PET imaging.
Computer aided detection of pulmonary nodules using genetic programmingWookjin Choi
This document describes a method for detecting pulmonary nodules in CT scans using genetic programming. It first segments the lung regions from CT images and extracts nodule candidates. Features are then extracted from the candidates. Genetic programming is used to classify candidates as nodules or non-nodules by optimizing combinations of features. The method was tested on a publicly available lung image database, achieving a true positive rate of over 90% and low false positive rate.
computer aided detection of pulmonary nodules in ct scansWookjin Choi
The document discusses computer aided detection of pulmonary nodules in CT scans. It introduces lung cancer as a major health problem and describes how detecting nodules early can improve survival rates. It then provides an overview of pulmonary nodule detection CAD systems, describing their general structure and evaluating various approaches in the literature. Key contributions are genetic programming and shape-based classifiers and a hierarchical block analysis method that achieved high performance on a publicly available lung image database.
automatic detection of pulmonary nodules in lung ct imagesWookjin Choi
The document discusses lung cancer detection using CT scans and pulmonary nodule detection systems. It describes how CT scans are used to detect lung nodules early and increase survival rates. It then discusses the challenges of evaluating large CT data sets and the use of pulmonary nodule detection CAD systems to assist radiologists. The document goes on to describe a proposed CAD system that includes lung segmentation, nodule candidate detection using multi-thresholding and feature extraction, and a genetic programming based classifier to analyze features and detect nodules. Experimental results on a publicly available lung image database show the system achieved over 80% accuracy on test data for nodule detection.
MEDICAL IMAGE PROCESSING METHODOLOGY FOR LIVER TUMOUR DIAGNOSISijsc
Apply the Image processing techniques to analyse the medical images may assist medical professionals as well as patients, especially in this research apply the algorithms to diagnose the liver tumours from the abdominal CT image. This research proposes a software solution to illustrate the automated liver
segmentation and tumour detection using artificial intelligent techniques. Evaluate the results of the liver segmentation and tumour detection, in-cooperation with the radiologists by using the prototype of the proposed system. This research overcomes the challenges in medical image processing. The 100 samples
collected from ten patients and received 90% accuracy rate.
Image quality assessment of contrast-enhanced 4D-CT for pancreatic adenocarci...Wookjin Choi
Quantitative and qualitative assessment of the image qualities in contrast-enhanced (CE) 3D-CT, 4D-CT and CE 4D-CT to identify feasibility for replacing the current clinical standard simulation with a single CE 4D-CT for pancreatic adenocarcinoma (PDA) in radiotherapy simulation.
Individually Optimized Contrast-Enhanced 4D-CT for Radiotherapy Simulation in...Wookjin Choi
To develop an individually optimized contrast-enhanced (CE) 4D-CT for radiotherapy simulation in pancreatic ductal adenocarcinomas (PDA).
http://scitation.aip.org/content/aapm/journal/medphys/43/6/10.1118/1.4958261
Segmentation of cysts in kidney and 3 d volume calculation from ct images ijcga
This paper proposes a segmentation method and a three-dimensional (3-D) volume calculation method of
cysts in kidney from a number of computer tomography (CT) slice images. The input CT slice images
contain both sides of kidneys. There are two segmentation steps used in the proposed method: kidney
segmentation and cyst segmentation. For kidney segmentation, kidney regions are segmented from CT slice
images by using a graph-cut method that is applied to the middle slice of input CT slice images. Then, the
same method is used for the remaining CT slice images. In cyst segmentation, cyst regions are segmented
from the kidney regions by using fuzzy C-means clustering and level-set methods that can reduce noise of
non-cyst regions. For 3-D volume calculation, cyst volume calculation and 3-D volume visualization are
used. In cyst volume calculation, the area of cyst in each CT slice image equals to the number of pixels in
the cyst regions multiplied by spatial density of CT slice images, and then the volume of cysts is calculated
by multiplying the cyst area and thickness (interval) of CT slice images. In 3-D volume visualization, a 3-D
visualization technique is used to show the distribution of cysts in kidneys by using the result of cyst volume
calculation. The total 3-D volume is the sum of the calculated cyst volume in each CT slice image.
Experimental results show a good performance of 3-D volume calculation. The proposed cyst segmentation
and 3-D volume calculation methods can provide practical supports to surgery options and medical
practice to medical students
Segmentation of cysts in kidney and 3 d volume calculation from ct imagesbioejjournal
Statistics based optimization, Plackett–Burman design (PBD) and response surface methodology
(RSM) were employed to screen and optimize the media components for the production of
clavulanic acid from Streptomyces clavuligerus MTCC 1142, using solid state fermentation. jackfruit
seed powder was used as both the solid support and carbon source for the growth of Streptomyces
clavuligerus MTCC 1142. Based on the positive influence of the Pareto chart obtained from PBD on
clavulanic acid production, five media components – yeast extract, beef extract, sucrose, malt extract
and ferric chloride were screened. Central composite design (CCD) was employed using these five
media components- yeast extract 2.5%, beef extract 0.5%, sucrose 2.5%, malt extract 0.25% and ferric
chloride nutritional factors at three levels, for further optimization, and the second order polynomial
equation was derived, based on the experimental data. Response surface methodology showed that
the concentrations of yeast extract 2.5%, beef extract 0.5%, sucrose 2.5%, malt extract 0.25% and ferric
chloride 2.5% were the optimal levels for maximal clavulanic acid production (19.37 mg /gds) which were validated through experiments.
Computer-aided Detection of Pulmonary Nodules using Genetic ProgrammingWookjin Choi
This document describes a study that used genetic programming to develop an accurate classifier for detecting pulmonary nodules on CT scans. The proposed method involved segmenting the lungs, detecting nodule candidates, extracting features, and using genetic programming to evolve a combination of features and functions to classify nodules versus non-nodules. When tested on 153 nodules across 32 CT scans, the genetic programming classifier achieved a sensitivity of 92% and specificity of 86%.
Image segmentation is still an active reason of research, a relevant research area
in computer vision and hundreds of image segmentation techniques have been proposed by
the researchers. All proposed techniques have their own usability and accuracy. In this paper
we are going present a review of some best lung nodule existing detection and segmentation
techniques. Finally, we conclude by focusing one of the best methods that may have high
level accuracy and can be used in detection of lung very small nodules accurately.
Identification of Robust Normal Lung CT Texture FeaturesWookjin Choi
Normal lung CT texture features have been used for the prediction of radiation-induced lung disease (radiation pneumonitis and radiation fibrosis). For these features to be clinically useful, they need to be relatively invariant (robust) to tumor size and not correlated with normal lung volume.
http://scitation.aip.org/content/aapm/journal/medphys/43/6/10.1118/1.4955803
Individually Optimized Contrast-Enhanced 4D-CT for Radiotherapy Simulation in...Wookjin Choi
Purpose/Objectives: To develop an individually optimized contrast-enhanced (CE) 4D-CT for radiotherapy simulation in pancreatic adenocarcinoma (PDA).
Materials/Methods: Ten PDA patients were enrolled and underwent three CT scans: a 4D-CT immediately following a CE 3D-CT, and an individually optimized CE 4D-CT using a test injection to estimate the peak contrast enhancement time and to optimize the delay time. Three physicians contoured the tumor and pancreatic tissues. We compared image quality scores, tumor volume, motion, image noise, tumor-to-pancreas contrast, and contrast-to- noise ratio (CNR) in the three CTs. We also evaluated inter-observer variations in contouring the tumor using simultaneous truth and performance level estimation (STAPLE).
Results: The average image quality scores for CE 3D-CT and CE 4D-CT were comparable (4.0 and 3.8, p=0.47), and both were significantly better than that for 4D-CT (2.6, p<0.001). The tumor-to- pancreas contrast in CE 3D-CT and CE 4D-CT were comparable (15.5 and 16.7 HU, p=0.71), and the later was significantly higher than that in 4D-CT (9.2 HU, p=0.03). Image noise in CE 3D-CT (12.5 HU) was significantly lower than that in CE 4D-CT (22.1 HU, p<0.001) and 4D-CT (19.4 HU, p=0.005). The CNR in CE 3D-CT and CE 4D-CT were comparable (1.4 and 0.8, p=0.23), and the former was significantly better than that in 4D-CT (0.6, p=0.04). The average tumor volume was smaller in CE 3D-CT (29.8 cm 3 ) and CE 4D-CT (22.8 cm 3 ) than in 4D-CT (42.0 cm 3 ), though the differences were not statistically significant. The tumor motion was comparable in 4D-CT and CE 4D-CT (7.2 and 6.2 mm, p=0.23). The inter-observer variations were comparable in CE 3D-CT and CE 4D-CT (Jaccard index 66.0% and 61.9%), and the former was significantly smaller than that of 4D-CT (55.6%, p=0.047).
Conclusions: The CE 4D-CT demonstrated largely comparable characteristics to the CE 3D-CT. It has high potential for simultaneously delineating the tumor and quantifying the tumor motion with a single scan.
The International Journal of Engineering and Science (The IJES)theijes
This document summarizes a research paper on developing a computer-aided diagnosis system for early detection of liver cancer from CT chest images. The proposed system involves extracting features from segmented liver regions of CT images using techniques like noise removal, segmentation, and morphological operations. Features are then extracted and can be classified using Hidden Markov Models to identify liver cancer cells at an early stage and improve diagnosis. The authors suggest future work to refine cancer cell classification and reduce time complexity for improved diagnosis confidence.
A New Approach to the Detection of Mammogram Boundary IJECEIAES
Mammography is a method used for the detection of breast cancer. computer-aided diagnostic (CAD) systems help the radiologist in the detection and interpretation of mass in breast mammography. One of the important information of a mass is its contour and its form because it provides valuable information about the abnormality of a mass. The accuracy in the recognition of the shape of a mass is related to the accuracy of the detected mass contours. In this work we propose a new approach for detecting the boundaries of lesion in mammography images based on region growing algorithm without using the threshold, the proposed method requires an initial rectangle surrounding the lesion selected manually by the radiologist (Region Of Interest), where the region growing algorithm applies on lines segments that attach each pixel of this rectangle with the seed point, such as the ends (seeds) of each line segment grow in a direction towards one another. The proposed approach is evaluated on a set of data with 20 masses of the MIAS base whose contours are annotated manually by expert radiologists. The performance of the method is evaluated in terms of specificity, sensitivity, accuracy and overlap. All the findings and details of approach are presented in detail.
Optimal fuzzy rule based pulmonary nodule detectionWookjin Choi
The document describes a lung cancer detection system that uses CT scans. It discusses (1) segmenting the lungs from CT images using adaptive thresholding and connected component analysis, (2) detecting nodule candidate regions using multi-thresholding and rule-based pruning, and (3) optimizing the rule-based pruning using a genetic algorithm trained fuzzy inference system to reduce false positives while maintaining high sensitivity. Experimental results on a publicly available lung image database show the optimized fuzzy system achieved better performance than a conventional rule-based approach.
Quantitative Image Feature Analysis of Multiphase Liver CT for Hepatocellular...Wookjin Choi
To identify the effective quantitative image features (radiomics features) for prediction of response, survival, recurrence and metastasis of hepatocellular carcinoma (HCC) in radiotherapy.
