Flexible cystoscopy is an examination that allows physicians to look inside the bladder. In flexible cystoscopy, beginner physicians tend to lose track of the observation due to complex handling patterns of a flexible cystoscope and poor characteristics of the bladder. In this paper, as a diagnostic support tool for beginner physicians in flexible cystoscopy, we propose a system for tracking the observation using cystoscopic images. Our system discriminates three handling patterns of a flexible cystoscope, namely bending, rotation, or insertion. To discriminate the handling patterns accurately, we propose to use the degree of bending, rotation, or insertion as features for the discrimination as well as ZNCC-based optical flows. These features are learned by a Random Forest classifier. The classifier discriminates sequential handling patterns of the cystoscope by a time-series analysis. Experimental results on ten videos obtained in flexible cystoscopy show that each of the three handling patterns were correctly discriminated over 90% in average. In addition, we reproduced the observation in a virtual bladder we propose.
The Indian Dental Academy is the Leader in continuing dental education , training dentists in all aspects of dentistry and
offering a wide range of dental certified courses in different formats.for more details please visit
www.indiandentalacademy.com
ASSESSING THE EFFECT OF UNICONDYLAR KNEE ARTHROPLASTY ON PROXIMAL TIBIA BONE ...ijbesjournal
ABSTRACT
In order to develop computational models of implanted constructs to predict prosthesis performance,robust experimental tests need to be devised. In the case of unicondylar knee arthroplasty (UKA), whereuptake of the procedure has been relatively low compared to traditional total knee arthroplasty techniques,computational modelling can give an insight into the factors affect theperformance of UKA if verified withappropriate, preferably data rich, experimental simulations. In the present work, an image based strainanalysis technique was applied for the assessment of the effect of UKA implantation on the strainsdeveloped in cortical bone of the proximal tibia. The results indicated the presence of increased strains inthe proximal portion of the bone, which could be exacerbated in the case of poor implant positioning, or for patients with diminished bone quality.
KEYWORDS
Unicondylar Knee Arthroplasty, Orthopaedics, Implantation, Cadaver Bone, Strain, Digital ImageCorrelation
Reeb Graph for Automatic 3D CephalometryCSCJournals
The purpose of this study is to present a method of three-dimensional computed tomographic (3D-CT) cephalometrics and its use to study cranio/maxilla-facial malformations. We propose a system for automatic localization of cephalometric landmarks using reeb graphs. Volumetric images of a patient were reconstructed into 3D mesh. The proposed method is carried out in three steps: we begin by applying 3d mesh skull simplification, this mesh was reconstructed from a head volumetric medical image, and then we extract a reeb graph. Reeb graph mesh extraction represents a skeleton composed in a number of nodes and arcs. We are interested in the node position; we noted that some reeb nodes could be considered as cephalometric landmarks under specific conditions. The third step is to identify these nodes automatically by using elastic mesh registration using “thin plate” transformation and clustering. Preliminary results show a landmarks recognition rate of more than 90%, very close to the manually provided landmarks positions made by a medical stuff.
Proposition of local automatic algorithm for landmark detection in 3D cephalo...journalBEEI
This study proposes a new contribution to solve the problem of automatic landmarks detection in three-dimensional cephalometry. 3D images obtained from CBCT (cone beam computed tomography) equipment were used for automatic identification of twelve landmarks. The proposed method is based on a local geometry and intensity criteria of skull structures. After the step of preprocessing and binarization, the algorithm segments the skull into three structures using the geometry information of nasal cavity and intensity information of the teeth. Each targeted landmark was detected using local geometrical information of the volume of interest containing this landmark. The ICC and confidence interval (95% CI) for each direction were 0, 91 (0.75 to 0.96) for x- direction; 0.92 (0.83 to 0.97) for y-direction; 0.92 (0.79 to 0.97) for z-direction. The mean error of detection was calculated using the Euclidian distance between the 3D coordinates of manually and automatically detected landmarks. The overall mean error of the algorithm was 2.76 mm with a standard deviation of 1.43 mm. Our proposed approach for automatic landmark identification in 3D cephalometric was capable of detecting 12 landmarks on 3D CBCT images which can be facilitate the use of 3D cephalometry to orthodontists.
