Identification of Focal Cortical Dysplasia (FCD) can be difficult due to the subtle MRI changes. Though sequences like FLAIR (fluid attenuated inversion recovery) can detect a large majority of these lesions, there are smaller lesions without signal changes that can easily go unnoticed by the naked eye. The aim of this study is to improve the visibility of Focal Cortical Dysplasia lesions in the T1 weighted brain MRI images. In the proposed method, we used a complex diffusion based approach for calculating the FCD affected areas.
Retinal Macular Edema Detection Using Optical Coherence Tomography ImagesIOSRJVSP
Macular Edema affects around 20 million people of the world each year. Optical Coherence Tomography (OCT), a non-invasive eye-imaging modality, is capable of detecting Macular Edema both in its early and advanced stages. In this paper, an algorithm which detects Macular Edema from OCT images has been presented. Initially the images are filtered to de-noise them. Then, the retinal layers - Inner Limiting Membrane (ILM) and Retinal Pigment Epithelium (RPE) are segmented using Graph Theory method. Region splitting is performed on the OCT scan and the thickness between the two layers in the different regions are determined. Area enclosed between the two layers is also estimated. Support Vector Machine, a binary classifier is used to draw a classification between normal and abnormal OCT scans. Region-wise thickness, a few Haralick’s features, area between ILM and RPE and a few wavelet features are used to train the classifier. The classifier yielded an accuracy of 95% and a sensitivity of 100%. Thus, this algorithm can be used by ophthalmologists in early detection of Macular Edema.
Segmentation and Classification of Brain MRI Images Using Improved Logismos-B...IJERA Editor
Automated reconstruction and diagnosis of brain MRI images is one of the most challenging problems in medical imaging. Accurate segmentation of MRI images is a key step in contouring during radiotherapy analysis. Computed tomography (CT) and Magnetic resonance (MR) imaging are the most widely used radiographic techniques in diagnosis and treatment planning. Segmentation techniques used for the brain Magnetic Resonance Imaging (MRI) is one of the methods used by the radiographer to detect any abnormality specifically in brain. The method also identifies important regions in brain such as white matter (WM), gray matter (GM) and cerebrospinal fluid spaces (CSF). These regions are significant for physician or radiographer to analyze and diagnose the disease. We propose a novel clustering algorithm, improved LOGISMOS-B to classify tissue regions based on probabilistic tissue classification, generalized gradient vector flows with cost and distance function. The LOGISMOS graph segmentation framework. Expand the framework to allow regionally-aware graph construction and segmentation.
Fuzzy Clustering Based Segmentation of Vertebrae in T1-Weighted Spinal MR Imagesijfls
Image segmentation in the medical domain is a challenging field owing to poor resolution and limited
contrast. The predominantly used conventional segmentation techniques and the thresholding methods
suffer from limitations because of heavy dependence on user interactions. Uncertainties prevalent in an
image cannot be captured by these techniques. The performance further deteriorates when the images are
corrupted by noise, outliers and other artifacts. The objective of this paper is to develop an effective robust
fuzzy C- means clustering for segmenting vertebral body from magnetic resonance image owing to its
unsupervised form of learning. The motivation for this work is detection of spine geometry and proper
localisation and labelling will enhance the diagnostic output of a physician. The method is compared with
Otsu thresholding and K-means clustering to illustrate the robustness.The reference standard for validation
was the annotated images from the radiologist, and the Dice coefficient and Hausdorff distance measures
were used to evaluate the segmentation
A Novel Approach for Diabetic Retinopthy ClassificationIJERA Editor
Sustainable Diabetic Mellitus may lead to several complications towards patients. One of the complications is
diabetic retinopathy. Diabetic retinopathy is the type of complication towards the retinal and interferes with
patient’s sight. Medical examination toward patients with diabetic retinopathy is observed directly through
retinal images using fundus camera. Diabetic retinopathy is classified into four classes based on severity, which
are: normal, non-proliferative diabetic retinopathy (NPDR), proliferative diabetic retinopathy (PDR), and
macular edema (ME). The aim of this research is to develop a method which can be used to classify the level of
severity of diabetic retinopathy based on patient’s retinal images. Seven texture features were extracted from
retinal images using gray level co-occurence matrix three dimensional method (3D-GLCM). These features are
maximum probability, correlation, contrast, energy, homogeneity, and entropy; subsequently trained using
Levenberg-Marquardt Backpropagation Neural Network (LMBP). This study used 600 data of patient’s retinal
images, consist of 450 data retinal images for training and 150 data retinal images for testing. Based on the result
of this test, the method can classify the severity of diabetic retinopathy with sensitivity of 97.37%, specificity of
75% and accuracy of 91.67%
International Journal of Computational Engineering Research(IJCER)ijceronline
International Journal of Computational Engineering Research(IJCER) is an intentional online Journal in English monthly publishing journal. This Journal publish original research work that contributes significantly to further the scientific knowledge in engineering and Technology.
Automated Diagnosis of Glaucoma using Haralick Texture FeaturesIOSR Journals
Abstract : Glaucoma is the second leading cause of blindness worldwide. It is a disease in which fluid
pressure in the eye increases continuously, damaging the optic nerve and causing vision loss. Computational
decision support systems for the early detection of glaucoma can help prevent this complication. The retinal
optic nerve fibre layer can be assessed using optical coherence tomography, scanning laser polarimetry, and
Heidelberg retina tomography scanning methods. In this paper, we present a novel method for glaucoma
detection using an Haralick Texture Features from digital fundus images. K Nearest Neighbors (KNN)
classifiers are used to perform supervised classification. Our results demonstrate that the Haralick Texture
Features has Database and classification parts, in Database the image has been loaded and Gray Level Cooccurrence
Matrix (GLCM) and thirteen haralick features are combined to extract the image features, performs
better than the other classifiers and correctly identifies the glaucoma images with an accuracy of more than
98%. The impact of training and testing is also studied to improve results. Our proposed novel features are
clinically significant and can be used to detect glaucoma accurately.
Keywords: Glaucoma, Haralick Texture features, KNN Classifiers, Feature Extraction
Retinal Macular Edema Detection Using Optical Coherence Tomography ImagesIOSRJVSP
Macular Edema affects around 20 million people of the world each year. Optical Coherence Tomography (OCT), a non-invasive eye-imaging modality, is capable of detecting Macular Edema both in its early and advanced stages. In this paper, an algorithm which detects Macular Edema from OCT images has been presented. Initially the images are filtered to de-noise them. Then, the retinal layers - Inner Limiting Membrane (ILM) and Retinal Pigment Epithelium (RPE) are segmented using Graph Theory method. Region splitting is performed on the OCT scan and the thickness between the two layers in the different regions are determined. Area enclosed between the two layers is also estimated. Support Vector Machine, a binary classifier is used to draw a classification between normal and abnormal OCT scans. Region-wise thickness, a few Haralick’s features, area between ILM and RPE and a few wavelet features are used to train the classifier. The classifier yielded an accuracy of 95% and a sensitivity of 100%. Thus, this algorithm can be used by ophthalmologists in early detection of Macular Edema.
Segmentation and Classification of Brain MRI Images Using Improved Logismos-B...IJERA Editor
Automated reconstruction and diagnosis of brain MRI images is one of the most challenging problems in medical imaging. Accurate segmentation of MRI images is a key step in contouring during radiotherapy analysis. Computed tomography (CT) and Magnetic resonance (MR) imaging are the most widely used radiographic techniques in diagnosis and treatment planning. Segmentation techniques used for the brain Magnetic Resonance Imaging (MRI) is one of the methods used by the radiographer to detect any abnormality specifically in brain. The method also identifies important regions in brain such as white matter (WM), gray matter (GM) and cerebrospinal fluid spaces (CSF). These regions are significant for physician or radiographer to analyze and diagnose the disease. We propose a novel clustering algorithm, improved LOGISMOS-B to classify tissue regions based on probabilistic tissue classification, generalized gradient vector flows with cost and distance function. The LOGISMOS graph segmentation framework. Expand the framework to allow regionally-aware graph construction and segmentation.
Fuzzy Clustering Based Segmentation of Vertebrae in T1-Weighted Spinal MR Imagesijfls
Image segmentation in the medical domain is a challenging field owing to poor resolution and limited
contrast. The predominantly used conventional segmentation techniques and the thresholding methods
suffer from limitations because of heavy dependence on user interactions. Uncertainties prevalent in an
image cannot be captured by these techniques. The performance further deteriorates when the images are
corrupted by noise, outliers and other artifacts. The objective of this paper is to develop an effective robust
fuzzy C- means clustering for segmenting vertebral body from magnetic resonance image owing to its
unsupervised form of learning. The motivation for this work is detection of spine geometry and proper
localisation and labelling will enhance the diagnostic output of a physician. The method is compared with
Otsu thresholding and K-means clustering to illustrate the robustness.The reference standard for validation
was the annotated images from the radiologist, and the Dice coefficient and Hausdorff distance measures
were used to evaluate the segmentation
A Novel Approach for Diabetic Retinopthy ClassificationIJERA Editor
Sustainable Diabetic Mellitus may lead to several complications towards patients. One of the complications is
diabetic retinopathy. Diabetic retinopathy is the type of complication towards the retinal and interferes with
patient’s sight. Medical examination toward patients with diabetic retinopathy is observed directly through
retinal images using fundus camera. Diabetic retinopathy is classified into four classes based on severity, which
are: normal, non-proliferative diabetic retinopathy (NPDR), proliferative diabetic retinopathy (PDR), and
macular edema (ME). The aim of this research is to develop a method which can be used to classify the level of
severity of diabetic retinopathy based on patient’s retinal images. Seven texture features were extracted from
retinal images using gray level co-occurence matrix three dimensional method (3D-GLCM). These features are
maximum probability, correlation, contrast, energy, homogeneity, and entropy; subsequently trained using
Levenberg-Marquardt Backpropagation Neural Network (LMBP). This study used 600 data of patient’s retinal
images, consist of 450 data retinal images for training and 150 data retinal images for testing. Based on the result
of this test, the method can classify the severity of diabetic retinopathy with sensitivity of 97.37%, specificity of
75% and accuracy of 91.67%
International Journal of Computational Engineering Research(IJCER)ijceronline
International Journal of Computational Engineering Research(IJCER) is an intentional online Journal in English monthly publishing journal. This Journal publish original research work that contributes significantly to further the scientific knowledge in engineering and Technology.