A magnetic resonance spectroscopy driven initialization scheme for active sha...TRS Telehealth Services
Segmentation of the prostate boundary on clinical images is useful in a large number of applications including cal
culation of prostate volume pre- and post-treatment, to detect extra-capsular spread, and for creating patient-specific
anatomical models. Manual segmentation of the prostate boundary is, however, time consuming and subject to inter
and intra-reader variability.
A Magnetic Resonance Spectroscopy Driven Initialization Scheme
for Active Shape Model Based Prostate Segmentation.
Robert Toth1, Pallavi Tiwari1, Mark Rosen2, Galen Reed3, John Kurhanewicz3,
Arjun Kalyanpur4, Sona Pungavkar5, and Anant Madabhushi1
1Rutgers, The State University of New Jersey,
Department of Biomedical Engineering, Piscataway, NJ 08854, USA.
2 University of Pennsylvania,
Department of Radiology, Philadelphia, PA 19104, USA.
3 University of California,
San Francisco, CA, USA.
4 Teleradiology Solutions,
Bangalore, 560048, India.
5 Dr. Balabhai Nanavati Hospital,
Mumbai, 400056, India.
Radiomics and Deep Learning for Lung Cancer ScreeningWookjin Choi
The document summarizes research on using radiomics and deep learning approaches for lung cancer screening. It describes:
1) Using radiomic features like shape, texture, and intensity from lung nodules on CT scans and an SVM-LASSO model to classify nodules with 87.9% sensitivity and 78.2% specificity, outperforming the Lung-RADS system.
2) A deep learning model developed for a Kaggle competition that achieved 67.4% accuracy on nodule classification but only ranked 99th due to overfitting issues without enough data.
3) Future work could integrate quantification of nodule characteristics like spiculation with plasma biomarkers to improve diagnostic accuracy.
A survey on enhancing mammogram image saradha arumugam academiaPunit Karnani
This document summarizes research on enhancing mammogram images to improve the detection of breast cancer. It discusses how mammogram images have low contrast and are noisy, making it difficult to identify microcalcifications that could indicate cancer. Various image enhancement techniques are reviewed that aim to improve contrast, reduce noise, and sharpen edges to make microcalcifications more visible. The techniques discussed include nonlinear unsharp masking, wavelet-based enhancement, adaptive contrast enhancement, and integrated wavelet decompositions. Evaluation of the techniques suggests they can improve cancer diagnosis by enhancing image details and increasing radiologist performance.
Spams are unwanted and also undesirable emails which are mass sent to the numerous victims. Further
penetration of spams into electronic processors and communication equipments such as computers and
mobiles as well as lack of control on the information shared on the internet and other communication
networks and also inefficiency of the spam detecting methods developed for Persian contexts are among the
main challenging issues of the Persian subscribers. This paper presents a novel and efficient method for
thematic identification of Persian spams. The proposed method is capable of identifying the Persian, spams
and also “Penglish” spams. “Penglish” is made up of two words Persian and English and demonstrates a
Persian text which is written by English alphabetic letters. Based on the experimental analysis of the 10000
spams of different type the efficiency of the proposed method is evaluated to be more than 98%. The
presented method is also capable of updating its databases taking the advantage of the feedbacks received
from the users.
With the surge in modern research focus towards Pervasive Computing, lot of techniques and challenges
needs to be addressed so as to effectively create smart spaces and achieve miniaturization. In the process of
scaling down to compact devices, the real things to ponder upon are the Information Retrieval challenges.
In this work, we discuss the aspects of multimedia which makes information access challenging. An
Example Pattern Recognition scenario is presented and the mathematical techniques that can be used to
model uncertainty are also presented for developing a system that can sense, compute and communicate in
a way that can make human life easy with smart objects assisting from around his surroundings.
Regularized Weighted Ensemble of Deep Classifiers ijcsa
Ensemble of classifiers increases the performance of the classification since the decision of many experts
are fused together to generate the resultant decision for prediction making. Deep learning is a classification algorithm where along with the basic learning technique, fine tuning learning is done for improved precision of learning. Deep classifier ensemble learning is having a good scope of research.Feature subset selection is another for creating individual classifiers to be fused for ensemble learning. All these ensemble techniques faces ill posed problem of overfitting. Regularized weighted ensemble of deep support vector machine performs the prediction analysis on the three UCI repository problems IRIS,Ionosphere and Seed data set, thereby increasing the generalization of the boundary plot between the
classes of the data set. The singular value decomposition reduced norm 2 regularization with the two level
deep classifier ensemble gives the best result in our experiments.
Decreasing of quantity of radiation de fects inijcsa
Recently we introduced an approach to increase sharpness of diffusion-junction and implanted-junction
heterorectifiers. The heterorectifiers could by single and as a part of heterobipolar transistors. However
manufacturing p-n-junctions by ion implantation leads to generation of radiation defects in materials of
heterostructure. In this paper we introduce an approach to use an overlayer and optimization of annealing
of radiation defects to decrease quantity of radiation defects.
PERFORMANCE EVALUATION OF TUMOR DETECTION TECHNIQUES ijcsa
Automatic segmentation of tumor plays a vital role in diagnosis and surgical planning. This paper deals
with techniques which providing solution for detecting hepatic tumor in Computed Tomography (CT)
images. The main aim of this work is to analyze performance of tumor detection techniques like Knowledge
Based Constraints, Graph Cut Method and Gradient Vector Flow active contour. These three techniques
are computed using sensitivity, specificity and accuracy. From the evaluated result, knowledge based
constraints method is better than other graph cut method and gradient vector flow active contour.
MEDICAL IMAGE PROCESSING METHODOLOGY FOR LIVER TUMOUR DIAGNOSISijsc
Apply the Image processing techniques to analyse the medical images may assist medical professionals as well as patients, especially in this research apply the algorithms to diagnose the liver tumours from the abdominal CT image. This research proposes a software solution to illustrate the automated liver
segmentation and tumour detection using artificial intelligent techniques. Evaluate the results of the liver segmentation and tumour detection, in-cooperation with the radiologists by using the prototype of the proposed system. This research overcomes the challenges in medical image processing. The 100 samples
collected from ten patients and received 90% accuracy rate.
Image quality assessment of contrast-enhanced 4D-CT for pancreatic adenocarci...Wookjin Choi
Quantitative and qualitative assessment of the image qualities in contrast-enhanced (CE) 3D-CT, 4D-CT and CE 4D-CT to identify feasibility for replacing the current clinical standard simulation with a single CE 4D-CT for pancreatic adenocarcinoma (PDA) in radiotherapy simulation.
Individually Optimized Contrast-Enhanced 4D-CT for Radiotherapy Simulation in...Wookjin Choi
To develop an individually optimized contrast-enhanced (CE) 4D-CT for radiotherapy simulation in pancreatic ductal adenocarcinomas (PDA).
http://scitation.aip.org/content/aapm/journal/medphys/43/6/10.1118/1.4958261
Segmentation of cysts in kidney and 3 d volume calculation from ct images ijcga
This paper proposes a segmentation method and a three-dimensional (3-D) volume calculation method of
cysts in kidney from a number of computer tomography (CT) slice images. The input CT slice images
contain both sides of kidneys. There are two segmentation steps used in the proposed method: kidney
segmentation and cyst segmentation. For kidney segmentation, kidney regions are segmented from CT slice
images by using a graph-cut method that is applied to the middle slice of input CT slice images. Then, the
same method is used for the remaining CT slice images. In cyst segmentation, cyst regions are segmented
from the kidney regions by using fuzzy C-means clustering and level-set methods that can reduce noise of
non-cyst regions. For 3-D volume calculation, cyst volume calculation and 3-D volume visualization are
used. In cyst volume calculation, the area of cyst in each CT slice image equals to the number of pixels in
the cyst regions multiplied by spatial density of CT slice images, and then the volume of cysts is calculated
by multiplying the cyst area and thickness (interval) of CT slice images. In 3-D volume visualization, a 3-D
visualization technique is used to show the distribution of cysts in kidneys by using the result of cyst volume
calculation. The total 3-D volume is the sum of the calculated cyst volume in each CT slice image.
Experimental results show a good performance of 3-D volume calculation. The proposed cyst segmentation
and 3-D volume calculation methods can provide practical supports to surgery options and medical
practice to medical students
Segmentation of cysts in kidney and 3 d volume calculation from ct imagesbioejjournal
Statistics based optimization, Plackett–Burman design (PBD) and response surface methodology
(RSM) were employed to screen and optimize the media components for the production of
clavulanic acid from Streptomyces clavuligerus MTCC 1142, using solid state fermentation. jackfruit
seed powder was used as both the solid support and carbon source for the growth of Streptomyces
clavuligerus MTCC 1142. Based on the positive influence of the Pareto chart obtained from PBD on
clavulanic acid production, five media components – yeast extract, beef extract, sucrose, malt extract
and ferric chloride were screened. Central composite design (CCD) was employed using these five
media components- yeast extract 2.5%, beef extract 0.5%, sucrose 2.5%, malt extract 0.25% and ferric
chloride nutritional factors at three levels, for further optimization, and the second order polynomial
equation was derived, based on the experimental data. Response surface methodology showed that
the concentrations of yeast extract 2.5%, beef extract 0.5%, sucrose 2.5%, malt extract 0.25% and ferric
chloride 2.5% were the optimal levels for maximal clavulanic acid production (19.37 mg /gds) which were validated through experiments.
Computer-aided Detection of Pulmonary Nodules using Genetic ProgrammingWookjin Choi
This document describes a study that used genetic programming to develop an accurate classifier for detecting pulmonary nodules on CT scans. The proposed method involved segmenting the lungs, detecting nodule candidates, extracting features, and using genetic programming to evolve a combination of features and functions to classify nodules versus non-nodules. When tested on 153 nodules across 32 CT scans, the genetic programming classifier achieved a sensitivity of 92% and specificity of 86%.
Image segmentation is still an active reason of research, a relevant research area
in computer vision and hundreds of image segmentation techniques have been proposed by
the researchers. All proposed techniques have their own usability and accuracy. In this paper
we are going present a review of some best lung nodule existing detection and segmentation
techniques. Finally, we conclude by focusing one of the best methods that may have high
level accuracy and can be used in detection of lung very small nodules accurately.
Identification of Robust Normal Lung CT Texture FeaturesWookjin Choi
Normal lung CT texture features have been used for the prediction of radiation-induced lung disease (radiation pneumonitis and radiation fibrosis). For these features to be clinically useful, they need to be relatively invariant (robust) to tumor size and not correlated with normal lung volume.
http://scitation.aip.org/content/aapm/journal/medphys/43/6/10.1118/1.4955803
Individually Optimized Contrast-Enhanced 4D-CT for Radiotherapy Simulation in...Wookjin Choi
Purpose/Objectives: To develop an individually optimized contrast-enhanced (CE) 4D-CT for radiotherapy simulation in pancreatic adenocarcinoma (PDA).