The Indian Dental Academy is the Leader in continuing dental education , training dentists in all aspects of dentistry and
offering a wide range of dental certified courses in different formats.for more details please visit
www.indiandentalacademy.com
ASSESSING THE EFFECT OF UNICONDYLAR KNEE ARTHROPLASTY ON PROXIMAL TIBIA BONE ...ijbesjournal
ABSTRACT
In order to develop computational models of implanted constructs to predict prosthesis performance,robust experimental tests need to be devised. In the case of unicondylar knee arthroplasty (UKA), whereuptake of the procedure has been relatively low compared to traditional total knee arthroplasty techniques,computational modelling can give an insight into the factors affect theperformance of UKA if verified withappropriate, preferably data rich, experimental simulations. In the present work, an image based strainanalysis technique was applied for the assessment of the effect of UKA implantation on the strainsdeveloped in cortical bone of the proximal tibia. The results indicated the presence of increased strains inthe proximal portion of the bone, which could be exacerbated in the case of poor implant positioning, or for patients with diminished bone quality.
KEYWORDS
Unicondylar Knee Arthroplasty, Orthopaedics, Implantation, Cadaver Bone, Strain, Digital ImageCorrelation
Reeb Graph for Automatic 3D CephalometryCSCJournals
The purpose of this study is to present a method of three-dimensional computed tomographic (3D-CT) cephalometrics and its use to study cranio/maxilla-facial malformations. We propose a system for automatic localization of cephalometric landmarks using reeb graphs. Volumetric images of a patient were reconstructed into 3D mesh. The proposed method is carried out in three steps: we begin by applying 3d mesh skull simplification, this mesh was reconstructed from a head volumetric medical image, and then we extract a reeb graph. Reeb graph mesh extraction represents a skeleton composed in a number of nodes and arcs. We are interested in the node position; we noted that some reeb nodes could be considered as cephalometric landmarks under specific conditions. The third step is to identify these nodes automatically by using elastic mesh registration using “thin plate” transformation and clustering. Preliminary results show a landmarks recognition rate of more than 90%, very close to the manually provided landmarks positions made by a medical stuff.
Proposition of local automatic algorithm for landmark detection in 3D cephalo...journalBEEI
This study proposes a new contribution to solve the problem of automatic landmarks detection in three-dimensional cephalometry. 3D images obtained from CBCT (cone beam computed tomography) equipment were used for automatic identification of twelve landmarks. The proposed method is based on a local geometry and intensity criteria of skull structures. After the step of preprocessing and binarization, the algorithm segments the skull into three structures using the geometry information of nasal cavity and intensity information of the teeth. Each targeted landmark was detected using local geometrical information of the volume of interest containing this landmark. The ICC and confidence interval (95% CI) for each direction were 0, 91 (0.75 to 0.96) for x- direction; 0.92 (0.83 to 0.97) for y-direction; 0.92 (0.79 to 0.97) for z-direction. The mean error of detection was calculated using the Euclidian distance between the 3D coordinates of manually and automatically detected landmarks. The overall mean error of the algorithm was 2.76 mm with a standard deviation of 1.43 mm. Our proposed approach for automatic landmark identification in 3D cephalometric was capable of detecting 12 landmarks on 3D CBCT images which can be facilitate the use of 3D cephalometry to orthodontists.
Automated Detection of Optic Disc in Retinal FundusImages Using PCAiosrjce
IOSR Journal of Pharmacy and Biological Sciences(IOSR-JPBS) is a double blind peer reviewed International Journal that provides rapid publication (within a month) of articles in all areas of Pharmacy and Biological Science. The journal welcomes publications of high quality papers on theoretical developments and practical applications in Pharmacy and Biological Science. Original research papers, state-of-the-art reviews, and high quality technical notes are invited for publications.