Automated Diagnosis of Glaucoma using Haralick Texture FeaturesIOSR Journals
Abstract : Glaucoma is the second leading cause of blindness worldwide. It is a disease in which fluid
pressure in the eye increases continuously, damaging the optic nerve and causing vision loss. Computational
decision support systems for the early detection of glaucoma can help prevent this complication. The retinal
optic nerve fibre layer can be assessed using optical coherence tomography, scanning laser polarimetry, and
Heidelberg retina tomography scanning methods. In this paper, we present a novel method for glaucoma
detection using an Haralick Texture Features from digital fundus images. K Nearest Neighbors (KNN)
classifiers are used to perform supervised classification. Our results demonstrate that the Haralick Texture
Features has Database and classification parts, in Database the image has been loaded and Gray Level Cooccurrence
Matrix (GLCM) and thirteen haralick features are combined to extract the image features, performs
better than the other classifiers and correctly identifies the glaucoma images with an accuracy of more than
98%. The impact of training and testing is also studied to improve results. Our proposed novel features are
clinically significant and can be used to detect glaucoma accurately.
Keywords: Glaucoma, Haralick Texture features, KNN Classifiers, Feature Extraction
SEGMENTATION OF MULTIPLE SCLEROSIS LESION IN BRAIN MR IMAGES USING FUZZY C-MEANSijaia
Magnetic resonance images (MRI) play an important role in supporting and substituting clinical
information in the diagnosis of multiple sclerosis (MS) disease by presenting lesion in brain MR images. In
this paper, an algorithm for MS lesion segmentation from Brain MR Images has been presented. We revisit
the modification of properties of fuzzy -c means algorithms and the canny edge detection. By changing and
reformed fuzzy c-means clustering algorithms, and applying canny contraction principle, a relationship
between MS lesions and edge detection is established. For the special case of FCM, we derive a sufficient
condition and clustering parameters, allowing identification of them as (local) minima of the objective
function.
Abstract:
A technique for exudate detectionin fundus image is been presented in this paper. Due to diabetic retinopathy an abnormality is caused known as exudates.The loss of vision can be prevented by detecting the exudates as early as possible. The work mainly aims at detecting exudates which is present in the green channel of the RGB image by applying few preprocessing steps, DWT and feature extraction. The extracted features are fed to 3 different classifiers such as KNN, SVM and NN. Based on the classifier result if an exudate is present the extraction of exudate ROI is done based on canny edge detection followed by morphological operations. The severity of the exudates is established on the area of the detected exudate.
Keywords:Exudates, Fundus image, Diabetic retinopathy, DWT, KNN, SVM, NN, Canny edge detection, Morphological operations.
MULTIPLE SCLEROSIS DIAGNOSIS WITH FUZZY C-MEANScscpconf
Magnetic resonance imaging (MRI) can support and substitute clinical information in the
diagnosis of multiple sclerosis (MS) by presenting lesion. In this paper, we present an algorithm
for MS lesion segmentation. We revisit the modification of properties of fuzzy c means
algorithms and the canny edge detection. Using reformulated fuzzy c means algorithms, apply
canny contraction principle, and establish a relationship between MS lesions and edge
detection. For the special case of FCM, we derive a sufficient condition for fixed lesions,
allowing identification of them as (local) minima of the objective function.
Segmentation and Labelling of Human Spine MR Images Using Fuzzy Clustering csandit
Computerized medical image segmentation is a challenging area because of poor resolution
and weak contrast. The predominantly used conventional clustering techniques and the
thresholding methods suffer from limitations owing to their heavy dependence on user
interactions. Uncertainties prevalent in an image cannot be captured by these techniques. The
performance further deteriorates when the images are corrupted by noise, outliers and other
artifacts. The objective of this paper is to develop an effective robust fuzzy C- means clustering
for segmenting vertebral body from magnetic resonance images. The motivation for this work is
that spine appearance, shape and geometry measurements are necessary for abnormality
detection and thus proper localisation and labelling will enhance the diagnostic output of a
physician. The method is compared with Otsu thresholding and K-means clustering to illustrate
the robustness. The reference standard for validation was the annotated images from the
radiologist, and the Dice coefficient and Hausdorff distance measures were used to evaluate the
segmentation.
There are three major complications of diabetes which lead to blindness. They are retinopathy, cataracts, and glaucoma among which diabetic retinopathy is considered as the most serious complication affecting the blood vessels in the retina. Diabetic retinopathy (DR) occurs when tiny vessels swell and leak fluid or abnormal new blood vessels grow hampering normal vision.
Diabetic retinopathy is a widespread problem of visual impairment. The abnormalities like microaneurysms, hemorrhages and exudates are the key symptoms which play an important role in diagnosis of diabetic retinopathy. Early detection of these abnormalities may prevent the blurred vision or vision loss due to diabetic retinopathy. Basically exudates are lipid lesions able to be seen in optical images. Exudates are categorized into hard exudates and soft exudates based on its appearance. Hard exudates come out as intense yellow regions and soft exudates have fuzzy manifestations. Automatic detection of exudates may aid ophthalmologists in diagnosis of diabetic retinopathy and its early treatment. Fig. 1 shows the key symptoms of diabetic retinopathy.
Brain Tumor Extraction from T1- Weighted MRI using Co-clustering and Level Se...CSCJournals
The aim of the paper is to propose effective technique for tumor extraction from T1-weighted magnetic resonance brain images with combination of co-clustering and level set methods. The co-clustering is the effective region based segmentation technique for the brain tumor extraction but have a drawback at the boundary of tumors. While, the level set without re-initialization which is good edge based segmentation technique but have some drawbacks in providing initial contour. Therefore, in this paper the region based co-clustering and edge-based level set method are combined through initially extracting tumor using co-clustering and then providing the initial contour to level set method, which help in cancelling the drawbacks of co-clustering and level set method. The data set of five patients, where one slice is selected from each data set is used to analyze the performance of the proposed method. The quality metrics analysis of the proposed method is proved much better as compared to level set without re-initialization method.
Glaucoma Disease Diagnosis Using Feed Forward Neural Network ijcisjournal
Glaucoma is an eye disease which damages the optic nerve and or loss of the field of vision which leads to
complete blindness caused by the pressure buildup by the fluid of the eye i.e. the intraocular pressure
(IOP). This optic disorder with a gradual loss of the field of vision leads to progressive and irreversible
blindness, so it should be diagnosed and treated properly at an early stage. In this paper,
thedaubechies(db3) or symlets (sym3)and reverse biorthogonal (rbio3.7) wavelet filters are employed for
obtaining average and energy texture feature which are used to classify glaucoma disease with high
accuracy. The Feed-Forward neural network classifies the glaucoma disease with an accuracy of 96.67%.
In this work, the computational complexity is minimized by reducing the number of filters while retaining
the same accuracy.
Efficient Brain Tumor Detection Using Wavelet TransformIJERA Editor
Brain tumor detection is a challenging task and its very important to analyze the structure of the tumor correctly so a automatic method is used now a days for the detection of the tumor. This method saves time as well as it reduces the error which occurs in the method of manual detection. In this paper the tumor is detected using wavelet transform. MRI is an important tool used in many fields of medicine and is capable of generating a detailed image of any part of the human body. The tumor is segmented from the MRI images, features are extracted and then the area of the tumor is determined. PNN can successfully handle the process of brain tumor classification
In this paper we present a recently developed tool named BrainAssist, which can be used for the study and analysis of brain abnormalities like Focal Cortical Dysplasia (FCD), Heterotopia and Multiple Sclerosis (MS). For the analysis of FCD and Heterotopia we used T1 weighted MR images and for the analysis of Multiple Sclerosis we used Proton Density (PD) images. 52 patients were studied. Out of 52 cases 36 were affected with FCDs, 6 with MS lesions and 10 normal cases. Preoperative MR images were acquired on a 1.5-T scanner (Siemens Medical Systems, Germany).
MELANOMA CELL DETECTION IN LYMPH NODES HISTOPATHOLOGICAL IMAGES USING DEEP LE...sipij
Histopathological images are widely used to diagnose diseases including skin cancer. As digital histopathological images are typically of very large size, in the order of several billion pixels, automated identification of all abnormal cell nuclei and their distribution within multiple tissue sections would assist
rapid comprehensive diagnostic assessment. In this paper, we propose a technique, using deep learning algorithms, to segment the cell nuclei in Hematoxylin and Eosin (H&E) stained images and detect the abnormal melanocytes within histopathological images. The Nuclear segmentation is done by using a Convolutional Neural Network (CNN) and hand-crafted features are extracted for each nucleus. The segmented nuclei are then classified into normal and abnormal nuclei using a Support Vector Machine classifier. Experimental results show that the CNN can segment the nuclei with more than 90% accuracy. The proposed technique has a low computational complexity.
Brain tumor classification using artificial neural network on mri imageseSAT Journals
Abstract
In this paper, an attempt has been made to summarize the multi-resolution transformation and the different classifiers useful to
analyze the brain tumor using MRI. X-ray, MRI, Ultrasound etc. are different techniques used to scan brain tumor images.
Radiologist prefers MRI to get detail information about tumor to help him diagnoses. In this paper we have used MRI of brain
tumor for analysis. We have used Digital image processing tool for detection of the tumor. The identification, detection and
classification of brain tumor have been done by extracting features from MRI with the help of wavelet transformation. The MRI of
brain tumor is classified into two categories normal and abnormal brain. In this work Digital image processing has been used as
a tool for getting clear and exact details about tumor in earlier stages. This helps the physicians and practitioners for diagnoses.
Key word – Brain tumor, Wavelet transform, segmentation.
A novel equalization scheme for the selective enhancement of optical disc and...TELKOMNIKA JOURNAL
The ratio of the diameters of Optic Cup (OC) and Optic Disc (OD), termed as ‘Cup to Disc Ratio’
(CDR), derived from the fundus imagery is a popular biomarker used for the diagnosis of glaucoma.
Demarcation of OC and OD either manually or through automated image processing algorithms is error
prone because of poor grey level contrast and their vague boundaries. A dedicated equalization which
simultaneously compresses the dynamic range of the background and stretches the range of ODis
proposed in this paper. Unlike the conventional GHE, in the proposed equalization, the original histogram
is inverted and weighted nonlinearly before computing the Cumulative Probability Density (CPD).
The equalization scheme is compared with Adaptive Histogram Equalization (AHE), Global Histogram
Equalization (GHE) and Contrast Limited Adaptive Histogram Equalization (CLAHE) in terms of the
difference between the mean grey levels of OD and the background, using a quantitative metric known as
Contrast Improvement Index (CII). The CII exhibited by CLAHE, GHE and the proposed scheme are
1.1977 ± 0.0326, 1.0862 ± 0.0304 and 1.3312 ± 0.0486, respectively.The proposed method is observed to
be superior to CLAHE, GHE and AHE and it can be employed in Computerized Clinical Decision Support
Systems (CCDSS) to improve the accuracy of localizing the OD and the computation of CDR.