Materials/Methods: Ten PDA patients were enrolled and underwent three CT scans: a 4D-CT immediately following a CE 3D-CT, and an individually optimized CE 4D-CT using a test injection to estimate the peak contrast enhancement time and to optimize the delay time. Three physicians contoured the tumor and pancreatic tissues. We compared image quality scores, tumor volume, motion, image noise, tumor-to-pancreas contrast, and contrast-to- noise ratio (CNR) in the three CTs. We also evaluated inter-observer variations in contouring the tumor using simultaneous truth and performance level estimation (STAPLE).
Results: The average image quality scores for CE 3D-CT and CE 4D-CT were comparable (4.0 and 3.8, p=0.47), and both were significantly better than that for 4D-CT (2.6, p<0.001). The tumor-to- pancreas contrast in CE 3D-CT and CE 4D-CT were comparable (15.5 and 16.7 HU, p=0.71), and the later was significantly higher than that in 4D-CT (9.2 HU, p=0.03). Image noise in CE 3D-CT (12.5 HU) was significantly lower than that in CE 4D-CT (22.1 HU, p<0.001) and 4D-CT (19.4 HU, p=0.005). The CNR in CE 3D-CT and CE 4D-CT were comparable (1.4 and 0.8, p=0.23), and the former was significantly better than that in 4D-CT (0.6, p=0.04). The average tumor volume was smaller in CE 3D-CT (29.8 cm 3 ) and CE 4D-CT (22.8 cm 3 ) than in 4D-CT (42.0 cm 3 ), though the differences were not statistically significant. The tumor motion was comparable in 4D-CT and CE 4D-CT (7.2 and 6.2 mm, p=0.23). The inter-observer variations were comparable in CE 3D-CT and CE 4D-CT (Jaccard index 66.0% and 61.9%), and the former was significantly smaller than that of 4D-CT (55.6%, p=0.047).
Conclusions: The CE 4D-CT demonstrated largely comparable characteristics to the CE 3D-CT. It has high potential for simultaneously delineating the tumor and quantifying the tumor motion with a single scan.
The International Journal of Engineering and Science (The IJES)theijes
This document summarizes a research paper on developing a computer-aided diagnosis system for early detection of liver cancer from CT chest images. The proposed system involves extracting features from segmented liver regions of CT images using techniques like noise removal, segmentation, and morphological operations. Features are then extracted and can be classified using Hidden Markov Models to identify liver cancer cells at an early stage and improve diagnosis. The authors suggest future work to refine cancer cell classification and reduce time complexity for improved diagnosis confidence.
A New Approach to the Detection of Mammogram Boundary IJECEIAES
Mammography is a method used for the detection of breast cancer. computer-aided diagnostic (CAD) systems help the radiologist in the detection and interpretation of mass in breast mammography. One of the important information of a mass is its contour and its form because it provides valuable information about the abnormality of a mass. The accuracy in the recognition of the shape of a mass is related to the accuracy of the detected mass contours. In this work we propose a new approach for detecting the boundaries of lesion in mammography images based on region growing algorithm without using the threshold, the proposed method requires an initial rectangle surrounding the lesion selected manually by the radiologist (Region Of Interest), where the region growing algorithm applies on lines segments that attach each pixel of this rectangle with the seed point, such as the ends (seeds) of each line segment grow in a direction towards one another. The proposed approach is evaluated on a set of data with 20 masses of the MIAS base whose contours are annotated manually by expert radiologists. The performance of the method is evaluated in terms of specificity, sensitivity, accuracy and overlap. All the findings and details of approach are presented in detail.
Optimal fuzzy rule based pulmonary nodule detectionWookjin Choi
The document describes a lung cancer detection system that uses CT scans. It discusses (1) segmenting the lungs from CT images using adaptive thresholding and connected component analysis, (2) detecting nodule candidate regions using multi-thresholding and rule-based pruning, and (3) optimizing the rule-based pruning using a genetic algorithm trained fuzzy inference system to reduce false positives while maintaining high sensitivity. Experimental results on a publicly available lung image database show the optimized fuzzy system achieved better performance than a conventional rule-based approach.
Quantitative Image Feature Analysis of Multiphase Liver CT for Hepatocellular...Wookjin Choi
To identify the effective quantitative image features (radiomics features) for prediction of response, survival, recurrence and metastasis of hepatocellular carcinoma (HCC) in radiotherapy.
A magnetic resonance spectroscopy driven initialization scheme for active sha...TRS Telehealth Services
Segmentation of the prostate boundary on clinical images is useful in a large number of applications including cal
culation of prostate volume pre- and post-treatment, to detect extra-capsular spread, and for creating patient-specific
anatomical models. Manual segmentation of the prostate boundary is, however, time consuming and subject to inter
and intra-reader variability.
A Magnetic Resonance Spectroscopy Driven Initialization Scheme
for Active Shape Model Based Prostate Segmentation.
Robert Toth1, Pallavi Tiwari1, Mark Rosen2, Galen Reed3, John Kurhanewicz3,
Arjun Kalyanpur4, Sona Pungavkar5, and Anant Madabhushi1
1Rutgers, The State University of New Jersey,
Department of Biomedical Engineering, Piscataway, NJ 08854, USA.
2 University of Pennsylvania,
Department of Radiology, Philadelphia, PA 19104, USA.
3 University of California,
San Francisco, CA, USA.
4 Teleradiology Solutions,
Bangalore, 560048, India.
5 Dr. Balabhai Nanavati Hospital,
Mumbai, 400056, India.
Radiomics and Deep Learning for Lung Cancer ScreeningWookjin Choi
The document summarizes research on using radiomics and deep learning approaches for lung cancer screening. It describes:
1) Using radiomic features like shape, texture, and intensity from lung nodules on CT scans and an SVM-LASSO model to classify nodules with 87.9% sensitivity and 78.2% specificity, outperforming the Lung-RADS system.
2) A deep learning model developed for a Kaggle competition that achieved 67.4% accuracy on nodule classification but only ranked 99th due to overfitting issues without enough data.
3) Future work could integrate quantification of nodule characteristics like spiculation with plasma biomarkers to improve diagnostic accuracy.
A survey on enhancing mammogram image saradha arumugam academiaPunit Karnani
This document summarizes research on enhancing mammogram images to improve the detection of breast cancer. It discusses how mammogram images have low contrast and are noisy, making it difficult to identify microcalcifications that could indicate cancer. Various image enhancement techniques are reviewed that aim to improve contrast, reduce noise, and sharpen edges to make microcalcifications more visible. The techniques discussed include nonlinear unsharp masking, wavelet-based enhancement, adaptive contrast enhancement, and integrated wavelet decompositions. Evaluation of the techniques suggests they can improve cancer diagnosis by enhancing image details and increasing radiologist performance.
Spams are unwanted and also undesirable emails which are mass sent to the numerous victims. Further
penetration of spams into electronic processors and communication equipments such as computers and
mobiles as well as lack of control on the information shared on the internet and other communication
networks and also inefficiency of the spam detecting methods developed for Persian contexts are among the
main challenging issues of the Persian subscribers. This paper presents a novel and efficient method for
thematic identification of Persian spams. The proposed method is capable of identifying the Persian, spams
and also “Penglish” spams. “Penglish” is made up of two words Persian and English and demonstrates a
Persian text which is written by English alphabetic letters. Based on the experimental analysis of the 10000
spams of different type the efficiency of the proposed method is evaluated to be more than 98%. The
presented method is also capable of updating its databases taking the advantage of the feedbacks received
from the users.
With the surge in modern research focus towards Pervasive Computing, lot of techniques and challenges
needs to be addressed so as to effectively create smart spaces and achieve miniaturization. In the process of
scaling down to compact devices, the real things to ponder upon are the Information Retrieval challenges.
In this work, we discuss the aspects of multimedia which makes information access challenging. An
Example Pattern Recognition scenario is presented and the mathematical techniques that can be used to
model uncertainty are also presented for developing a system that can sense, compute and communicate in
a way that can make human life easy with smart objects assisting from around his surroundings.
Regularized Weighted Ensemble of Deep Classifiers ijcsa
Ensemble of classifiers increases the performance of the classification since the decision of many experts
are fused together to generate the resultant decision for prediction making. Deep learning is a classification algorithm where along with the basic learning technique, fine tuning learning is done for improved precision of learning. Deep classifier ensemble learning is having a good scope of research.Feature subset selection is another for creating individual classifiers to be fused for ensemble learning. All these ensemble techniques faces ill posed problem of overfitting. Regularized weighted ensemble of deep support vector machine performs the prediction analysis on the three UCI repository problems IRIS,Ionosphere and Seed data set, thereby increasing the generalization of the boundary plot between the
classes of the data set. The singular value decomposition reduced norm 2 regularization with the two level
deep classifier ensemble gives the best result in our experiments.
Decreasing of quantity of radiation de fects inijcsa
Recently we introduced an approach to increase sharpness of diffusion-junction and implanted-junction
heterorectifiers. The heterorectifiers could by single and as a part of heterobipolar transistors. However
manufacturing p-n-junctions by ion implantation leads to generation of radiation defects in materials of
heterostructure. In this paper we introduce an approach to use an overlayer and optimization of annealing
of radiation defects to decrease quantity of radiation defects.
PERFORMANCE EVALUATION OF TUMOR DETECTION TECHNIQUES ijcsa
Automatic segmentation of tumor plays a vital role in diagnosis and surgical planning. This paper deals
with techniques which providing solution for detecting hepatic tumor in Computed Tomography (CT)
images. The main aim of this work is to analyze performance of tumor detection techniques like Knowledge
Based Constraints, Graph Cut Method and Gradient Vector Flow active contour. These three techniques
are computed using sensitivity, specificity and accuracy. From the evaluated result, knowledge based
constraints method is better than other graph cut method and gradient vector flow active contour.
IMPLEMENTATION OF SECURITY PROTOCOL FOR WIRELESS SENSORijcsa
Intrusion Detection is one of the methods of defending against these attacks. In the proposed a security protocol for homogeneous wireless sensor network; network with all nodes are of same type. Clustering is used to improve the energy efficiency. Zone-Based Cluster Protocol (ZBCA) is used for selection of cluster head which is effective in scalability and energy consumption. Single hop technique is used for
communication within normal nodes and cluster head to base station. Simulation of proposed algorithm is performed in MATLAB. Sleep Deprivation Attack has been analyzed where attacker changes the environmental values by an artificial event. Attacker produces an event in environment due to which nodes have to sense the environment more than once in the same round that increase the power consumption of
the node. This interrupt reduces the network life time as nodes are not allowed to go in sleep mode and they are not able to perform their function of data collection and reporting to Cluster head and Base Station properly. Proposed protocol identifies this attack and prevents it from happening by solating the attacker node.
Experimental analysis of channel interference in ad hoc networkijcsa
In recent times, the use of ad hoc networks is a common research area among a researcher. Designing an
efficient and reliable network is not easy task. Network engineer faces many problems at the time of
deploying a network such as interference; Signal coverage, proper location of access point etc. channel
interference in one of them which must be considered at the time of deploying WLAN indoor environments
because channel interference impacts the network throughput and degrade the network performance.