A MEMS BASED OPTICAL COHERENCE TOMOGRAPHY IMAGING SYSTEM AND OPTICAL BIOPSY P...Ping Hsu
A fully-functional, real-time optical coherence tomography (OCT) system based on a high-speed, gimbal-less micromachined scanning
mirror is presented. The designed MEMS control architecture allows the MEMS based imaging probes to be connected to a time-domain, a
Fourier domain or a spectral domain OCT system. Furthermore, a variety of probes optimized for specific laboratory or clinical
applications including various minimally invasive endoscopic, handheld or lab-bench mounted probes may be switched between effortlessly
and important driving parameters adjusted in real-time. In addition, artifact free imaging speeds of 33μs per voxel have been achieved
while imaging a 1.4mm×1.4mm×1.4mm region with 5μm×5μm×5μm sampling resolution (SD-OCT system.)
Lec6: Pre-Processing for Nuclear Medicine ImagesUlaş Bağcı
2017 Spring, UCF Medical Image Computing
1. The use of PET/SPECT, PET/CT and MRI/PET Images
2. What to measure from Nuclear Medicine Images?
3. Denoising Nuclear MedicineI mages
4. PartialVolumeCorrection
ImageEnhancement • Filtering • Smoothing • Introduction to Medical Image Computing and Toolkits • Image Filtering, Enhancement, Noise Reduction, and Signal Processing • MedicalImageRegistration • MedicalImageSegmentation • MedicalImageVisualization • Machine Learning in Medical Imaging • Shape Modeling/Analysis of Medical Images Deep Learning in Radiology
PERFORMANCE EVALUATION OF TUMOR DETECTION TECHNIQUES ijcsa
Automatic segmentation of tumor plays a vital role in diagnosis and surgical planning. This paper deals
with techniques which providing solution for detecting hepatic tumor in Computed Tomography (CT)
images. The main aim of this work is to analyze performance of tumor detection techniques like Knowledge
Based Constraints, Graph Cut Method and Gradient Vector Flow active contour. These three techniques
are computed using sensitivity, specificity and accuracy. From the evaluated result, knowledge based
constraints method is better than other graph cut method and gradient vector flow active contour.
Medical Image Synthesis with Improved Cycle-GAN: CT from CECT BoahKim2
Presentation file for "Medical Image Synthesis with Improved Cycle-GAN: CT from CECT" presented at the Workshop on Deep Learning for Biomedical Image Reconstruction of IEEE Internetional Symposium on Biomedical Imaging, ISBI 2020.
Optic Disc and Macula Localization from Retinal Optical Coherence Tomography ...IJECEIAES
This research used images from Optical Coherence Tomography (OCT) examination as well as fundus images to localize the optical disc and macular layer of retina. The researchers utilized the OCT and fundus image to interpret the distance between macular center and optic disc in the image. The distance will express the area of macula that can be employed for further research. This distance could recognize the thickness of macula parameters diameter that will be used in localizing process of optic disc and macula. The parameters are the circle radius, the size of window’s filter, the constant value and the size of optic disc element structure as well as the size of macula. The results of this study are expected to improve the accuracy of macula detection that experience the edema.
Lec8: Medical Image Segmentation (II) (Region Growing/Merging)Ulaş Bağcı
. Region Growing algorithm
• Homogeneity Criteria
• Split/Merge algorithm
• Examples from CT, MRI, PET
• Limitations
Image Filtering, Enhancement, Noise Reduction, and Signal Processing • MedicalImageRegistration • MedicalImageSegmentation • MedicalImageVisualization • Machine Learning in Medical Imaging • Shape Modeling/Analysis of Medical Images Deep Learning in Radiology
Unsupervised Deformable Image Registration Using Cycle-Consistent CNNBoahKim2
Presentation file for "Unsupervised Deformable Image Registration Using Cycle-Consistent CNN" presented at the International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2019.
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.