AUTOMATED DETECTION OF HARD EXUDATES IN FUNDUS IMAGES USING IMPROVED OTSU THR...IJCSES Journal
One common cause of visual impairment among people of working age in the industrialized countries is
Diabetic Retinopathy (DR). Automatic recognition of hard exudates (EXs) which is one of DR lesions in
fundus images can contribute to the diagnosis and screening of DR.The aim of this paper was to
automatically detect those lesions from fundus images. At first,green channel of each original fundus image
was segmented by improved Otsu thresholding based on minimum inner-cluster variance, and candidate
regions of EXs were obtained. Then, we extracted features of candidate regions and selected a subset which
best discriminates EXs from the retinal background by means of logistic regression (LR). The selected
features were subsequently used as inputs to a SVM to get a final segmentation result of EXs in the image.
Our database was composed of 120 images with variable color, brightness, and quality. 70 of them were
used to train the SVM and the remaining 50 to assess the performance of the method. Using a lesion based
criterion, we achieved a mean sensitivity of 95.05% and a mean positive predictive value of 95.37%. With
an image-based criterion, our approach reached a 100% mean sensitivity, 90.9% mean specificity and
96.0% mean accuracy. Furthermore, the average time cost in processing an image is 8.31 seconds. These
results suggest that the proposed method could be a diagnostic aid for ophthalmologists in the screening
for DR.
A Wavelet Based Automatic Segmentation of Brain Tumor in CT Images Using Opti...CSCJournals
This paper presents an automated segmentation of brain tumors in computed tomography images (CT) using combination of Wavelet Statistical Texture features (WST) obtained from 2-level Discrete Wavelet Transformed (DWT) low and high frequency sub bands and Wavelet Co-occurrence Texture features (WCT) obtained from two level Discrete Wavelet Transformed (DWT) high frequency sub bands. In the proposed method, the wavelet based optimal texture features that distinguish between the brain tissue, benign tumor and malignant tumor tissue is found. Comparative studies of texture analysis is performed for the proposed combined wavelet based texture analysis method and Spatial Gray Level Dependence Method (SGLDM). Our proposed system consists of four phases i) Discrete Wavelet Decomposition (ii) Feature extraction (iii) Feature selection (iv) Classification and evaluation. The combined Wavelet Statistical Texture feature set (WST) and Wavelet Co-occurrence Texture feature (WCT) sets are derived from normal and tumor regions. Feature selection is performed by Genetic Algorithm (GA). These optimal features are used to segment the tumor. An Probabilistic Neural Network (PNN) classifier is employed to evaluate the performance of these features and by comparing the classification results of the PNN classifier with the Feed Forward Neural Network classifier(FFNN).The results of the Probabilistic Neural Network, FFNN classifiers for the texture analysis methods are evaluated using Receiver Operating Characteristic (ROC) analysis. The performance of the algorithm is evaluated on a series of brain tumor images. The results illustrate that the proposed method outperforms the existing methods.
Hippocampus’s volume calculation on coronal slice’s for strengthening the dia...TELKOMNIKA JOURNAL
Alzheimer’s is one of the most common types of dementia in the world. Although not a contagious disease, this disease has many impacts, especially in socio-economic life. In diagnosing Alzheimer’s and using interview techniques, physical examination methods are also used, namely using an magnetic resonance imaging (MRI) machine to get a clear image of the patient’s brain condition, with a focus on the hippocampus and ventricular area. In this paper, we discuss the calculation of the volume of the hippocampus, especially the coronal slice, to provide information to doctors in making decisions on diagnosing the severity of Alzheimer’s. Using the basis of volume calculations, we made a 3D visualization reconstruction of the coronal hippocampus slice area in order to make it easier for doctors to analyze the condition of the hippocampus area, which in the end will be used as a recommendation in the classification of the severity of Alzheimer’s. Our experimental results show, the lower the severity, the bigger the volume, the more slices, and the longer the counting time.
Classification and Segmentation of Glaucomatous Image Using Probabilistic Neu...ijsrd.com
The gradual visual field loss and there is a characteristic type of damage to the retinal nerve fiber layer associated with the progression of the disease glaucoma. Texture features within images are actively pursued for accurate and efficient glaucoma classification. Energy distribution over wavelet subband is applied to find these important texture features. In this paper, we investigate the discriminatory potential of wavelet features obtained from the Daubechies (db3), symlets (sym3), and biorthogonal (bio3.3, bio3.5, and bio3.7) wavelet filters. We propose a novel technique to extract energy signatures obtained using 2-D discrete wavelet transform, and subject these signatures to different feature ranking and feature selection strategies. Here my project aims at the use of Probabilistic Neural Network (PNN), Fuzzy C-means (FCM) and K-means helps for the detection of glaucoma disease. For this, fuzzy c-means clustering algorithm and k-means algorithm is used. Fuzzy c-means results faster and reliably good clustering when compare to k-means.
AN EFFICIENT WAVELET BASED FEATURE REDUCTION AND CLASSIFICATION TECHNIQUE FOR...ijcseit
This research paper proposes an improved feature reduction and classification technique to identify mild and severe dementia from brain MRI data. The manual interpretation of changes in brain volume based on visual examination by radiologist or a physician may lead to missing diagnosis when a large number of MRIs are analyzed. To avoid the human error, an automated intelligent classification system is proposed
which caters the need for classification of brain MRI after identifying abnormal MRI volume, for the diagnosis of dementia. In this research work, advanced classification techniques using Support Vector Machines based on Particle Swarm Optimisation and Genetic algorithm are compared. Feature reduction
by wavelets and PCA are analysed. From this analysis, it is observed that the proposed classification of SVM based PSO is found to be efficient than SVM trained with GA and wavelet based feature reduction technique yields better results than PCA.
SEGMENTATION OF MULTIPLE SCLEROSIS LESION IN BRAIN MR IMAGES USING FUZZY C-MEANSijaia
Magnetic resonance images (MRI) play an important role in supporting and substituting clinical
information in the diagnosis of multiple sclerosis (MS) disease by presenting lesion in brain MR images. In
this paper, an algorithm for MS lesion segmentation from Brain MR Images has been presented. We revisit
the modification of properties of fuzzy -c means algorithms and the canny edge detection. By changing and
reformed fuzzy c-means clustering algorithms, and applying canny contraction principle, a relationship
between MS lesions and edge detection is established. For the special case of FCM, we derive a sufficient
condition and clustering parameters, allowing identification of them as (local) minima of the objective
function.
Abstract:
A technique for exudate detectionin fundus image is been presented in this paper. Due to diabetic retinopathy an abnormality is caused known as exudates.The loss of vision can be prevented by detecting the exudates as early as possible. The work mainly aims at detecting exudates which is present in the green channel of the RGB image by applying few preprocessing steps, DWT and feature extraction. The extracted features are fed to 3 different classifiers such as KNN, SVM and NN. Based on the classifier result if an exudate is present the extraction of exudate ROI is done based on canny edge detection followed by morphological operations. The severity of the exudates is established on the area of the detected exudate.
Keywords:Exudates, Fundus image, Diabetic retinopathy, DWT, KNN, SVM, NN, Canny edge detection, Morphological operations.
MULTIPLE SCLEROSIS DIAGNOSIS WITH FUZZY C-MEANScscpconf
Magnetic resonance imaging (MRI) can support and substitute clinical information in the
diagnosis of multiple sclerosis (MS) by presenting lesion. In this paper, we present an algorithm
for MS lesion segmentation. We revisit the modification of properties of fuzzy c means
algorithms and the canny edge detection. Using reformulated fuzzy c means algorithms, apply
canny contraction principle, and establish a relationship between MS lesions and edge
detection. For the special case of FCM, we derive a sufficient condition for fixed lesions,
allowing identification of them as (local) minima of the objective function.
Segmentation and Labelling of Human Spine MR Images Using Fuzzy Clustering csandit
Computerized medical image segmentation is a challenging area because of poor resolution
and weak contrast. The predominantly used conventional clustering techniques and the
thresholding methods suffer from limitations owing to their heavy dependence on user
interactions. Uncertainties prevalent in an image cannot be captured by these techniques. The
performance further deteriorates when the images are corrupted by noise, outliers and other
artifacts. The objective of this paper is to develop an effective robust fuzzy C- means clustering
for segmenting vertebral body from magnetic resonance images. The motivation for this work is
that spine appearance, shape and geometry measurements are necessary for abnormality
detection and thus proper localisation and labelling will enhance the diagnostic output of a
physician. The method is compared with Otsu thresholding and K-means clustering to illustrate
the robustness. The reference standard for validation was the annotated images from the
radiologist, and the Dice coefficient and Hausdorff distance measures were used to evaluate the
segmentation.
There are three major complications of diabetes which lead to blindness. They are retinopathy, cataracts, and glaucoma among which diabetic retinopathy is considered as the most serious complication affecting the blood vessels in the retina. Diabetic retinopathy (DR) occurs when tiny vessels swell and leak fluid or abnormal new blood vessels grow hampering normal vision.
Diabetic retinopathy is a widespread problem of visual impairment. The abnormalities like microaneurysms, hemorrhages and exudates are the key symptoms which play an important role in diagnosis of diabetic retinopathy. Early detection of these abnormalities may prevent the blurred vision or vision loss due to diabetic retinopathy. Basically exudates are lipid lesions able to be seen in optical images. Exudates are categorized into hard exudates and soft exudates based on its appearance. Hard exudates come out as intense yellow regions and soft exudates have fuzzy manifestations. Automatic detection of exudates may aid ophthalmologists in diagnosis of diabetic retinopathy and its early treatment. Fig. 1 shows the key symptoms of diabetic retinopathy.
Brain Tumor Extraction from T1- Weighted MRI using Co-clustering and Level Se...CSCJournals
The aim of the paper is to propose effective technique for tumor extraction from T1-weighted magnetic resonance brain images with combination of co-clustering and level set methods. The co-clustering is the effective region based segmentation technique for the brain tumor extraction but have a drawback at the boundary of tumors. While, the level set without re-initialization which is good edge based segmentation technique but have some drawbacks in providing initial contour. Therefore, in this paper the region based co-clustering and edge-based level set method are combined through initially extracting tumor using co-clustering and then providing the initial contour to level set method, which help in cancelling the drawbacks of co-clustering and level set method. The data set of five patients, where one slice is selected from each data set is used to analyze the performance of the proposed method. The quality metrics analysis of the proposed method is proved much better as compared to level set without re-initialization method.
Glaucoma Disease Diagnosis Using Feed Forward Neural Network ijcisjournal
Glaucoma is an eye disease which damages the optic nerve and or loss of the field of vision which leads to
complete blindness caused by the pressure buildup by the fluid of the eye i.e. the intraocular pressure
(IOP). This optic disorder with a gradual loss of the field of vision leads to progressive and irreversible
blindness, so it should be diagnosed and treated properly at an early stage. In this paper,
thedaubechies(db3) or symlets (sym3)and reverse biorthogonal (rbio3.7) wavelet filters are employed for
obtaining average and energy texture feature which are used to classify glaucoma disease with high
accuracy. The Feed-Forward neural network classifies the glaucoma disease with an accuracy of 96.67%.