In this experiment, we design a two WLAN BSS1 and BSS2 and investigate the impact of interference on
nodes. BSS1 contains three FTP clients and BSS2 contains two FTP client and their jobs is to upload data
to FTP Server Initially, they are far from each other. BSS1 moves toward BSS2 and after some time at
particular position both BSSs overlaps to each other. When BSSs overlaps to each other interference is
high and decrease network performance and increase upload time.
A ROBUST APPROACH FOR DATA CLEANING USED BY DECISION TREEijcsa
Now a day’s every second trillion of bytes of data is being generated by enterprises especially in internet.To achieve the best decision for business profits, access to that data in a well-situated and interactive way is always a dream of business executives and managers. Data warehouse is the only viable solution that can bring the dream into veracity. The enhancement of future endeavours to make decisions depends on the availability of correct information that is based on quality of data underlying. The quality data can only be produced by cleaning data prior to loading into data warehouse since the data collected from different sources will be dirty. Once the data have been pre-processed and cleansed then it produces accurate results on applying the data mining query. Therefore the accuracy of data is vital for well-formed and reliable decision making. In this paper, we propose a framework which implements robust data quality to ensure consistent and correct loading of data into data warehouses which ensures accurate and reliable data analysis, data mining and knowledge discovery.
This work is proposed the feed forward neural network with symmetric table addition method to design the
neuron synapses algorithm of the sine function approximations, and according to the Taylor series
expansion. Matlab code and LabVIEW are used to build and create the neural network, which has been
designed and trained database set to improve its performance, and gets the best a global convergence with
small value of MSE errors and 97.22% accuracy.
IMPROVING PACKET DELIVERY RATIO WITH ENHANCED CONFIDENTIALITY IN MANETijcsa
In Mobile Ad Hoc Network (MANET), the collection of mobile nodes gets communicated without the need of any customary infrastructure. In MANET, repeated topology changes and intermittent link breakage
causes the failure of existing path. This leads to rediscovery of new route by broadcasting RREQ packet.The number of RREQ packet in the network gets added due to the increased amount of link failures. This result in increased routing overhead which degrades the packet delivery ratio in MANET. While designing
routing protocols for MANET, it is indispensable to reduce the overhead in route discovery. In our previous
work[17], routing protocol based on neighbour details and probabilistic knowledge is utilized, additionally
the symmetric cipher AES is used for securing the data packet. Through this protocol, packet delivery ratio
gets increased and confidentiality is ensured. But there is a problem in secure key exchange among the
source and destination while using AES. To resolve that problem, hybrid cryptographic system i.e.,
combination of AES and RSA is proposed in this paper. By using this hybrid cryptographic scheme and the
routing protocol based on probability and neighbour knowledge, enhanced secure packet delivery is
ensured in MANET
3 d single gaas co axial nanowire solar cell for nanopillar-array photovoltai...ijcsa
Nanopillar array photovoltaics give unique advantages over today’s planar thin films in the areas of
optical properties and carrier collection, arising from their 3D geometry. The choice of the material
system, however, is essential in order to gain the advantage of the large surface/interface area associated
with nanopillars. Therefore, a well known Si and GaAs material are used in the design and studied in this
nanowire application. This work calculates and analyses the performance of the coaxial GaAs nanowire
and compared with that of Si nanowire using a semi-classical method. The current-voltage characteristics
are investigated for both under dark and AM1.5G illumination. It is found that GaAs nanowire gives almost
double efficiency with its counterpart Si nanowire. Their TCAD simulations can be validated reasonably
with that of published experimental result.
A HYBRID COA-DEA METHOD FOR SOLVING MULTI-OBJECTIVE PROBLEMS ijcsa
The Cuckoo optimization algorithm (COA) is developed for solving single-objective problems and it cannot be used for solving multi-objective problems. So the multi-objective cuckoo optimization algorithm based on data envelopment analysis (DEA) is developed in this paper and it can gain the efficient Pareto frontiers. This algorithm is presented by the CCR model of DEA and the output-oriented approach of it.The selection criterion is higher efficiency for next iteration of the proposed hybrid method. So the profit function of the COA is replaced by the efficiency value that is obtained from DEA. This algorithm is
compared with other methods using some test problems. The results shows using COA and DEA approach for solving multi-objective problems increases the speed and the accuracy of the generated solutions.
Nowadays peoples are actively involved in giving comments and reviews on social networking websites
and other websites like shopping websites, news websites etc. large number of people everyday share
their opinion on the web, results is a large number of user data is collected .users also find it trivial task
to read all the reviews and then reached into the decision. It would be better if these reviews are
classified into some category so that the user finds it easier to read. Opinion Mining or Sentiment
Analysis is a natural language processing task that mines information from various text forms such as
reviews, news, and blogs and classify them on the basis of their polarity as positive, negative or neutral.
But, from the last few years, user content in Hindi language is also increasing at a rapid rate on the Web.
So it is very important to perform opinion mining in Hindi language as well. In this paper a Hindi
language opinion mining system is proposed. The system classifies the reviews as positive, negative and
neutral for Hindi language. Negation is also handled in the proposed system. Experimental results using
reviews of movies show the effectiveness of the system.
Basic survey on malware analysis, tools and techniquesijcsa
The term malware stands for malicious software. It is a program installed on a system without the
knowledge of owner of the system. It is basically installed by the third party with the intention to steal some
private data from the system or simply just to play pranks. This in turn threatens the computer’s security,
wherein computer are used by one’s in day-to-day life as to deal with various necessities like education,
communication, hospitals, banking, entertainment etc. Different traditional techniques are used to detect
and defend these malwares like Antivirus Scanner (AVS), firewalls, etc. But today malware writers are one
step forward towards then Malware detectors. Day-by-day they write new malwares, which become a great
challenge for malware detectors. This paper focuses on basis study of malwares and various detection
techniques which can be used to detect malwares.
Artificial neural network approach for more accurate solar cell electrical ci...ijcsa
The implementation of a neural network especially for improving the accuracy of the electrical equivalent
circuit parameters of a solar cell is proposed. These electrical parameters mainly depend on solar
irradiation and temperature, but their relationship is nonlinear and cannot be easily expressed by any
analytical equation. Therefore, the proposed neural network is trained once by using some measured
current–voltage curves, and the equivalent circuit parameters are estimated by only reading the samples of
solar irradiation and temperature very quickly. Taking the effect of sunlight irradiance and ambient
temperature into consideration, the output current and power characteristics of PV model are simulated
and optimized. Finally, the proposed model has been validated with datasheet and experimental data from
commercial PV module, Kotak PV-KM0060 (60Wp).The comparison show the higher accuracy of the ANN
model than the conventional one diode circuit model for all operating conditions.
Crime and violence are inherent in our political and social system. With the moving pace of technology, the
popularity of internet grows continuously, with not only changing our views of life, but also changing the
way crime takes place all over the world. We need a technology that can be used to bring justice to those
who are responsible for conducting attacks on computer systems across the globe. In this paper, we present
various measures being taken in order to control and deal with the crime related to digital devices. This
paper gives an insight of Digital Forensics and current situation of India in handling such type of crimes.
A NEW IMPROVED QUANTUM EVOLUTIONARY ALGORITHM WITH MULTIPLICATIVE UPDATE FUNC...ijcsa
The Quantum Evolutionary Algorithm (QEA) is a new subcategory of evolutionary computation in which
the principles and concepts of quantum computation are used, and to display the solutions it utilizes a
probabilistic structure.Therefore, it causes an increase in the solution space. This algorithm has two major
problems: hitchhiking phenomenon and slow convergence speed.In this paper, to solve the problems a
multiplicative update function called quantum gate is proposed that in addition to considering the best
global solution ، considers the best solution of each generation. The results of one max and knapsack
problems and five famous numerical functions show that the proposed method has a significant advantage
compared with the basic algorithm in terms of performance, quality of solutions and convergence speed.
MODELLING FOR CROSS IGNITION TIME OF A TURBULENT COLD MIXTURE IN A MULTI BURN...ijcsa
This document presents a computational model for determining the cross-ignition time of a turbulent cold mixture in a multi-burner combustor. The model accounts for various parameters that influence heat transfer during the cross-ignition process. Experimental validation was conducted using a simple test rig with two burners. Preliminary results from the experiments show relationships between cross-ignition time and factors like the flow area between burners, distance between burners, and flame properties. Computational fluid dynamics simulations will also be used to further investigate heat transfer in high-velocity regions and validate the model under more conditions.
COMPARISON AND EVALUATION DATA MINING TECHNIQUES IN THE DIAGNOSIS OF HEART DI...ijcsa
Heart disease is one of the biggest health problems in the world because of high mortality and morbidity
caused by the disease. The use of data mining on medical data brought valuable and effective life
achievements and can enhance medical knowledge to make necessary decisions. Data mining plays an
important role in the field of medical science to solve health problems and diagnose ailments in critical
conditions and in normal conditions. For this reason, in this paper, data mining techniques are used to
diagnose heart disease from a dataset that includes 200 samples from different patients. Techniques used to
diagnose heart disease include Bagging, AdaBoostM1, Random Forest, Naive Bayes, RBF Network, IBK,
and NNge that all the techniques used to diagnose heart disease use Weka tool. Then these techniques are
compared to determine which is more accurate in the diagnosis of heart disease that according to the
results, it was found that the RBF Network with the accuracy of 88.2% is the most accurate classification in
the diagnosis of heart disease.
The i.sawan Residential Spa and Club at the Grand Hyatt Erawan in Thailand offers an ultra-luxurious spa experience. It has over 7,000 square meters of indoor and outdoor facilities including treatment rooms, a pool, and six guest cottages. The cottages provide a luxurious setting with amenities like high-speed internet and three TVs. Guests can receive treatments without leaving their cottage, and enjoy meals delivered to their room. The spa aims to provide highly personalized care and attention to guests to create a world-class spa experience.
Terrain generation finds many applications such as
in CGI movies, animations and video games. This
paper describes a new and simple-to-implement terra
in generator called the Uplift Model. It is based o
n
the theory of crustal deformations by uplifts in Ge
ology. When a number of uplifts are made on the Ear
th’s
surface, the final net effect is an average of the
influence of each uplift at each point on the terra
in. The
result of applying this model from Nature is a very
realistic looking effect in the generated terrains
. The
model uses 6 parameters which allow for a great var
iety in landscape types produced. Comparisons are
made with other existing terrain generation algorit
hms. Averaging causes erosion of the surface wherea
s
fractal surfaces tend to be very jagged and more su
ited to alien worlds.
Reinforcing optimization enabled interactive approach for liver tumor extrac...IJECEIAES
This document presents a method for liver tumor extraction from computed tomography (CT) images using an optimization-based approach. It first preprocesses the CT images using median filtering to reduce noise. It then segments the images using a fuzzy c-means clustering algorithm. To improve segmentation accuracy, it incorporates a grey wolf optimization metaheuristic to determine the optimal clustering threshold. Experimental results on both public and simulated datasets show the method can extract liver tumors while minimizing user input, outperforming other state-of-the-art segmentation algorithms.
The International Journal of Engineering and Science (The IJES)theijes
The International Journal of Engineering & Science is aimed at providing a platform for researchers, engineers, scientists, or educators to publish their original research results, to exchange new ideas, to disseminate information in innovative designs, engineering experiences and technological skills. It is also the Journal's objective to promote engineering and technology education. All papers submitted to the Journal will be blind peer-reviewed. Only original articles will be published.
The papers for publication in The International Journal of Engineering& Science are selected through rigorous peer reviews to ensure originality, timeliness, relevance, and readability.
IJRET : International Journal of Research in Engineering and Technology is an international peer reviewed, online journal published by eSAT Publishing House for the enhancement of research in various disciplines of Engineering and Technology. The aim and scope of the journal is to provide an academic medium and an important reference for the advancement and dissemination of research results that support high-level learning, teaching and research in the fields of Engineering and Technology. We bring together Scientists, Academician, Field Engineers, Scholars and Students of related fields of Engineering and Technology
Comparative analysis of edge based and region based active contour using leve...eSAT Publishing House
IJRET : International Journal of Research in Engineering and Technology is an international peer reviewed, online journal published by eSAT Publishing House for the enhancement of research in various disciplines of Engineering and Technology. The aim and scope of the journal is to provide an academic medium and an important reference for the advancement and dissemination of research results that support high-level learning, teaching and research in the fields of Engineering and Technology. We bring together Scientists, Academician, Field Engineers, Scholars and Students of related fields of Engineering and Technology.
Comparative analysis of edge based and region based active contour using leve...eSAT Journals
This document compares edge-based and region-based active contour segmentation methods using level sets on CT images of liver cancer. Edge-based methods use gradient information to stop contours at edges, while region-based methods use average intensity inside and outside contours. The document applies both methods to 10 CT images and evaluates them quantitatively using correlation, variation of information, global consistency error, and Rand index. Results show edge-based segmentation with distance regularized level sets performs better than region-based at locating cancer regions in images with indistinct edges.
An approach for cross-modality guided quality enhancement of liver imageIJECEIAES
This paper proposes a novel approach for enhancing the contrast of CT liver images using MRI liver images as a guide. The approach has three steps: 1) transforming the CT and MRI images to the same range, 2) adjusting the CT histogram to match the MRI histogram, and 3) applying adaptive histogram equalization to the CT image using two sigmoid functions. Experimental results show that the proposed approach improves image quality metrics like contrast and brightness preservation compared to traditional techniques. Both subjective and objective assessments indicate the approach enhances image contrast while preserving mean brightness and details.
BATCH NORMALIZED CONVOLUTION NEURAL NETWORK FOR LIVER SEGMENTATIONsipij
With the huge innovative improvement in all lifestyles, it has been important to build up the clinical fields, remembering the finding for which treatment is done; where the fruitful treatment relies upon the preoperative. Models for the preoperative, for example, planning to understand the complex internal structure of the liver and precisely localize the liver surface and its tumors; there are various algorithms proposed to do the automatic liver segmentation. In this paper, we propose a Batch Normalization After All - Convolutional Neural Network (BATA-Convnet) model to segment the liver CT images using Deep Learning Technique. The proposed liver segmentation model consists of four main steps: pre-processing, training the BATA-Convnet, liver segmentation, and the postprocessing step to maximize the result efficiency. Medical Image Computing and Computer Assisted Intervention (MICCAI) dataset and 3DImage Reconstruction for Comparison of Algorithm Database (3D-RCAD) were used in the experimentation and the average results using MICCAI are 0.91% for Dice, 13.44% for VOE, 0.23% for RVD, 0.29mm for ASD, 1.35mm for RMSSD and 0.36mm for MaxASD. The average results using 3DIRCAD dataset are 0.84% for Dice, 13.24% for VOE, 0.16% for RVD, 0.32mm for ASD, 1.17mm for RMSSD and 0.33mm for MaxASD.
Batch Normalized Convolution Neural Network for Liver Segmentationsipij
With the huge innovative improvement in all lifestyles, it has been important to build up the clinical fields, remembering the finding for which treatment is done; where the fruitful treatment relies upon the
preoperative. Models for the preoperative, for example, planning to understand the complex internal structure of the liver and precisely localize the liver surface and its tumors; there are various algorithms proposed to do the automatic liver segmentation. In this paper, we propose a Batch Normalization After All - Convolutional Neural Network (BATA-Convnet) model to segment the liver CT images using Deep
Learning Technique. The proposed liver segmentation model consists of four main steps: pre-processing, training the BATA-Convnet, liver segmentation, and the postprocessing step to maximize the result
efficiency. Medical Image Computing and Computer Assisted Intervention (MICCAI) dataset and 3DImage Reconstruction for Comparison of Algorithm Database (3D-IRCAD) were used in the experimentation and the average results using MICCAI are 0.91% for Dice, 13.44% for VOE, 0.23% for RVD, 0.29mm for ASD, 1.35mm for RMSSD and 0.36mm for MaxASD. The average results using 3DIRCAD dataset are 0.84% for Dice, 13.24% for VOE, 0.16% for RVD, 0.32mm for ASD, 1.17mm for
RMSSD and 0.33mm for MaxASD.
Medical Image Processing Methodology for Liver Tumour Diagnosis ijsc
Apply the Image processing techniques to analyse the medical images may assist medical professionals as well as patients, especially in this research apply the algorithms to diagnose the liver tumours from the abdominal CT image. This research proposes a software solution to illustrate the automated liver segmentation and tumour detection using artificial intelligent techniques. Evaluate the results of the liver segmentation and tumour detection, in-cooperation with the radiologists by using the prototype of the proposed system. This research overcomes the challenges in medical image processing. The 100 samples collected from ten patients and received 90% accuracy rate.
Brain Tumor Area Calculation in CT-scan image using Morphological Operationsiosrjce
IOSR Journal of Computer Engineering (IOSR-JCE) is a double blind peer reviewed International Journal that provides rapid publication (within a month) of articles in all areas of computer engineering and its applications. The journal welcomes publications of high quality papers on theoretical developments and practical applications in computer technology. Original research papers, state-of-the-art reviews, and high quality technical notes are invited for publications.
This document presents a method for calculating the area of brain tumors in CT scan images using morphological operations. The proposed method involves 10 steps: 1) inputting a CT scan image, 2) cropping the image, 3) converting to grayscale, 4) applying morphological gradient filtering, 5) histogram equalization, 6) selecting the region of interest, 7) subtracting and thresholding, 8) morphological closing, and 9) calculating the tumor area. The method is tested on different tumor images and more accurately calculates tumor area compared to radiologists who assume tumor shapes. The algorithm provides automated tumor highlighting and area calculation to assist physicians.
This document proposes a computer-aided lung cancer classification system using curvelet features and an ensemble classifier. It first pre-processes CT images using adaptive histogram equalization to improve contrast. Then it segments the images using kernelized fuzzy c-means clustering. Curvelet features are extracted from the segmented regions and an ensemble classifier is applied to classify regions as benign or malignant. The proposed approach achieves reliable and accurate classification results compared to existing methods, with better performance metrics like accuracy, sensitivity and specificity.
IRJET- Image Processing based Lung Tumor Detection System for CT ImagesIRJET Journal
This document presents a method for detecting lung tumors in CT scan images using image processing techniques. The proposed method involves preprocessing images using median filtering for noise removal and contrast adjustment for enhancement. The lungs are then segmented from the images using mathematical morphology. Geometric and textural features are extracted from the segmented region of interest and used as input for an SVM classifier to detect lung cancer. The methodology was tested on a dataset from The Cancer Imaging Archive and was able to successfully detect lung tumors in CT images.
Prediction of Lung Cancer Using Image Processing Techniques: A Reviewaciijournal
The document discusses previous research on predicting lung cancer using image processing techniques. Various studies are reviewed that used techniques like segmentation, feature extraction and classification on CT scan images to detect lung cancer. Classification approaches discussed include support vector machines, neural networks, fuzzy logic and genetic algorithms. Accuracy of prediction ranged from 80-99% depending on the techniques and image datasets used. The summary highlights several studies that applied methods like segmentation, feature extraction and neural network or SVM classification to CT images to detect lung nodules and predict cancer.
Prediction of Lung Cancer Using Image Processing Techniques: A Reviewaciijournal
Prediction of lung cancer is most challenging problem due to structure of cancer cell, where most of the
cells are overlapped each other. The image processing techniques are mostly used for prediction of lung
cancer and also for early detection and treatment to prevent the lung cancer. To predict the lung cancer
various features are extracted from the images therefore, pattern recognition based approaches are useful
to predict the lung cancer. Here, a comprehensive review for the prediction of lung cancer by previous
researcher using image processing techniques is presented. The summary for the prediction of lung cancer
by previous researcher using image processing techniques is also presented.
Prediction of lung cancer is most challenging problem due to structure of cancer cell, where most of the cells are overlapped each other. The image processing techniques are mostly used for prediction of lung cancer and also for early detection and treatment to prevent the lung cancer. To predict the lung cancer various features are extracted from the images therefore, pattern recognition based approaches are useful to predict the lung cancer. Here, a comprehensive review for the prediction of lung cancer by previous researcher using image processing techniques is presented. The summary for the prediction of lung cancer by previous researcher using image processing techniques is also presented.
A novel CAD system to automatically detect cancerous lung nodules using wav...IJECEIAES
A novel cancerous nodules detection algorithm for computed tomography images (CT-images) is presented in this paper. CT-images are large size images with high resolution. In some cases, number of cancerous lung nodule lesions may missed by the radiologist due to fatigue. A CAD system that is proposed in this paper can help the radiologist in detecting cancerous nodules in CT- images. The proposed algorithm is divided to four stages. In the first stage, an enhancement algorithm is implement to highlight the suspicious regions. Then in the second stage, the region of interest will be detected. The adaptive SVM and wavelet transform techniques are used to reduce the detected false positive regions. This algorithm is evaluated using 60 cases (normal and cancerous cases), and it shows a high sensitivity in detecting the cancerous lung nodules with TP ration 94.5% and with FP ratio 7 cluster/image.
Enhancing Segmentation Approaches from Super Pixel Division Algorithm to Hidd...Christo Ananth
Christo Ananth, S. Amutha, K. Niha, Djabbarov Botirjon Begimovich, “Enhancing Segmentation Approaches from Super Pixel Division Algorithm to Hidden Markov Random Fields with Expectation Maximization (HMRF-EM)”, International Journal of Early Childhood Special Education, Volume 14, Issue 05, 2022,pp. 2400-2410.
Christo Ananth et al. discussed that In surgical planning and cancer treatment, it is crucial to segment and measure a liver tumor's volume accurately. Because it would involve automation, standardisation, and the incorporation of complete volumetric information, accurate automatic liver tumor segmentation would substantially affect the processes for therapy planning and follow-up reporting. Based on the Hidden Markov random field, Automatic liver tumor detection in CT scans is possible using hidden Markov random fields (HMRF-EM). A CT scan of the liver may be too low-resolution for this software. CT liver tissue segmentation is based on the HMRF model. When building an accurate HMRF model, an accurate initial image estimate is crucial. Adaptive K-means clustering is often used for initial estimation. HMRF's performance can be greatly improved by clustering. This project aims to segment liver tissue quickly. This paper proposes an adaptive K-means clustering approach for estimating liver images in the HMRF-EM model. The previous strategy had flaws, so this one fixed them. We compare the current and proposed methods. The proposed method outperforms the currently used method.