In this work, the computational complexity is minimized by reducing the number of filters while retaining
the same accuracy.
Efficient Brain Tumor Detection Using Wavelet TransformIJERA Editor
Brain tumor detection is a challenging task and its very important to analyze the structure of the tumor correctly so a automatic method is used now a days for the detection of the tumor. This method saves time as well as it reduces the error which occurs in the method of manual detection. In this paper the tumor is detected using wavelet transform. MRI is an important tool used in many fields of medicine and is capable of generating a detailed image of any part of the human body. The tumor is segmented from the MRI images, features are extracted and then the area of the tumor is determined. PNN can successfully handle the process of brain tumor classification
In this paper we present a recently developed tool named BrainAssist, which can be used for the study and analysis of brain abnormalities like Focal Cortical Dysplasia (FCD), Heterotopia and Multiple Sclerosis (MS). For the analysis of FCD and Heterotopia we used T1 weighted MR images and for the analysis of Multiple Sclerosis we used Proton Density (PD) images. 52 patients were studied. Out of 52 cases 36 were affected with FCDs, 6 with MS lesions and 10 normal cases. Preoperative MR images were acquired on a 1.5-T scanner (Siemens Medical Systems, Germany).
MELANOMA CELL DETECTION IN LYMPH NODES HISTOPATHOLOGICAL IMAGES USING DEEP LE...sipij
Histopathological images are widely used to diagnose diseases including skin cancer. As digital histopathological images are typically of very large size, in the order of several billion pixels, automated identification of all abnormal cell nuclei and their distribution within multiple tissue sections would assist
rapid comprehensive diagnostic assessment. In this paper, we propose a technique, using deep learning algorithms, to segment the cell nuclei in Hematoxylin and Eosin (H&E) stained images and detect the abnormal melanocytes within histopathological images. The Nuclear segmentation is done by using a Convolutional Neural Network (CNN) and hand-crafted features are extracted for each nucleus. The segmented nuclei are then classified into normal and abnormal nuclei using a Support Vector Machine classifier. Experimental results show that the CNN can segment the nuclei with more than 90% accuracy. The proposed technique has a low computational complexity.
Brain tumor classification using artificial neural network on mri imageseSAT Journals
Abstract
In this paper, an attempt has been made to summarize the multi-resolution transformation and the different classifiers useful to
analyze the brain tumor using MRI. X-ray, MRI, Ultrasound etc. are different techniques used to scan brain tumor images.
Radiologist prefers MRI to get detail information about tumor to help him diagnoses. In this paper we have used MRI of brain
tumor for analysis. We have used Digital image processing tool for detection of the tumor. The identification, detection and
classification of brain tumor have been done by extracting features from MRI with the help of wavelet transformation. The MRI of
brain tumor is classified into two categories normal and abnormal brain. In this work Digital image processing has been used as
a tool for getting clear and exact details about tumor in earlier stages. This helps the physicians and practitioners for diagnoses.
Key word – Brain tumor, Wavelet transform, segmentation.
A novel equalization scheme for the selective enhancement of optical disc and...TELKOMNIKA JOURNAL
The ratio of the diameters of Optic Cup (OC) and Optic Disc (OD), termed as ‘Cup to Disc Ratio’
(CDR), derived from the fundus imagery is a popular biomarker used for the diagnosis of glaucoma.
Demarcation of OC and OD either manually or through automated image processing algorithms is error
prone because of poor grey level contrast and their vague boundaries. A dedicated equalization which
simultaneously compresses the dynamic range of the background and stretches the range of ODis
proposed in this paper. Unlike the conventional GHE, in the proposed equalization, the original histogram
is inverted and weighted nonlinearly before computing the Cumulative Probability Density (CPD).
The equalization scheme is compared with Adaptive Histogram Equalization (AHE), Global Histogram
Equalization (GHE) and Contrast Limited Adaptive Histogram Equalization (CLAHE) in terms of the
difference between the mean grey levels of OD and the background, using a quantitative metric known as
Contrast Improvement Index (CII). The CII exhibited by CLAHE, GHE and the proposed scheme are
1.1977 ± 0.0326, 1.0862 ± 0.0304 and 1.3312 ± 0.0486, respectively.The proposed method is observed to
be superior to CLAHE, GHE and AHE and it can be employed in Computerized Clinical Decision Support
Systems (CCDSS) to improve the accuracy of localizing the OD and the computation of CDR.
AUTOMATED DETECTION OF HARD EXUDATES IN FUNDUS IMAGES USING IMPROVED OTSU THR...IJCSES Journal
One common cause of visual impairment among people of working age in the industrialized countries is
Diabetic Retinopathy (DR). Automatic recognition of hard exudates (EXs) which is one of DR lesions in
fundus images can contribute to the diagnosis and screening of DR.The aim of this paper was to
automatically detect those lesions from fundus images. At first,green channel of each original fundus image
was segmented by improved Otsu thresholding based on minimum inner-cluster variance, and candidate
regions of EXs were obtained. Then, we extracted features of candidate regions and selected a subset which
best discriminates EXs from the retinal background by means of logistic regression (LR). The selected
features were subsequently used as inputs to a SVM to get a final segmentation result of EXs in the image.
Our database was composed of 120 images with variable color, brightness, and quality. 70 of them were
used to train the SVM and the remaining 50 to assess the performance of the method. Using a lesion based
criterion, we achieved a mean sensitivity of 95.05% and a mean positive predictive value of 95.37%. With
an image-based criterion, our approach reached a 100% mean sensitivity, 90.9% mean specificity and
96.0% mean accuracy. Furthermore, the average time cost in processing an image is 8.31 seconds. These
results suggest that the proposed method could be a diagnostic aid for ophthalmologists in the screening
for DR.
A Wavelet Based Automatic Segmentation of Brain Tumor in CT Images Using Opti...CSCJournals
This paper presents an automated segmentation of brain tumors in computed tomography images (CT) using combination of Wavelet Statistical Texture features (WST) obtained from 2-level Discrete Wavelet Transformed (DWT) low and high frequency sub bands and Wavelet Co-occurrence Texture features (WCT) obtained from two level Discrete Wavelet Transformed (DWT) high frequency sub bands. In the proposed method, the wavelet based optimal texture features that distinguish between the brain tissue, benign tumor and malignant tumor tissue is found. Comparative studies of texture analysis is performed for the proposed combined wavelet based texture analysis method and Spatial Gray Level Dependence Method (SGLDM). Our proposed system consists of four phases i) Discrete Wavelet Decomposition (ii) Feature extraction (iii) Feature selection (iv) Classification and evaluation. The combined Wavelet Statistical Texture feature set (WST) and Wavelet Co-occurrence Texture feature (WCT) sets are derived from normal and tumor regions. Feature selection is performed by Genetic Algorithm (GA). These optimal features are used to segment the tumor. An Probabilistic Neural Network (PNN) classifier is employed to evaluate the performance of these features and by comparing the classification results of the PNN classifier with the Feed Forward Neural Network classifier(FFNN).The results of the Probabilistic Neural Network, FFNN classifiers for the texture analysis methods are evaluated using Receiver Operating Characteristic (ROC) analysis. The performance of the algorithm is evaluated on a series of brain tumor images. The results illustrate that the proposed method outperforms the existing methods.
Hippocampus’s volume calculation on coronal slice’s for strengthening the dia...TELKOMNIKA JOURNAL
Alzheimer’s is one of the most common types of dementia in the world. Although not a contagious disease, this disease has many impacts, especially in socio-economic life. In diagnosing Alzheimer’s and using interview techniques, physical examination methods are also used, namely using an magnetic resonance imaging (MRI) machine to get a clear image of the patient’s brain condition, with a focus on the hippocampus and ventricular area. In this paper, we discuss the calculation of the volume of the hippocampus, especially the coronal slice, to provide information to doctors in making decisions on diagnosing the severity of Alzheimer’s. Using the basis of volume calculations, we made a 3D visualization reconstruction of the coronal hippocampus slice area in order to make it easier for doctors to analyze the condition of the hippocampus area, which in the end will be used as a recommendation in the classification of the severity of Alzheimer’s. Our experimental results show, the lower the severity, the bigger the volume, the more slices, and the longer the counting time.
Classification and Segmentation of Glaucomatous Image Using Probabilistic Neu...ijsrd.com
The gradual visual field loss and there is a characteristic type of damage to the retinal nerve fiber layer associated with the progression of the disease glaucoma. Texture features within images are actively pursued for accurate and efficient glaucoma classification. Energy distribution over wavelet subband is applied to find these important texture features. In this paper, we investigate the discriminatory potential of wavelet features obtained from the Daubechies (db3), symlets (sym3), and biorthogonal (bio3.3, bio3.5, and bio3.7) wavelet filters. We propose a novel technique to extract energy signatures obtained using 2-D discrete wavelet transform, and subject these signatures to different feature ranking and feature selection strategies. Here my project aims at the use of Probabilistic Neural Network (PNN), Fuzzy C-means (FCM) and K-means helps for the detection of glaucoma disease. For this, fuzzy c-means clustering algorithm and k-means algorithm is used. Fuzzy c-means results faster and reliably good clustering when compare to k-means.
AN EFFICIENT WAVELET BASED FEATURE REDUCTION AND CLASSIFICATION TECHNIQUE FOR...ijcseit
This research paper proposes an improved feature reduction and classification technique to identify mild and severe dementia from brain MRI data. The manual interpretation of changes in brain volume based on visual examination by radiologist or a physician may lead to missing diagnosis when a large number of MRIs are analyzed. To avoid the human error, an automated intelligent classification system is proposed
which caters the need for classification of brain MRI after identifying abnormal MRI volume, for the diagnosis of dementia. In this research work, advanced classification techniques using Support Vector Machines based on Particle Swarm Optimisation and Genetic algorithm are compared. Feature reduction
by wavelets and PCA are analysed. From this analysis, it is observed that the proposed classification of SVM based PSO is found to be efficient than SVM trained with GA and wavelet based feature reduction technique yields better results than PCA.
AN EFFICIENT WAVELET BASED FEATURE REDUCTION AND CLASSIFICATION TECHNIQUE FOR...ijcseit
This research paper proposes an improved feature reduction and classification technique to identify mild
and severe dementia from brain MRI data. The manual interpretation of changes in brain volume based on
visual examination by radiologist or a physician may lead to missing diagnosis when a large number of
MRIs are analyzed. To avoid the human error, an automated intelligent classification system is proposed
which caters the need for classification of brain MRI after identifying abnormal MRI volume, for the
diagnosis of dementia. In this research work, advanced classification techniques using Support Vector
Machines based on Particle Swarm Optimisation and Genetic algorithm are compared. Feature reduction
by wavelets and PCA are analysed. From this analysis, it is observed that the proposed classification of
SVM based PSO is found to be efficient than SVM trained with GA and wavelet based feature reduction
technique yields better results than PCA.