International Journal of Computational Engineering Research (IJCER) is dedicated to protecting personal information and will make every reasonable effort to handle collected information appropriately. All information collected, as well as related requests, will be handled as carefully and efficiently as possible in accordance with IJCER standards for integrity and objectivity.
Similar to PERFORMANCE EVALUATION OF TUMOR DETECTION TECHNIQUES (20)
Using recycled concrete aggregates (RCA) for pavements is crucial to achieving sustainability. Implementing RCA for new pavement can minimize carbon footprint, conserve natural resources, reduce harmful emissions, and lower life cycle costs. Compared to natural aggregate (NA), RCA pavement has fewer comprehensive studies and sustainability assessments.
Embedded machine learning-based road conditions and driving behavior monitoringIJECEIAES
Car accident rates have increased in recent years, resulting in losses in human lives, properties, and other financial costs. An embedded machine learning-based system is developed to address this critical issue. The system can monitor road conditions, detect driving patterns, and identify aggressive driving behaviors. The system is based on neural networks trained on a comprehensive dataset of driving events, driving styles, and road conditions. The system effectively detects potential risks and helps mitigate the frequency and impact of accidents. The primary goal is to ensure the safety of drivers and vehicles. Collecting data involved gathering information on three key road events: normal street and normal drive, speed bumps, circular yellow speed bumps, and three aggressive driving actions: sudden start, sudden stop, and sudden entry. The gathered data is processed and analyzed using a machine learning system designed for limited power and memory devices. The developed system resulted in 91.9% accuracy, 93.6% precision, and 92% recall. The achieved inference time on an Arduino Nano 33 BLE Sense with a 32-bit CPU running at 64 MHz is 34 ms and requires 2.6 kB peak RAM and 139.9 kB program flash memory, making it suitable for resource-constrained embedded systems.
Low power architecture of logic gates using adiabatic techniquesnooriasukmaningtyas
The growing significance of portable systems to limit power consumption in ultra-large-scale-integration chips of very high density, has recently led to rapid and inventive progresses in low-power design. The most effective technique is adiabatic logic circuit design in energy-efficient hardware. This paper presents two adiabatic approaches for the design of low power circuits, modified positive feedback adiabatic logic (modified PFAL) and the other is direct current diode based positive feedback adiabatic logic (DC-DB PFAL). Logic gates are the preliminary components in any digital circuit design. By improving the performance of basic gates, one can improvise the whole system performance. In this paper proposed circuit design of the low power architecture of OR/NOR, AND/NAND, and XOR/XNOR gates are presented using the said approaches and their results are analyzed for powerdissipation, delay, power-delay-product and rise time and compared with the other adiabatic techniques along with the conventional complementary metal oxide semiconductor (CMOS) designs reported in the literature. It has been found that the designs with DC-DB PFAL technique outperform with the percentage improvement of 65% for NOR gate and 7% for NAND gate and 34% for XNOR gate over the modified PFAL techniques at 10 MHz respectively.
Literature Review Basics and Understanding Reference Management.pptxDr Ramhari Poudyal
Three-day training on academic research focuses on analytical tools at United Technical College, supported by the University Grant Commission, Nepal. 24-26 May 2024
A review on techniques and modelling methodologies used for checking electrom...nooriasukmaningtyas
The proper function of the integrated circuit (IC) in an inhibiting electromagnetic environment has always been a serious concern throughout the decades of revolution in the world of electronics, from disjunct devices to today’s integrated circuit technology, where billions of transistors are combined on a single chip. The automotive industry and smart vehicles in particular, are confronting design issues such as being prone to electromagnetic interference (EMI). Electronic control devices calculate incorrect outputs because of EMI and sensors give misleading values which can prove fatal in case of automotives. In this paper, the authors have non exhaustively tried to review research work concerned with the investigation of EMI in ICs and prediction of this EMI using various modelling methodologies and measurement setups.
Understanding Inductive Bias in Machine LearningSUTEJAS
This presentation explores the concept of inductive bias in machine learning. It explains how algorithms come with built-in assumptions and preferences that guide the learning process. You'll learn about the different types of inductive bias and how they can impact the performance and generalizability of machine learning models.
The presentation also covers the positive and negative aspects of inductive bias, along with strategies for mitigating potential drawbacks. We'll explore examples of how bias manifests in algorithms like neural networks and decision trees.
By understanding inductive bias, you can gain valuable insights into how machine learning models work and make informed decisions when building and deploying them.
DEEP LEARNING FOR SMART GRID INTRUSION DETECTION: A HYBRID CNN-LSTM-BASED MODELgerogepatton
As digital technology becomes more deeply embedded in power systems, protecting the communication
networks of Smart Grids (SG) has emerged as a critical concern. Distributed Network Protocol 3 (DNP3)
represents a multi-tiered application layer protocol extensively utilized in Supervisory Control and Data
Acquisition (SCADA)-based smart grids to facilitate real-time data gathering and control functionalities.
Robust Intrusion Detection Systems (IDS) are necessary for early threat detection and mitigation because
of the interconnection of these networks, which makes them vulnerable to a variety of cyberattacks. To
solve this issue, this paper develops a hybrid Deep Learning (DL) model specifically designed for intrusion
detection in smart grids. The proposed approach is a combination of the Convolutional Neural Network
(CNN) and the Long-Short-Term Memory algorithms (LSTM). We employed a recent intrusion detection
dataset (DNP3), which focuses on unauthorized commands and Denial of Service (DoS) cyberattacks, to
train and test our model. The results of our experiments show that our CNN-LSTM method is much better
at finding smart grid intrusions than other deep learning algorithms used for classification. In addition,
our proposed approach improves accuracy, precision, recall, and F1 score, achieving a high detection
accuracy rate of 99.50%.
ACEP Magazine edition 4th launched on 05.06.2024Rahul
This document provides information about the third edition of the magazine "Sthapatya" published by the Association of Civil Engineers (Practicing) Aurangabad. It includes messages from current and past presidents of ACEP, memories and photos from past ACEP events, information on life time achievement awards given by ACEP, and a technical article on concrete maintenance, repairs and strengthening. The document highlights activities of ACEP and provides a technical educational article for members.
CHINA’S GEO-ECONOMIC OUTREACH IN CENTRAL ASIAN COUNTRIES AND FUTURE PROSPECTjpsjournal1
The rivalry between prominent international actors for dominance over Central Asia's hydrocarbon
reserves and the ancient silk trade route, along with China's diplomatic endeavours in the area, has been
referred to as the "New Great Game." This research centres on the power struggle, considering
geopolitical, geostrategic, and geoeconomic variables. Topics including trade, political hegemony, oil
politics, and conventional and nontraditional security are all explored and explained by the researcher.
Using Mackinder's Heartland, Spykman Rimland, and Hegemonic Stability theories, examines China's role
in Central Asia. This study adheres to the empirical epistemological method and has taken care of
objectivity. This study analyze primary and secondary research documents critically to elaborate role of
china’s geo economic outreach in central Asian countries and its future prospect. China is thriving in trade,
pipeline politics, and winning states, according to this study, thanks to important instruments like the
Shanghai Cooperation Organisation and the Belt and Road Economic Initiative. According to this study,
China is seeing significant success in commerce, pipeline politics, and gaining influence on other
governments. This success may be attributed to the effective utilisation of key tools such as the Shanghai
Cooperation Organisation and the Belt and Road Economic Initiative.
PERFORMANCE EVALUATION OF TUMOR DETECTION TECHNIQUES
1. International Journal on Computational Sciences & Applications (IJCSA) Vol.5, No.3, June 2015
DOI:10.5121/ijcsa.2015.5303 25
PERFORMANCE EVALUATION OF TUMOR
DETECTION TECHNIQUES
A.Sinduja1
and Dr.A.Suruliandi2
1
Department of Computer Science and Engineering, Manonmaniam Sundaranar
University, Tirunelveli-627012, India
2
Manonmaniam Sundaranar University, Tirunelveli-627012, India
ABSTRACT
Automatic segmentation of tumor plays a vital role in diagnosis and surgical planning. This paper deals
with techniques which providing solution for detecting hepatic tumor in Computed Tomography (CT)
images. The main aim of this work is to analyze performance of tumor detection techniques like Knowledge
Based Constraints, Graph Cut Method and Gradient Vector Flow active contour. These three techniques
are computed using sensitivity, specificity and accuracy. From the evaluated result, knowledge based
constraints method is better than other graph cut method and gradient vector flow active contour.
KEYWORDS
Tumor detection, knowledge based constraints, graph cut method, gradient vector flow active contour,
hepatic tumor
1.INTRODUCTION
The liver is the largest gland in the body which plays a vital role in keeping us alive. Its function
include storing vitamins and nutrient, producing proteins used for blood clotting, and creating bile
used for digestion. Two types of cancer affect the liver, primary cancer and Metastasized cancer.
The American cancer society’s estimates for primary liver cancer and intra hepatic bile duct
cancer in the United States are about 33,190 new cases will be diagnosed and about 23,000
people will die of these cancers. The percentage of liver cancer has been rising slowly for several
decades. Liver cancers is seen more often in men than in women.
CT imaging remains the best modality for liver metastases. The imaging of hepatic metastases
answers several roles: the evaluation of suspected lesions, preoperative planning, and the
monitoring of treatment and post-treatment follow up. Computed Tomography (CT) provides
significantly better image quality than Ultra Sound (US) and is most widely used imaging
techniques. CT is currently the most common imaging modality for detecting and characterizing
liver lesions.
The segmentation of the liver tumors in the CT images is a challenging task that is even
complicated by the clinical prospect of this work. Indeed, the structures to segment vary widely
and often a striking resemblance exists between tumors and other tissues. Hence the images from
2. International Journal on Computational Sciences & Applications (IJCSA) Vol.5, No.3, June 2015
26
clinical routine should be handled by the tumor detection techniques. To detect the tumor in the
CT images three methods of tumor detection techniques are used they are Knowledge based
constraints, Graph cut method and Gradient Vector Flow active contour. In this study, datasets of
different patients were processed using the above automatic mentioned method. The results were
computed using sensitivity, specificity and accuracy by comparing it with respect to manual
segmentation carried out by an expert radiologist
1.1.MOTIVATION AND JUSTIFICATION
The detection of liver tumour is a challenging task and is achieved on CT images because this
modality is the most common for the diagnosis of hepatic tumour. Without the use of Computed
Tomography (CT) medical decision are rarely taken. The trained clinicians were manually done
the segmentation of tumour lesions from CT slice image to identify tumour .
S.-J. Park, K.-S. Seo, and J.-A. Park.[14], proposed a method that first apply intensity histogram
transformation for segmentation of the liver and a posteriori classification resulting in a binary
mask. After morphological processing is applied to the mask, the tumors are located within the
mask area. K.-S. Seo.[6], proposed a method it segments the liver first, then by using the optimal
threshold value hepatic tumor is segmented with minimum total probability error. This approach
produces diverse false positives, but promising results are obtained.