MR Image Segmentation of Patients’ Brain Using Disease Specific a priori Know...CSCJournals
Segmentation of high quality brain MR images using a priori knowledge about brain structures enables a more accurate and comprehensive interpretation. Benefits of applying a priori knowledge about the brain structures may also be employed for image segmentation of specific brain and neural patients. Such procedure may be performed to determine the disease stage or monitor its gradual progression over time. However segmenting brain images of patients using general a priori knowledge which corresponds to healthy subjects would result in inaccurate and unreliable interpretation in the regions which are affected by the disease. In this paper, a technique is proposed for extracting a priori knowledge about structural distribution of different brain tissues affected by a specific disease to be applied for accurate segmentation of the patients’ brain images. For this purpose, extracted a priori knowledge is gradually represented as disease specific probability maps throughout an iterative process, and then is utilized in a statistical approach for segmentation of new patients’ images. Experiments conducted on a large set of images acquired from patients with a similar neurodegenerative disease implied success of the proposed technique for representing meaningful a priori knowledge as disease specific probability maps. Promising results obtained also indicated an accurate segmentation of brain MR images of the new patients using the represented a priori knowledge, into three tissue classes of gray matter, white matter, and cerebrospinal fluid. This enables an accurate estimation of tissues’ thickness and volumes and can be counted as a substantial forward step for more reliable monitoring and interpretation of progression in specific brain and neural diseases.
Improving radiologists’ and orthopedists’ QoE in diagnosing lumbar disk herni...IJECEIAES
This article studies and analyzes the use of 3D models, built from magnetic reso- nance imaging (MRI) axial scans of the lumbar intervertebral disk, that are needed for the diagnosis of disk herniation. We study the possibility of assisting radiologists and orthopedists and increasing their quality of experience (QoE) during the diagnosis process. The main aim is to build a 3D model for the desired area of interest and ask the specialists to consider the 3D models in the diagnosis process instead of considering multiple axial MRI scans. We further propose an automated framework to diagnose the lumber disk herniation using the constructed 3D models. We evaluate the effectiveness of increasing the specialists QoE by conducting a questionnaire on 14 specialists with different experiences ranging from residents to consultants. We then evaluate the effectiveness of the automated diagnosis framework by training it with a set of 83 cases and then testing it on an unseen test set. The results show that the the use of 3D models increases doctors QoE and the automated framework gets 90% of diagnosis accuracy.
SEGMENTATION AND LABELLING OF HUMAN SPINE MR IMAGES USING FUZZY CLUSTERINGcscpconf
Computerized medical image segmentation is a challenging area because of poor resolution and weak contrast. The predominantly used conventional clustering techniques and the thresholding methods suffer from limitations owing to their heavy dependence on user interactions. Uncertainties prevalent in an image cannot be captured by these techniques. The performance further deteriorates when the images are corrupted by noise, outliers and other artifacts. The objective of this paper is to develop an effective robust fuzzy C- means clustering for segmenting vertebral body from magnetic resonance images. The motivation for this work is that spine appearance, shape and geometry measurements are necessary for abnormality detection and thus proper localisation and labelling will enhance the diagnostic output of a physician. The method is compared with Otsu thresholding and K-means clustering to illustrate the robustness. The reference standard for validation was the annotated images from the radiologist, and the Dice coefficient and Hausdorff distance measures were used to evaluate the segmentation.
Early detection of glaucoma through retinal nerve fiber layer analysis using ...eSAT Journals
Abstract The retinal nerve fiber layer (RNFL) is a vital part of human visual system, which can be directly observed by the fundus camera. This paper describes a method for glaucomatous retina detection based on Texture and Fractal description, followed by classification using support vector machine classifier. The color fundus images are used, in which the region of retinal nerve fibers are analyzed. It is shown that Texture & Fractal dimensions are correlated and linear correlation coefficient values are estimated at 0.35, 0.57, and 0.87 for healthy RNFL, medium loss and severe loss of RNFL respectively. The features are measured at 303 RNFL regions retinal positions in the peri-papillary area from 50 non-glaucomatous and 24 glaucomatous retinal fundus images. The presented method can also be used for glaucoma detection. Keywords- Retinal nerve Fiber layer, Glaucoma, Fractal dimension, texture feature, Box counting method
Brain Image Fusion using DWT and Laplacian Pyramid Approach and Tumor Detecti...INFOGAIN PUBLICATION
Image fusion is the process of combining important information from two or more images into a single image. The resulting image will be more enhanced than any of the input pictures. The idea of combining multiple image modalities to furnish a single, more enhanced image is well established, special fusion methods have been proposed in literature. This paper is based on image fusion using laplacian pyramid and Discreet Wavelet Transform (DWT) methods. This system uses an easy and effective algorithm for multi-focus image fusion which uses fusion rules to create fused image. Subsequently, the fused image is obtained by applying inverse discreet wavelet transform. After fused image is obtained, watershed segmentation algorithm is applied to detect the tumor part in fused image.
Visible watermarking within the region of non interest of medical images base...csandit
Transfer of medical information amongst various hospitals and diagnostic centers for mutual
availability of diagnostic and therapeutic case studies is a very common process. Watermarking
is adding “ownership” information in multimedia contents to verify signal integrity, prove
authenticity and achieve control over the copy process. Distortion in Region of Interest (ROI) of
a bio-medical image caused by watermarking may lead to wrong diagnosis and treatment.
Therefore, proper selection of Region of Non-Interest (RONI) in a medical image is very crucial
for adding watermark. First part of the present work proposes proper selection of Region of
Non-Interest based on Fuzzy C-Means segmentation and Harris corner detection, to improve
retention of diagnostic value lost in embedding ownership information. The second part of the
work presents watermark embedding in the selected area of RONI based on alpha blending
technique. In this approach, the generated watermarked image having an acceptable level of
imperceptibility and distortion is compared to the original image. The Peak Signal to Noise
Ratio (PSNR) of the original image vs. watermarked image is calculated to prove the efficacy of
the proposed method.
Similar to Focal Cortical Dysplasia Lesion Analysis with Complex Diffusion Approach (20)
The complexity of Medical image reconstruction requires tens to hundreds of billions of computations per second. Until few years ago, special purpose processors designed especially for such applications were used. Such processors require significant design effort and are thus difficult to change as new algorithms in reconstructions evolve and have limited parallelism. Hence the demand for flexibility in medical applications motivated the use of stream processors with massively parallel architecture. Stream processing architectures offers data parallel kind of parallelism.
As data processing requirements increased with new applications, new processing technologies like Stream computing and parallel execution came into being. This write‐up briefly compares two competing performance architectures for data parallelism – Cell Broadband Engine (Cell BE in short) and the GPU (Graphics Processing Unit). The Cell BE Processor architecture was developed in collaboration between IBM, Sony and Toshiba. Development started in 2001 and first set of products based on this architecture started appearing in 2005.
A Set-top-Box (STB) is a very common name heard in the consumer electronics market. It is a device that is attached to a Television for enhancing its functions or the quality of its functions. On the other side, the STB is connected to an external source of signal, like satellite, cable, terrestrial or internet. The STB processes the signal it receives, turns it into content, which is then displayed on the television screen or other display device. There are different types of STBs based on what kind of signals it can receive and what kind of processing it can do. The most widely used STBs are DVB STBs, which receive DVB (Digital Video Broadcast) transmission.
Fast and robust tracking of multiple faces is receiving increased attention from computer vision researchers as it finds potential applications in many fields like video surveillance and computer mediated video conferencing. Real-time tracking of multiple faces in high resolution videos involve three basic tasks namely initialization, tracking and display. Among these, tracking is quite compute intensive as it involves particle filtering that won’t yield a real time performance if we use a conventional CPU based system alone.
This paper presents a study of the efficiency and performance speedup achieved by applying Graphics Processing Units for Face Recognition Solutions. We explore one of the possibilities of parallelizing and optimizing a well-known Face Recognition algorithm, Principal Component Analysis (PCA) with Eigenfaces. In recent years, the Graphics Processing Units (GPU) has been the subject of extensive research and the computation speed of GPUs has been rapidly increasing.
In today’s competitive software development scenario, the customer demands a testing coverage which not only ensures the stated requirements but also the implied ones. This situation calls for an exhaustive testing which may not be always possible due to various reasons. Testing, due to its last position in SDLC, often gets crunched due to the cumulative schedule slippages. Hence Tester is faced with a challenge to make testing as efficient as possible within a short time span due to cost constraints. With selective testing an only option, test leads usually go for the age-old approach of Random Testing. Random testing does not ensure coverage in a scientific manner.
SOM (Self-Organizing Map) is one of the most popular artificial neural network algorithms in the unsupervised learning category. For efficient construction of large maps searching the best-matching unit is usually the computationally heaviest operation in the SOM. The parallel nature of the algorithm and the huge computations involved makes it a good target for GPU based parallel implementation. This paper presents an overall idea of the optimization strategies used for the parallel implementation of Basic-SOM on GPU using CUDA programming paradigm.
This paper provides an overview of Universal Plug and Play (UPnP) and how it works to build a digital home network. UPnP network technology allows personal computer and consumer electronics devices to advertise and offer their services to network clients. UPnP can be viewed as the technological foundation of the digital home, enabling innovative usage models, higher levels of automation, and easier integration of devices from different manufacturers. UPnP technology is all about making home networking simple and affordable for users.
In this paper a PDE based hybrid method for image denoising is introduced. The method is a bi-stage filter with anisotropic diffusion followed by wavelet based bayesian shrinkage. Here efficient denoising is achieved by reducing the convergence time of anisotropic diffusion.
In this paper we discuss the speckle reduction in images with the recently proposed Wavelet Embedded Anisotropic Diffusion (WEAD) and Wavelet Embedded Complex Diffusion (WECD). Both these methods are improvements over anisotropic and complex diffusion by adding wavelet based bayes shrink in its second stage. Both WEAD and WECD produce excellent results when compared with the existing speckle reduction filters.
Software Defined Networking (SDN) is an emerging trend in the networking and communication industry and promises to deliver enormous benefits, from reduced costs to more efficient network operations. It is a new approach that gives network operators and owners more control of the infrastructure, allowing optimization, customization and virtualization that enable the creation of new types of network services. This is done by decoupling the management and control planes that make decisions about where traffic is sent from (the control plane) the underlying hardware that forwards data traffic to the selected destination (the data plane).
Software Testing is the last phase in software development lifecycle which has high impact on the quality of the final product delivered to the customer. Even after being a critical phase, it was not given the importance as it actually deserves. The schedule constraints and slippage carry forwarded from the previous phase also make the testing phase more torrent. History reveals that the situation has changed with time, wherein testing is now visualized as one of the most critical, phase of software development. This makes software testing a discipline which demands for continuous and systematic growth. Software testing is a trade-off between Cost, Time and Quality.