Ciecholewski et al [10],proposed a method which segmented the liver using the contour model,
and then enhance the image using histrogam transformation. Region growing algorithm using
intensity distribution is proposed by Zhao et al. [17]. In Arakeri et al.[1], seed points are selected
automatically using modified region growing algorithm and also it proposed a method to find the
threshold value using automatic region growing method that incorporates fuzzy c-means
clustering algorithm. Massoptier and Casciaro.[12], firstly, the liver is segmented by the approach
statistical method then the tumor is classified by analyzing the wavelet. Chen and Metaxas [3],
used Markov Random Field (MRF) estimation and also Deformable models to segment the
tumors.
Shang et al. [16] presented active contour model with an embedded classifier, based on a
Gaussian mixture model fitted to the intensity distribution of the medical image to segment liver,
vessels and lesions.Lu et al. [9] , also used the active contour for which initial contour is
specified manually to obtain the tumor boundary. Segmentation based on Watershed algorithm
[5] in this method edge detection, the watershed algorithm and region merging approaches are
used. In Knowledge-Based Constraints[13] Firstly, the gray level intensity contrast was
enhanced; secondly, the image was added to itself, then, to isolate tumor directly the contrast
between liver and tumour is large enough to use the threshold ; thirdly, the image is segmented by
threshold based method; Since it is sensitive to noise ,the morphological filter was employed as
the post-processing.
Marius George Linguraru.[11], Firstly, Statistic adaptive threshold initialization is done. From the
threshold image user needs to select seed point with the label the object and background. Then
the image is represented in an undirected weighted graph. Every node of image represents each
pixel. The segmentation is obtained in cut of the graph. Each region represents a sub graph.
Sergio Casciaro[15] proposed a method in which the tumor is separated first by thresholding
method. Gradient Vector Flow (GVF) is the external force field for the active contour.
3. International Journal on Computational Sciences & Applications (IJCSA) Vol.5, No.3, June 2015
27
The main problem in tumour detection techniques are computational time, false positive rate,
initial assumption. The whole segmentation process take not more than few minutes in order to
satisfy the speed standards. The different approaches for liver tumour detection have been
developed but the best method is not identified by any research groups. Motivated by this
concept, an attempt is taken to implement and evaluate the tumour detection techniques for CT
liver image. The detection of tumour using CT image give good results compared to MRI scan
images. Justified by this, it is essential to evaluate the performance of such already existing
techniques for the practical application.
1.2.OUTLINE OF THE PROPOSED WORK
Fig 1 Block Diagram for performance evaluation of tumour detection techniques
Fig.1. is the overall working model for tumour detection techniques based on Knowledge Based
Constraints, Graph Cut Method and Gradient Vector Flow active contour methods. First, the CT
Liver image is given as the input. This input undergoes contrast stretching, pixel addition,
Gaussian smoothing, thresholding and morphological operation. Finally, segmented tumour area
is obtained.
Original Liver CT
image
Contrast Stretching
Adding Pixels to Itself
Gaussian Smoothing
Iso-data Threshold
Morphological Filters
Gaussian Filter
Thresholding
Seed Points
Gradient Vector Flow
active contour
Graph cut method
Segmented Liver Tumor Segmented Liver Tumor
Performance Evaluation
Segmented Liver Tumor
Liver tissue are
Noise removed image
Binary image
Image after noise
Finding source and
terminal node
Binary image
Finding tumour region (black)
and background(white)
4. International Journal on Computational Sciences & Applications (IJCSA) Vol.5, No.3, June 2015
28
Next, inputting the liver CT image to graph cut and Gradient Vector Flow active contour method.
First, the pre-processing process like Gaussian filter and thresh holding is done. From this seed
points are detected which undergoes further segmentation to detect accurate tumor area with
graph cut and Gradient Vector Flow Active contour method. Finally, performance is evaluated for
these three techniques using sensitivity, specificity and accuracy.
1.3.ORGANISATION OF THE PAPER
Rest of the paper is organized as follows. In the Section II various tumor detection techniques are
discussed. And section III is the experimental result and performance analysis. Finally, section IV
is the conclusion and the future scope.
2.VARIOUS TUMOR DETECTION METHODS
2.1.KNOWLEDGE BASED CONSTRAINTS(KBC)
First, Segmenting the liver structure from the CT image, then the similar gray level liver
parenchyma and tumor tissues is obtained by enhancing the contrast of the segmented slice. This
can be done by selecting the range for stretching, so that over enhancement is avoided. Here,
linear contrast stretching is used to increase the difference between liver tissue and tumors.
Fig 2 Liver with tumor
Fig 3 Liver after Linear Contrast Stretching
There are several methods for contrast stretching, such as Selective histogram equalization, direct
stretching with the linear relationship, linear stretching according to the fitting curve and
nonlinear stretching with the logarithmic transformation, direct stretching with the linear
relationship shows good result[19], formula applied to this performance is number (1):
ܫᇱ
=
′ܫ௫ − ′ܫ
ܫ௫ − ܫ
ሺܫ − ܫሻ + ′ܫ (1)
where, I is the gray level before transformation and I′ is the gray levels after transformation,
5. International Journal on Computational Sciences & Applications (IJCSA) Vol.5, No.3, June 2015
29
respect. I′max is the highest gray level after transformation and I′min is the lowest gray levels
after transformation, Imax is the maximum gray level and Imin is the minimum gray level in the
liver region before the transformation. Finally, liver and tumor is differentiated by its gray level
using linear contrast stretching. Formula(2), to add the enhanced image to itself is.
R (i, j) = I’(i, j) + I′( i, j) (2)
where R(i,j) is the image after applying this it will reach 255 and will appear as white. We obtain
the pixels of tumor tissues dark. The result of addition is image background as well liver tissue
that appears as white background with some pepper noise, and tumors that appears as dark spots
with range of gray levels. In order to remove the noise and make the region of tumor more
homogeneous as in shown Fig. 3(a),Gaussian smoothing is used as in (3), where ݔ is the distance
from the origin in the horizontal axis, ݕ is the distance from the origin in the vertical axis, and ߪ
is the standard deviation of the Gaussian distribution.
ܩሺ,ݔ ݕሻ =
1
2ߨߪଶ
݁
ି
௫మା௬మ
ଶఙమ
(3)
Next step is that isodata thresholding operation is done to turn the image into binary where the
tumor region is appeared as black blob on white space, followed by morphological hole filling
operation like erosion and dilation as in Fig 4.
Fig 4 Finally segmented tumor with Knowledge Based constraints
2.2.GRAPH CUT METHOD(GCM)
The graph-cut approach is applied to the segmented liver structure to find the hepatic tumors.[2].
In the basic form, graph cuts suffer from the shrinking bias problem, particularly for segmenting
elongated and small structures, such as the blood vessels and certain types of tumors. The
segmentation of abdominal organs improved with the help of graph cut [8]. From the medical
knowledge that tumors are generally circle like mask. For image segmentation ,the input data is
given from which it will find a globally optimal cut through the graph. The initial Graph G = {V,
E} where V is the nodes and E is the edges. Here we consider the pixel of image as nodes and
there are two other additional nodes they are S(source) for object and T(terminal) for the
background .4-neighbourhood system is considered to convert Image in to weighted graph. For
each pair of nodes an edge is formed called n-link. The following formula(4) is to calculate
weight function that is assigned for each n-link,
ܤ, = ܿ݁ݔ ቀ−
ሺூିூሻమ
ଶఙమ ቁ1/݀݅ݐݏሺ, ݍሻ. (4)
6. International Journal on Computational Sciences & Applications (IJCSA) Vol.5, No.3, June 2015
30
where ܤ, is a measure of the similarity of image intensities at pixels p and q. The seed points
were generated automatically for the tumors by finding thresholds in the data. The initialization
is supplied as a set of pixels in foreground (object) and a set of pixel in background .The
initialized regions are used to define the histograms for the region based segmentation. Here
negative log likelihood’s are used to define the weights
Rp(“obj”)=-lnPr(Ip|O); (5)
Rp(“bkg”)=-lnPr(Ip|B) (6)
where Pr(Ip|O),Pr(Ip|B) are the intensity histograms of object and background respectively.
Implementation of this Graph cut segmentation uses the maxflow algorithm to segment the tumor
from liver.
Fig 5 Liver with tumour
Fig 6 Final segmented tumour with graph cut
2.3.GRADIENT VECTOR FLOW(GVF) ACTIVE CONTOUR
The Gradient Vector Flow(GVF) active contour approach is applied to the segmented liver
structure to find the hepatic tumors.Before the segmentation of liver tumour in CT is done, the
main step is to eliminate unwanted noise, so for that it is necessary to apply the Gaussian filter for
removing all unwanted noises. After that seed points were generated automatically for the tumors
by finding thresholds in the data.
Active contour is an energy minimizing spline, its energy depends on its shape an location within
the image. An energy function E(c) can be defined on the contour as:
Eሺcሻ = ܧ௧ + ܧ௫௧ (7)
where, ܧ௧ and ܧ௧ denote the internal and external energies respectively. The internal
energy function determines the regularity, i.e. smooth shape, of the contour. A common choice
for the internal energy is a quadratic functional given by
7. International Journal on Computational Sciences & Applications (IJCSA) Vol.5, No.3, June 2015
31
ܧ௧ = ߙ|ܿሺݏሻ|ଶ
+ ߚ|ܿሺݏሻ|ଶ
݀ݏ
ଵ
(8)
Here,ߙ controls the tension of the contour, andߚ controls the rigidity of the countour. The
external energy term that determines the criteria of contour evaluation depending on the image
I(x,y),and can be defined as
ܧ௫௧ = ܧ൫ܿሺݏሻ൯݀ݏ
ଵ
(8)
ܧሺ,ݔ ݕሻ,denotes a scalar function defined on the image plane, so that local minimum of ܧ
attracts the snakes to edges. Solving the problem of snakes is to ind the contour that minimizes
the total energy term E using greedy algorithm with the given set of weights
ߙ ܽ݊݀ ߚ.Initialization of object boundary is the limitation to use this model for segmentation,
which can be overcome by other models. GVF snake [4] has been defined as an external forces to
push the snake into object concavity.
It is a 2D vector field V(s)=[u(s),v(s)], which minimizes the following energy functional
E=∬ ߟ൫ݑ௫
ଶ
+ݒ௫
ଶ
+ ݑ௬
ଶ
+ ݒ௬
ଶ
൯ + |∇݂|ଶ|v∇݂|ଶ
݀ݕ݀ݔ (9)
where,ݑ௫, ݒ௫, ݑ௬, ܽ݊݀ ݑ௬ are the spatial derivative of the field , ߟ is the regularization parameter,
which should be set according to the amount of noise of the image and f is the gradient of the
edge map which is defined as the negative external force i.e. F=- ܧ௫௧.The behavior of thee using
Gradient Vector Flow (GVF) Snake Algorithm approach that is able to converge to boundary
concavity can be explained from the Euler equation used to find the GVF field. These Euler
equations are.