Test Automation is an accepted technique which is adapted by the industry for increasing the effectiveness of the testing phase. The recurring tasks are being automated by the tools thus simplifying the human efforts and results in increased quality of product under test. A study of test automation programmes in the industry reveals the fact that a good percentage of them fail to find the intended results.
Complex digital and analog circuits and multiple clock signals used for design and development of modern systems usually make the job of engineers and designers a tedious one. While working with complex circuits and signals, a designer might encounter problems with circuit validation due to long simulation time. These complexities adversely affect the development time and hence increase time to market incurring higher production costs. By applying a new methodology in their Digital Phase-Locked Loop (Digital PLL) design, the engineers at QuEST reduced the simulation effort to one-by-third.
In software industry, test automation is a key solution for achieving volume verification and validation with optimal costs. Picking up the right automation tool and underlying scripting language has always been a challenge, balancing between cost factors and team’s expertise levels in various tools and scripting languages. A real solution would be one that allows full flexibility for team on these two core concern areas – test automation tool and scripting language. Flexi any Script any Tool (FaSaT) is a test automation framework which provides interoperability among multiple test automation tools and multiple scripting languages.
The improved hybrid model for molecular image denoising, proposed by NeST Software, can give a better SNR Molecular Image output. Read more on the proposed hybrid model.
Analog-to-Digital Converter (ADC) is an integral part of high-speed signal processing applications. This paper discusses about 10-bit SAR based ADC that enables very low power consumption and sampling rate as high as 165 MSPS.
Ground breaking innovations like Advanced Driver Assistance System (ADAS) makes driving easier and safer on congested roads. The whitepaper details how FPGA technology emerges as a complete solution for ADAS.
Reusable Video IP Cores give software engineering service providers flexibility and less time to market while catering to the ever increasing demands of customers. Read on to know more about the Reusable IP Cores developed by NeST Software.
More from QuEST Global (erstwhile NeST Software) (19)
One of the most developed cities of India, the city of Chennai is the capital of Tamilnadu and many people from different parts of India come here to earn their bread and butter. Being a metropolitan, the city is filled with towering building and beaches but the sad part as with almost every Indian city
Empowering ACOs: Leveraging Quality Management Tools for MIPS and BeyondHealth Catalyst
Join us as we delve into the crucial realm of quality reporting for MSSP (Medicare Shared Savings Program) Accountable Care Organizations (ACOs).
In this session, we will explore how a robust quality management solution can empower your organization to meet regulatory requirements and improve processes for MIPS reporting and internal quality programs. Learn how our MeasureAble application enables compliance and fosters continuous improvement.
Medical Technology Tackles New Health Care Demand - Research Report - March 2...pchutichetpong
M Capital Group (“MCG”) predicts that with, against, despite, and even without the global pandemic, the medical technology (MedTech) industry shows signs of continuous healthy growth, driven by smaller, faster, and cheaper devices, growing demand for home-based applications, technological innovation, strategic acquisitions, investments, and SPAC listings. MCG predicts that this should reflects itself in annual growth of over 6%, well beyond 2028.
According to Chris Mouchabhani, Managing Partner at M Capital Group, “Despite all economic scenarios that one may consider, beyond overall economic shocks, medical technology should remain one of the most promising and robust sectors over the short to medium term and well beyond 2028.”
There is a movement towards home-based care for the elderly, next generation scanning and MRI devices, wearable technology, artificial intelligence incorporation, and online connectivity. Experts also see a focus on predictive, preventive, personalized, participatory, and precision medicine, with rising levels of integration of home care and technological innovation.
The average cost of treatment has been rising across the board, creating additional financial burdens to governments, healthcare providers and insurance companies. According to MCG, cost-per-inpatient-stay in the United States alone rose on average annually by over 13% between 2014 to 2021, leading MedTech to focus research efforts on optimized medical equipment at lower price points, whilst emphasizing portability and ease of use. Namely, 46% of the 1,008 medical technology companies in the 2021 MedTech Innovator (“MTI”) database are focusing on prevention, wellness, detection, or diagnosis, signaling a clear push for preventive care to also tackle costs.
In addition, there has also been a lasting impact on consumer and medical demand for home care, supported by the pandemic. Lockdowns, closure of care facilities, and healthcare systems subjected to capacity pressure, accelerated demand away from traditional inpatient care. Now, outpatient care solutions are driving industry production, with nearly 70% of recent diagnostics start-up companies producing products in areas such as ambulatory clinics, at-home care, and self-administered diagnostics.
CHAPTER 1 SEMESTER V PREVENTIVE-PEDIATRICS.pdfSachin Sharma
This content provides an overview of preventive pediatrics. It defines preventive pediatrics as preventing disease and promoting children's physical, mental, and social well-being to achieve positive health. It discusses antenatal, postnatal, and social preventive pediatrics. It also covers various child health programs like immunization, breastfeeding, ICDS, and the roles of organizations like WHO, UNICEF, and nurses in preventive pediatrics.
ICH Guidelines for Pharmacovigilance.pdfNEHA GUPTA
The "ICH Guidelines for Pharmacovigilance" PDF provides a comprehensive overview of the International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH) guidelines related to pharmacovigilance. These guidelines aim to ensure that drugs are safe and effective for patients by monitoring and assessing adverse effects, ensuring proper reporting systems, and improving risk management practices. The document is essential for professionals in the pharmaceutical industry, regulatory authorities, and healthcare providers, offering detailed procedures and standards for pharmacovigilance activities to enhance drug safety and protect public health.
India Clinical Trials Market: Industry Size and Growth Trends [2030] Analyzed...Kumar Satyam
According to TechSci Research report, "India Clinical Trials Market- By Region, Competition, Forecast & Opportunities, 2030F," the India Clinical Trials Market was valued at USD 2.05 billion in 2024 and is projected to grow at a compound annual growth rate (CAGR) of 8.64% through 2030. The market is driven by a variety of factors, making India an attractive destination for pharmaceutical companies and researchers. India's vast and diverse patient population, cost-effective operational environment, and a large pool of skilled medical professionals contribute significantly to the market's growth. Additionally, increasing government support in streamlining regulations and the growing prevalence of lifestyle diseases further propel the clinical trials market.
Growing Prevalence of Lifestyle Diseases
The rising incidence of lifestyle diseases such as diabetes, cardiovascular diseases, and cancer is a major trend driving the clinical trials market in India. These conditions necessitate the development and testing of new treatment methods, creating a robust demand for clinical trials. The increasing burden of these diseases highlights the need for innovative therapies and underscores the importance of India as a key player in global clinical research.
The Impact of Meeting: How It Can Change Your Life
Focal Cortical Dysplasia Lesion Analysis with Complex Diffusion Approach
1. Focal Cortical Dysplasia (FCD) Lesion Analysis with Complex
Diffusion Approach
Jeny Rajan1
, K.Kannan1
, C. Kesavadas2
, Bejoy Thomas2
, A.K. Gupta2
1
Medical Imaging Research Group, NeST,
Technopark, Trivandrum, India
2
Dept. of Imaging Sciences & Interventional Radiology,
SCTIMST, Trivandrum, India
Corresponding author’s address:
Jeny Rajan
Medical Imaging Research Group
NeST (p) Ltd.
A-3, Periyar
Technopark Campus
Trivandrum – 695 581
INDIA
Email : jenyrajan@gmail.com
Ph.no.: +91 471 4060824
Fax: +91 471 2700442
* Manuscript
2. Abstract:
Identification of Focal Cortical Dysplasia (FCD) can be difficult due to the subtle MRI
changes. Though sequences like FLAIR (fluid attenuated inversion recovery) can detect a
large majority of these lesions, there are smaller lesions without signal changes that can
easily go unnoticed by the naked eye. The aim of this study is to improve the visibility of
Focal Cortical Dysplasia lesions in the T1 weighted brain MRI images. In the proposed
method, we used a complex diffusion based approach for calculating the FCD affected
areas. Based on the diffused image and thickness map, a complex map is created. From
this complex map; FCD areas can be easily identified. MRI brains of forty eight subjects
selected by neuroradiologists were given to computer scientists who developed the
complex map for identifying the cortical dysplasia. The scientists were blinded to the
MRI interpretation result of the neuroradiologist. The FCD could be identified in all the
patients in whom surgery was done, however three patients had false positive lesions.
More lesions were identified in patients in whom surgery was not performed and lesions
were seen in few of the controls. These were considered as false positive. This computer
aided detection technique using complex diffusion approach can help detect focal cortical
dysplasia in patients with epilepsy.
Keywords : Complex Diffusion, Cortical Thickening, Epilepsy, Focal Cortical
Dysplasia, MRI, Thickness Map
3. Introduction
Focal Cortical Dysplasia (FCD) is a neuronal migration disorder and is a major cause of
medically refractory epilepsy. The most carefully collected data from international
surveys indicate that about 1 adult in 200 suffers from recurrent epilepsy. Around 30% of
brain epilepsy is due to FCD [1]. The histologic features of FCD range from mild
disruption of the cortical organization to more severe forms with marked cortical
dyslamination, voluminous balloon cells littered throughout the cortex and astrocytosis
[2],[3]. During the last 35 years, developments in imaging, electroencephalography, and
electrocorticography have allowed more patients with medically refractory epilepsy to
undergo resective surgery [4].
Magnetic Resonance Imaging (MRI) plays a pivotal role in the presurgical evaluation
of patients with intractable epilepsy. Although MRI has allowed the recognition of FCD
in an increased number of patients, standard radiological evaluation fails to identify
lesions in a large number of cases due to their subtlety and the complexity of the cortex
convolution. Even with high resolution MRI, the lesional boundaries are often difficult to
delineate by neuroimaging or on the basis of the macroscopic appearance of the cortex
during the surgical procedure. On T1-weighted MRI sequence, FCD is usually
characterized by variable degrees of cortical thickening and reduced demarcation of the
gray-white matter junction [5]. FLAIR sequence shows hyperintensity of gray and
subcortical white matter. Recently many computational models and image processing
techniques have been developed to improve the lesion detection [5] - [8] [26]. In all these
models the above said FCD characteristics are used to identify the lesions. In the
4. proposed method we used a combination of complex diffusion [9] and cortical thickness
map for identifying FCD lesions.
Complex diffusion is a comparatively new Partial Differential Equation (PDE) based
method and can be applied for processing images. This is a generalization of diffusion
and free Schrodinger equations [9]. Analysis of linear complex diffusion shows that
generalized diffusion has properties of both forward and inverse diffusion. When
complex diffusion is applied to images, we will get details in real and imaginary planes.
In real plane we will get smoothed areas of the image and in imaginary plane, the edge
components of the image. When non linear complex diffusion is applied to images, intra
region smoothing will occur before inter region smoothing. This property of non linear
diffusion will help to detect the blurring effect in Gray-White matter junction. The
contrast difference between lesion tissues and non-lesional tissues can be increased by
applying complex diffusion.
II. Materials and Methods
The 3D T1 weighted sequence images of forty-eight subjects were selected by the
neuroradiologist. These included images of patients with intractable epilepsy due to
cortical dysplasia and of subjects with normal brain MRI. There were 27 males and 21
females of age group 0.3 years to 55 years old. These images were given to the computer
scientists, who were blinded about the MRI findings. The diagnosis of FCD was based on
the MRI imaging criteria of thickening of cortex, blurring of gray-white matter junction
and/or cortical and subcortical white matter hyperintensity. The MRI diagnosis of FCD
was made by neuroradiologists specializing in epilepsy imaging. The diagnosis was based
on an epilepsy MRI protocol consisting of axial and high resolution coronal T2 fast spin
5. echo, axial and coronal FLAIR, 3D T1 weighted spoiled gradient, diffusion weighted and
susceptibility weighted imaging. Out of 48 subjects, 35 had FCD and the rest had a
normal MRI. The subjects with the normal MRI were selected by the neuroradiologists as
the controls for the study. This age matched control subjects had undergone the same
MRI imaging protocol. The computer scientists were informed about the inclusion of
normal controls in the study but were not informed about the identity of imaging data of
the controls. Ten of the patients underwent surgery and the pathological diagnosis of
focal cortical dysplasia was made.
The MR images were acquired on a 1.5-T scanner (Siemens Medical Systems,
Germany). The acquired 3D MRI data set consisted of approximately 120 T1 weighted
coronal slices with TR/TE/FOV/ flip angle of 11s/4.94s/23cm/ 15°, matrix size 256 x
224, thickness 1.5 mm and pixel spacing 0.89 mm. Axial FLAIR images were acquired
with TR/TE/ FOV/ slice thickness/interslice thickness of 9000s/109s/23 cm/5mm/1.5mm
and matrix size of 256 x 224. In the proposed method we used a combination of complex
diffusion and thickness map to identify lesions.
The following preprocessing steps are applied first.
1. The images were intensity normalized using a subject specific linear multiplier
based on the median voxel-wise intensity [10], [11] of the image to an average
control brain.
2. The scalp and lipid layers were removed from each image of the entire volume.
Images are converted into axial slices before stripping scalp and lipid layers.
Morphological operations such as dilation, erosion and connected component
analysis are used for stripping scalp from brain MR images [12] - [14].
6. 3. The possibility of FCD in cerebellum as a cause for intractable epilepsy is
comparatively negligible, so cerebellum is removed before processing to reduce
false positives.
4. The next step is the segmentation of brain MRI into Gray Matter (GM), White
Matter (WM) and CSF. For this a Gaussian curve was used to fit each of the gray
and white matter peaks in the histogram. The intensity threshold between gray
and white matter was then automatically determined by the intersection of the two
Gaussian curves, eliminating the reliance on the local minimum between the gray
and white matter peaks [5].
5. Usually FCD affects the gray matter area. So searching for FCD is restricted to
gray matter area alone. The white matter and CSF is removed from the segmented
image.
Once the preprocessing operations are completed, thickness map, complex diffused
image and complex map are calculated. These methods are explained below.
A. Thickness Map Calculation
Cortical thickening and blurring effect in gray-white matter interface are the findings
seen in FCD. To calculate the cortical thickness, many methods are proposed in the
literature [14] – [19]. In the proposed method we used the method suggested earlier [16]
in which cortical surface is considered as equipotential surface used in the mathematical
description of electrostatic fields and described by Laplace’s equation. This method
solves Laplace’s equation to construct trajectories passing through the cortical sheet
connecting one surface to the other. A particular advantage of this approach is that for
7. any path or trajectory there is mutual correspondence between the points on the two
surfaces regardless of the trajectory’s starting point [20].
The method works by considering the cortex of each hemisphere to be a volume
bounded by two surfaces, S and '
S . Laplace’s equation (shown in eqn. 1) is solved over
the area between S and '
S to calculate the scalar filed [16].
02
2
2
2
2
2
2
zyx
(1)
In their method Jones. et. al [16] defined a potential every where between two lines
S and '
S such that = 0 on S and = 10,000 on '
S .The resulting profile of is a
smooth transition from = 0 V on S to 10,000 V on '
S . The significant property of
Laplace’s equation is that nonintersecting intermediate lines, or isopotentials, with
constant values between 0 V and 10,000 V must exist between S and '
S .Once the
solution of is obtained, filed lines are computed and normalized to
E
E
N (2)
where E and N represents a unit vector field defined everywhere between S and
'
S which always points perpendicularly to the sublayer on which it sits [16]. Integrating
in the direction defined by N at any point in the volume from one boundary to the other,
provides the length of the trajectory and hence the cortical thickness.
8. B. Calculating Complex Diffused Image
For calculating complex diffused image, we used non-linear complex diffusion
proposed by Gilboa et. al [9]. Complex diffusion is a generalization of diffusion and free
Schrodinger equations.
The nonlinear complex diffusion is of the form
)))(Im(( IIcIt (3)
where
2
)Im(
1
)Im(
k
I
e
Ic
i
(4)
is the gradient and k is the threshold parameter. In the experiment we used k as 1.5.
When image is processed with complex diffusion, we will get low frequency
components (plateaus) of the image in real plane and high frequency components (edges)
in the imaginary plane. As iteration continues more high frequency components will
move to imaginary plane. The component in the real and imaginary is equivalent to that
of the image convolved with Gaussian and Laplacian of Gaussian (LOG).
In the proposed method we applied complex diffusion to the segmented gray matter
with an iteration step of 40. The reasons for selecting non-linear complex diffusion is that
intra region smoothing will occur before inter region smoothing. So FCD areas and non-
FCD areas in gray matter will diffuse separately. The advantage of using complex
diffusion is that we can take advantage of imaginary part also. Figure 1 shows the real
and imaginary parts of an FCD affected brain image.
9. C. Calculating Complex Map
In both thickness map and real part of the diffused image, the areas where thickness is
more is represented with high intensity and in imaginary part the reverse. So to get a
better view of thickness area, we derived the following equation for calculating complex
map
diffusioncomplexofpartimaginary
diffusioncomplexofpartrealthicknesscortical
mapComplex
(3)
When non-linear complex diffusion is applied to an image, the areas having similar
properties will be grouped first (intra region smoothing occur before inter region
smoothing, a property of non linear diffusion). The contrast between FCD areas and non-
FCD areas will increase in the real plane after complex diffusion. The imaginary part of
complex diffusion is almost equal to Laplacian of Gaussian (LOG), in which the borders
will be highlighted. When the real part of the complex diffusion is divided with
imaginary part, all the smooth areas in the gray matter will also get enhanced (the areas
other than edges), but when this result is weighted with cortical thickness, only the areas
affected with FCDs highlighted. This process can be clearly understood from Fig 2. It can
also be seen from the Fig 2 that there is a significant increase in the contrast difference
between lesion and non lesion areas in the gray matter.
III. Results
The computer scientists, who were blinded to the MRI findings, identified FCDs in
38 patients of the total 48 patients. This included the ten patients in whom surgery was
done and a pathological diagnosis was made. The histopathology was Taylor- balloon
10. cell dysplasia in the majority of patients. Table-1 illustrates the findings in these ten
patients. In each patient the neuroradiologists reported the site of the lesion in T1
weighted and FLAIR sequences. There were two patients (patient 3 & 6) in whom the
lesion was not visualised in the T1 weighted sequence, but was seen in the FLAIR
sequence. In both these patients the lesion was small and subtle in the FLAIR sequence.
The FCD area was highlighted after post processing in nine out of the ten surgically
proved cases in the same location where the lesion was detected by the neuroradiologist.
In three patients (patient 3, 6, 7) more areas were identified by the processing technique.
In patient 6, the area identified by the neuroradiologist was not highlighted by the
postprocessing technique. Instead few other areas were highlighted. Since there was no
clinical, neuropsychological or video-EEG correlations for the site of seizure onset and
the additional areas highlighted by proposed method in the patients, these areas were
considered false positives.
Among the rest of thirty eight MRI given to the computer scientists, the lesion was
identified in twenty nine and nine were reported as normal. The results are summarised in
the Table-2. The neuroradiologists had identified a cortical dysplasia in only twenty five
of these 38 MRI. In five of these twenty five patients the neuroradiologists had reported
the dysplasia based on a thickened cortex with poor grey – white matter distinction in the
T1 weighted image. The FLAIR did not reveal any intralesional or subcortical signal
changes. The lesion was detected in the same location by the neuroradiologist and the
computer scientists in all the twenty five cases detected by both.
More lesions were shown by the computer scientists in eight patients. This
included 4 patients in whom the neuroradiologist had reported a normal MRI and was
11. taken as the controls of the study. After the report from the scientists the MRI images
were reviewed by the neuroradiologists. But again these areas were not diagnosed as
FCD. Hence these areas were considered to be radiological false positive area. Since
these areas were considered as false positive based on the MRI findings and not on
pathology, it is likely, though remote, that some of these areas may be truly areas of
dysplasia. Except in one patient who was 3 months old, the image processing steps
could be done without difficulty. In this infant the white matter myelination was not
complete and hence there was difficulty in segmentation. The FCD was large in this
patient and the lesion identified by the neuroradiologist corresponded with the lesion
identified by the scientist.
IV. Discussion
While many techniques are being developed [5]-[8],[22][23] to enhance FCD
lesions from MR images, our paper demonstrates the integration of two Partial
Differential Equation (PDE) based methods (Thickness measurement of Jones et.al and
Complex diffusion) to extract FCD areas. In most of the previous methods [5],[6],[22]
thickness map along with gradient techniques are used to compute FCD areas. But by
using complex diffusion both blurring effect and gradient change can be computed in a
single step. Experimental results show that this hybrid mathematical model is a good
candidate for FCD segmentation (Fig 3). The proposed method produces a mean contrast
change of around 16 times (between FCD areas and nearby tissues) than that of original
image. The contrast change is computed by taking the mean intensity difference of the
FCD areas and nearby tissues (gray matter) before and after processing. The t test
12. analysis done on the results also shows that increase in contrast was significant (P=
0.007).