ߟ∇ଶ
ݑ − ሺݑ − ݂௫ሻ൫݂௫
ଶ
+ ݂௬
ଶ
൯ = 0 (10)
ߟ∇ଶ
ݒ − ሺݒ − ݂௫ሻ൫݂௫
ଶ
+ ݂௬
ଶ
൯ = 0 (11)
where, ∇ଶ
is the Laplacian operator . Compared to the bollon force, the using GVF Snake
Algorithm approach is proven to converge relatively faster. This is caused by the external force
employed by the using GVF Snake Algorithm that make the capture range of the active contour
bigger. Since the GVF uses the classical formulation , its basic principle is to diffuse the edge
information from the object boundary to the rest of the image . The generation of using GVF
Snake algorithm is iterative and computationally intensive.
Fig 7 Liver with tumour
8. International Journal on Computational Sciences & Applications (IJCSA) Vol.5, No.3, June 2015
32
Fig 8 Final Segmented tumour with GVF Active contour
3.EXPERIMENTAL RESULTS AND PERFORMANCE ANALYSIS
In this, the tumour is segmented using KBC, GCM and GVF active contour. The result is the
segmented tumor in CT image, but the detected result may vary according to the algorithm’s
sensitivity, specificity and accuracy. If the specificity is high then, the tumour detection algorithm
gives the result correctly. While checking the percentage of accuracy, if the percentage is high it
will give the exact detection of liver tumour. Since this whole paper concentrated only on Liver
CT so best algorithm can be chosen according to how much tumour is detected correctly covered
by the algorithm
3.1.EXPERIMENTAL RESULT
This experiments calculate the performance of the tumour detection techniques for CT image.
The segmented tumour from the liver structure by three techniques is displayed in fig 9.
Test Images
Tumour Detection Techniques
KBC GCM GVF
9. International Journal on Computational Sciences & Applications (IJCSA) Vol.5, No.3, June 2015
33
Fig 9 Segmented Liver from CT image of different patients
3.2.PERFORMANCE EVALUATION
For performance measure the Sensitivity, Specificity and accuracy are computed as defined
below:
Sensitivity =
୬୳୫ୠୣ୰ ୭ ୲୰୳ୣ ୮୭ୱ୧୲୧୴ୣୱ
୬୳୫ୠୣ୰ ୭ ୲୰୳ୣ ୮୭ୱ୧୲୧୴ୣୱା୬୳୫ୠୣ୰ ୭ୟ୪ୱୣ ୬ୣୟ୲୧୴ୣୱ
(12)
Speciϐicity =
୬୳୫ୠୣ୰ ୭ ୲୰୳ୣ ୬ୣୟ୲୧୴ୣୱ
୬୳୫ୠୣ୰ ୭ ୲୰୳ୣ ୬ୣୟ୲୧୴ୣୱା୬୳୫ୠୣ୰ ୭ୟ୪ୱୣ ୮୭ୱ୧୲୧୴ୣୱ
(13)
Accuracy=
்௨ ௦௧௩ ା்௨ ே௧௩
୰୳ୣ ୭ୱ୧୲୧୴ୣା ୟ୪ୱୣ ୭ୱ୧୲୧୴ୣ ା ୰୳ୣ ୣୟ୲୧୴ୣ ା ୟ୪ୱୣ ୣୟ୲୧୴ୣ
(14)
Table 1 present the sensitivity, specificity and accuracy for segmented liver tumour image with
manually segmented liver tumour images.
Table 1 Sensitivity, Specificity and accuracy of various tumour detection techniques in percentage(%)
Image
KBC (%) GCM(%) GVF(%)
Sens Spec Acc Sens Spec Acc Sens Spec Acc
1 94 85 98 96 50 73 96 60 81
2 98 84 98 87 84 96 99 50 66
3 87 84 96 82 84 94 99 60 99
4 88 85 96 88 78 90 98 86 98
5 85 86 93 99 54 77 99 83 99
6 93 85 98 95 84 98 99 50 66
7 100 86 100 99 28 97 99 50 66
8 100 86 100 99 40 70 99 79 99
9 100 86 100 98 33 97 99 42 99
10 99 81 98 93 84 97 99 79 99
10. International Journal on Computational Sciences & Applications (IJCSA) Vol.5, No.3, June 2015
34
Here 10 test images are considered and sensitivity, specificity and accuracy of each image is
identified for above procedures and the average value is also calculated. It is observed that the
knowledge based constraints is providing best result of more than 80%.But the result for other
two methods provides lesser percentage for some images. So we go for calculating average
percentage value.
All three techniques are evaluated with sensitivity with the help of graph.
Fig 10 Performance evaluation based on Sensitivity
From fig 10, KBC and GCM performs better than GVF active contour.Now, Specificity is
calculated for all three methods.
Fig 11. Performance evaluation based on Specificity
From fig 11, KBC method performs better than other two methods. Finally, accuracy is calculated
for detected tumor by three techniques
Fig 12 Performance evaluation based on Specificity
From this evaluation KBC method performs better than the GCM and GVF active contour.
85%
90%
95%
KBC GVF GCM
SENSITIVITY
SENSITIVITY
0%
50%
100%
KBC GVF GCM
SPECIFICITY
SPECIFICITY
80%
90%
100%
KBC GVF GCM
ACCURACY
ACCURACY
11. International Journal on Computational Sciences & Applications (IJCSA) Vol.5, No.3, June 2015
35
4.CONCLUSION AND FUTURE SCOPE
In this paper the analysis was done on CT liver image for detecting the tumour using KBC, GVF
and GCM. Comparative study is done on these three techniques based on the sensitivity,
specificity and accuracy. From the experimental results it is noted that GVF and KBC method is
effective because it is having good accuracy, specificity and sensitivity.
Further studies are guaranteed to correct the small areas that were misclassified. These
corrections could be required in certain surgical planning application. This paper helps to obtain
almost the tumour detected area correctly. Further researches also strive towards improving
accuracy, precision, computational speed, automation and reduction of manual interaction.
REFERENCES
[1] Arakeri.M.P and Lakshmana. (2010) "Automatic segmentation of liver tumor on computed
tomography images," in Proceedings of the International Conference and Workshop on Emerging
Trends in Technology. Mumbai, Maharashtra, India: ACM.
[2] Boykov. Y and Kolmogorov. V.(Sep.2004)“An experimental comparison of mincut/max- flow
algorithms for energy minimization in vision,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 26, no. 9,
pp. 1124–1137.
[3] Chen.L and Metaxas.D.(2005) "A hybrid framework for 3D medical image segmentation," Journal of
Medical Image Analysis, vol. 9, pp. 547-565.
[4] Habib Zaidi, Torsten Rues , Frederic Schoenahl, and Marie-Louise Montandon.(2006)’Comparative
assessment of statistical brain MR image segmentation algorithms and their impact on partial volume
correction in PET’,Elsevier,Neuroimage 32,pp.1591-1607.
[5] Jianhua.L, Zhongyi.W, and Rui.Z.(2009) "Liver Cancer CT Image Segmentation Methods Based on
Watershed Algorithm," in International Conference on Computational Intelligence and Software
Engineering (CiSE '09). Wuhan, China, pp. 1-4.
[6] Kyung-Sik Seo and Tae-Woong Chung. (2005). ‘Automatic Boundary Tumor Segmentation of a
Liver’, In ICCSA 2005, pages 836–842.
[7] Liang. Y,Yang. L, and Fan. H.(2009)"Image enhancement for liver CT images," in International
Conference on Optical Instruments and Technology: Optoelectronic Imaging and Process
Technology, vol.7513, T. Yoshizawa, P. Wei, and J. Zheng, Eds. Shanghai, China:SPIE, pp.
75130K-75130K-8.
[8] Linguraru M.G et al.(2010)“Multi-organ segmentation from multi-Phase abdominal CT via 4D
graphs using enhancement, shape and Location optimization,” in MICCAI 2010, New York, vol.
6363, LNCS,pp. 89–96.
[9] Lu.R, Marziliano.P, and Hua Thng.C. (2005) "Liver Tumor Volume Estimation By Semi-Automatic
Segmentation Method," presented at 27th Annual International Conference of the Engineering in
Medicine and Biology Society (EMBS '05), Shanghai, China.
[10] Marcin Ciecholewski and Marek R. Ogiela. (2007) ‘Automatic Segmentation of Single and Multiple
Neoplastic Hepatic Lesions in CT Images’, In IWINAC ’07: Proceedings of the 2nd international
work-conference on Nature Inspired Problem-Solving Methods in Knowledge Engineering, pages 63–
71.
[11] Marius George Linguraru*, William J. Richbourg, Jianfei Liu, Jeremy M. Watt, Vivek
Pamulapati,Shijun Wang, and Ronald M. Summers.(2012)’Tumor Burden Analysis on Computed
Tomography by Automated Liver and Tumor Segmentation’ in Medical Imaging, IEEE Transactions
on (Volume:31 , Issue: 10 ) Page(s):1965 – 1976.
[12] Massoptier.L and Casciaro.S. (2008) "A New Fully Automatic and Robust Algorithm for Fast
Segmentation of Liver Tissue and Tumors from CT Scans," Journal of European Radiology, vol. 18,
pp. 1658-65.
12. International Journal on Computational Sciences & Applications (IJCSA) Vol.5, No.3, June 2015
36
[13] Nader H. Abdel-massieh, Mohiy M. Hadhoud, and Khalid M. Amin. (2010)’Automatic Liver Tumor
Segmentation from CT Scans with Knowledge-based Constraints’, in Biomedical Engineering
Conference (CIBEC), 2010 5th Cairo International ,Page(s): 215 – 218.
[14] Seung-Jin Park, Kyung-Sik Seo and Jong-An Park. (2005) ‘Automatic Hepatic Tumor Segmentation
Using Statistical Optimal Threshold’ In ICCS, pages 934–940.
[15] Sergio Casciaro, Roberto Franchini, Laurent Massoptier, Ernesto Casciaro, Francesco
Conversano,Antonio Malvasi, and Aimè Lay-Ekuakille.(2012)’ Fully Automatic Segmentations of
Liver and Hepatic Tumors From 3-D Computed Tomography Abdominal Images: Comparative
Evaluation of Two Automatic Methods’,in Sensors Journal, IEEE (Volume:12 , Issue: 3 )
Page(s):464 – 473.
[16] Shang.Y, Markova.A, Deklerck.R, Nyssen.E, Yang.X, and de Mey.J.(2010) "Liver segmentation
by an active contour model with embedded Gaussian mixture model based classifiers," in Optics,
Photonics, and Digital Technologies for Multimedia Applications, vol. 7723, P. Schelkens, T.
Ebrahimi, G. Cristobal, F. Truchetet, and P. Saarikko, Eds. Brussels, Belgium: SPIE, pp. 772313-7.
[17] Zhao.B, Schwartz.L.H, Jiang.L, Colville.J, Moskowitz.C, Wang.L, Leftowitz.R, Liu.F, and
Kalaigian.J, (2006) "Shape-Constraint Region Growing for Delineation of Hepatic Metastases on
Contrast-Enhanced Computed Tomography Scans," Journal of Investigative Radiology, vol.41, pp.
753-762.