In the surgically proved cases the proposed technique was correct in
highlighting the site of lesion in ninety percent of patients. Most of the previously
reported techniques had a detection rate of more than seventy percent [21]-[23]. In the
patient in whom the lesion was not seen by proposed technique, there was no significant
cortical thickening in the T1 images. However the neuroradiologist could diagnose FCD
based on the subcortical FLAIR hyperintensity. This is an important fact because the
automated image processing techniques base their detection mainly on T1 cortical
thickening and poor grey- white matter distinction. At the same time small Taylor balloon
cell cortical dysplasia can present with only subtle cortical hyperintensity in FLAIR and
these techniques usually fail in these cases. Non balloon cell dysplasia which are more
difficult to diagnose radiologically because of the absence of signal changes are the
lesions in which the image processing techniques can help. We feel that the image
processing technique can also help in detecting heterotopic grey matter, another easily
missed radiological entity. Studies have to be done to prove this. Detection of cortical
dysplasia by this technique can sometimes be difficult in infants especially before the
myelination is complete in the T1 weighted images. The segmentation of the grey and
white matter, which is one of the steps of this image processing technique, can be
difficult as was noted in one of our patients. Though methods have been described for
segmentation in the developing brain [27], we have not used this technique in our patient.
Although potential false positives were significantly reduced in the complex
map they were not completely eliminated. In some patients additional one or two areas
13. were highlighted in the complex map. There is possibility that some false positives may
in fact be true lesions of FCD. A review of the routine images and the map by the
radiologist will increase the detectability of these lesions. Even though there is a
significant reduction in false positives, its presence is still a problem in developing fully
automated system for FCD detection. Removal of cerebellum and caudate nucleus in the
pre-processing stage can further increase the accuracy of results. Recently some methods
have been developed for removing caudate nucleus from MR images, but most of them
depends on some atlas [24],[25].
V. Conclusion
Image processing techniques using complex diffusion approach can help detect focal
cortical dysplasia. These lesions can be easily missed in the MRI studies done in patients
as a work up for epilepsy. This computer aided detection technique can be used to
identify the abnormal areas, following which the radiologist can survey all the MRI
sequences to diagnose the area which are visually truly positive for the cortical dysplasia
and which agrees with the neurophysiology and Positron Emission Tomography (PET).
This technique is especially useful in patients in whom the routine MRI appears normal
and adds to the diagnostic armamentarium of these patients. Larger studies using this
technique with validation of results are needed to understand the role of this technique in
guiding placement of depth electrodes and serving as a complimentary technique to PET,
neurophysiology and magneto-encephalography.
14. Table 1: Report in the surgically proved cases
Pt.
No: Sex Age
Neuro
radiology Report
Location
Scientist
Report Histopathology
T1 FLAIR *
Site
1. F 03Y + + A FCD- b c
2. M 08Y + − A FCD
3. F 18Y + + A & C FCD- b c
4. M 25Y + + A FCD- b c
5. M 19Y + + A FCD- b c
6. F 25Y − + B & C FCD- b c
7. F 28Y + + A & C FCD- b c
8. M 27Y + + A FCD- b c
9. M 06Y + + A FCD- b c
10. M 37Y + − A FCD
*
Site: A. corresponds to that identified by neuroradiologist.
B. does not correspond to that identified by neuroradiologist
C. more sites than that identified by neuroradiologist (which may be false
positive)
+ - seen
− - not seen
FCD - bc focal cortical dysplasia – balloon cell type
FCD Focal cortical dysplasia
Table 2: Summary of results of MRI whose surgery was not performed.
Number of patients: 38
Neuroradiology report
Number of patients in whom dysplasia was detected: 25
Dysplasia detection based on FLAIR image alone: 20
Dysplasia detection based on T1 weighted & FLAIR image: 25
Normal MRI: 13
Scientist Report
Number of patients in whom dysplasia was detected: 29
Normal MRI: 9
Site of lesion detection
Same as the site of lesion as detected by neuroradiologist : 25
Additional sites detected: 8
15. Figure Legends
Fig 1: (a) Scalp & Cerebellum removed MR Brain image, (b) & (c) are real and
imaginary part of (a) after complex diffusion (5 iterations), (d) Gray matter segmented
from (a), (e) & (f) are real and imaginary part of (d) after complex diffusion (15
iterations). The contrast difference between lesion and non-lesion areas can be easily
identified from (e).
Fig 2: The proposed approach for FCD detection
Fig 3: Figure shows the original and processed T1 MR Brain images. It can be seen that
it is easy to identify the FCD areas from the images after processing it with the proposed
approach.
16. Acknowledgements
The authors thank Dr.P.Sankara Sarma, Additional Professor of Biostatistics, SCTIMST,
Trivandrum for advising on the statistical calculation in the paper. The authors also thank
the Director of SCTIMST and President of NeST for permitting collaboration between
Department of Imaging Sciences and Interventional Radiology & Medical Imaging
Research Group, NeST, Technopark, Trivandrum.
References
[1] Rahman A, Choudhary MI, Thomsen WJ., Bioassay Techniques for Drug
Development, Informa Health Care 2001, 97:98.
[2] Cesaroni E, Zamponi N, Rychlicki F, Scerrati M, Trignani R, Scarpelli M, Polonara
G., Focal Cortical Dysplasia: Neurophysiological Features and Surgical Outcome in
Children. 14th Meeting of the World Society for Stereotactic and Functional
Neurosurgery, 2005, 14:17.
[3] Wabg VY, Chang EF, Barbaro NM., Focal Cortical Dysplasia: A Review of
Pathological Features, Genetics and Surgical Outcome, Neurosurg Focus, 2006,
1:7,20.
[4] Double KL, Halliday GM, Kril JJ, Harasty JA, Cullen K, Brooks WS, Creasey H,
Broe GA.,Topography of brain atrophy during normal aging and Alzheimer's
disease. Neurobiol Aging ,1996,513:521,16.
[5] Antel SB, Bernasconi A, Bernasconi N, Collins D L, Kearney RE, Shinghal R,
Arnold DL. (2002) Computational Models of MRI Characteristics of Focal Cortical
Dysplasia Improve Lesion Detection. Neuroimage 17:1755-1760.
[6] Kassubek J, Huppertz HJ, Spreer J Bonhage AS. (2002) Detection and Localization
of Focal Cortical Dysplasia by Voxel based 3D MRI Analysis. Epilepsia, 43:596-
602.
17. [7] O. Colliot, T. Mansi, P. Besson, N. Bernasconi, A. Bernasconi, Improved
Segmentation of Focal Cortical Dysplasia Lesions on MRI using Expansion
Towards Cortical Boundaries, 3rd IEEE International Symposium on Biomedical
Imaging vol.6 no.9, pp. 323 – 326, 2006.
[8] O. Colliot, T. Mansi, N. Bernasconi, V. Naessens, D. Klironomos, and A.
Bernasconi, Segmentation of Focal Cortical Dysplasia Lesions on MRI using Level
Set Evolution, NeuroImage, vol. 32, no.4, pp. 1621-1630, 2006.
[9] Gilboa G, Sochen N, Zeevi YY. (2004) Image Enhancement and Denoising by
Complex Diffusion Processes. IEEE Trans Pattern Anal Mach Intell 26(8):1020-
1036.
[10] Brinkmann BH, Manduca A, Robb RA. (1998) Optimized homomorphic unsharp
masking for MRgrayscale inhomogeneity correction IEEE Trans Med Imaging
17:161–171.
[11] Vokurka EA, Thacker NA, Jackson A. (1999) A fast model independent method for
automatic correction of intensity nonuniformity in MRI data. J Magn Reson
Imaging 10(4):550-562.
[12] Atkins MS, Mackiewich BT. (1998) Fully Automatic Segmentation of the Brain in
MRI. IEEE Trans Med Imaging 17(1):98-107.
[13] Segonne F, Dale AM, Busa E,Glessner M, Salat D, Hahn HK, Fischl B. (2004) A
hybrid approach to the skull stripping problem in MRI. Neuroimage 22:1060–1075.
[14] Dogdas B, Shattuck D, Leahy RM. (2005) Segmentation of Skull and Scalp in 3D
Human MRI Using Mathematical Morphology. Hum Brain Mapp 26(4):273-285.
[15] Fischl B, Dale AM.(2000) Measuring the thickness of the human cerebral cortex
from magnetic resonance images. Proc Nat Acad Sci USA 97(20):11050-11055.
[16] Jones SE, Buchbinder BR, Aharon I. (2000) Three-dimensional mapping of cortical
thickness using laplace’s equation. Hum Brain Mapp 11:12–32.
18. [17] Kabani N, Goualher GL, MacDonald D, Evans AC. (2001) Measurement of cortical
thickness using an automated 3-d algorithm:A validation study. Neuroimage
13:375–380.
[18] Sowell ER, Peterson BS, Thompson PM, Welcome SE, Henkenius AL, Toga AW.
(2003) Mapping cortical change across the human life span. Nat Neurosci 6(3):309-
315.
[19] Han X, Jovicich J, Salat D, van der Kouwe A, Quinn B, Czanner S, Busa E, Pacheco
J, Albert M, Killiany R, Maguire P, Rosas D, Makris N, Dale A, Dickerson B,
Fischl B. (2006) Reliability of MRI-derived measurements of human cerebral
cortical thickness: The effects of field strength, scanner upgrade and manufacturer.
Journal of Neuroimage 32:180-194.
[20] Chloe Hutton, Enrico De Vita, John Ashburner, Ralf Deichmann, Robert Turner,
Voxel based cortical thickness measurements in the MRI, NeuroImage 40: 1701-
1710, 2008.
[21] S Srivastava, F Maes, D Vandermeulen, W V Paesschen, P Dupont, P Suetens,
Feature based statistical analysis of structural MR data for automatic detection of
focal cortical dysplastic lesions, Nuuroimage 27: 253-266, 2005.
[22] S B Antel, D L Collins, N Bernasconi, F Andermann, Rajjan Shinghal, R E
Kearney, D L Arnold, A Bernasconi, Automated detection of focal cortical
dysplasia lesions using computational models of their characteristics and texture
analysis, Neuroimage 19 : 1748-1759, 2003.
[23] O Colliot, N Bernasconi, N Khalili, S B Antel, V Naessens, A Bernasconi,
Individual voxel based analysis of gray matter in focal cortical dysplasia,
Neuroimage 29: 162-171, 2006.
19. [24] K. Kartheeswari, K. Muneeswaran, J. Rajan, K. Kannan, Automatic segmentation of
caudate nucleus from MR images, National conference on networks, image and
security, 2008.
[25] Y. Xia, K. Bettinger, L. Shen, A L Reiss, Automatic segmentation of the caudate
nucleus from from human brain MR images, IEEE Trans. Med. Imaging, 26: 509-
517, 2007.
[26] J.M. Bruggeman, M. Wilke, S.S Som, A.M.E. Bye, A. Bleasel, J A Lawson, Voxel-
based morphometry in the detection of dysplasia and neoplasia in childhood
epilepsy : Combined grey/white matter analysis augments detection, Epilepsy
Research, 77:93-101,2007.
[27] H. Xue, L. Srinivasan, S. Jiang, M. Rutherford, AD Edwards, D Rueckert, JV
Hajnal, Automatic segmentation and reconstruction of the cortex from neonatal
MRI, Neuroimage,38: 461-477, Nov 2007.