Brain Tumor Area Calculation in CT-scan image using Morphological Operationsiosrjce
IOSR Journal of Computer Engineering (IOSR-JCE) is a double blind peer reviewed International Journal that provides rapid publication (within a month) of articles in all areas of computer engineering and its applications. The journal welcomes publications of high quality papers on theoretical developments and practical applications in computer technology. Original research papers, state-of-the-art reviews, and high quality technical notes are invited for publications.
Segmentation of cysts in kidney and 3 d volume calculation from ct imagesbioejjournal
Statistics based optimization, Plackett–Burman design (PBD) and response surface methodology
(RSM) were employed to screen and optimize the media components for the production of
clavulanic acid from Streptomyces clavuligerus MTCC 1142, using solid state fermentation. jackfruit
seed powder was used as both the solid support and carbon source for the growth of Streptomyces
clavuligerus MTCC 1142. Based on the positive influence of the Pareto chart obtained from PBD on
clavulanic acid production, five media components – yeast extract, beef extract, sucrose, malt extract
and ferric chloride were screened. Central composite design (CCD) was employed using these five
media components- yeast extract 2.5%, beef extract 0.5%, sucrose 2.5%, malt extract 0.25% and ferric
chloride nutritional factors at three levels, for further optimization, and the second order polynomial
equation was derived, based on the experimental data. Response surface methodology showed that
the concentrations of yeast extract 2.5%, beef extract 0.5%, sucrose 2.5%, malt extract 0.25% and ferric
chloride 2.5% were the optimal levels for maximal clavulanic acid production (19.37 mg /gds) which were validated through experiments.
Automated Detection of Optic Disc in Retinal FundusImages Using PCAiosrjce
IOSR Journal of Pharmacy and Biological Sciences(IOSR-JPBS) is a double blind peer reviewed International Journal that provides rapid publication (within a month) of articles in all areas of Pharmacy and Biological Science. The journal welcomes publications of high quality papers on theoretical developments and practical applications in Pharmacy and Biological Science. Original research papers, state-of-the-art reviews, and high quality technical notes are invited for publications.
A MEMS BASED OPTICAL COHERENCE TOMOGRAPHY IMAGING SYSTEM AND OPTICAL BIOPSY P...Ping Hsu
A fully-functional, real-time optical coherence tomography (OCT) system based on a high-speed, gimbal-less micromachined scanning
mirror is presented. The designed MEMS control architecture allows the MEMS based imaging probes to be connected to a time-domain, a
Fourier domain or a spectral domain OCT system. Furthermore, a variety of probes optimized for specific laboratory or clinical
applications including various minimally invasive endoscopic, handheld or lab-bench mounted probes may be switched between effortlessly
and important driving parameters adjusted in real-time. In addition, artifact free imaging speeds of 33μs per voxel have been achieved
while imaging a 1.4mm×1.4mm×1.4mm region with 5μm×5μm×5μm sampling resolution (SD-OCT system.)
Lec6: Pre-Processing for Nuclear Medicine ImagesUlaş Bağcı
2017 Spring, UCF Medical Image Computing
1. The use of PET/SPECT, PET/CT and MRI/PET Images
2. What to measure from Nuclear Medicine Images?
3. Denoising Nuclear MedicineI mages
4. PartialVolumeCorrection
ImageEnhancement • Filtering • Smoothing • Introduction to Medical Image Computing and Toolkits • Image Filtering, Enhancement, Noise Reduction, and Signal Processing • MedicalImageRegistration • MedicalImageSegmentation • MedicalImageVisualization • Machine Learning in Medical Imaging • Shape Modeling/Analysis of Medical Images Deep Learning in Radiology
PERFORMANCE EVALUATION OF TUMOR DETECTION TECHNIQUES ijcsa
Automatic segmentation of tumor plays a vital role in diagnosis and surgical planning. This paper deals
with techniques which providing solution for detecting hepatic tumor in Computed Tomography (CT)
images. The main aim of this work is to analyze performance of tumor detection techniques like Knowledge
Based Constraints, Graph Cut Method and Gradient Vector Flow active contour. These three techniques
are computed using sensitivity, specificity and accuracy. From the evaluated result, knowledge based
constraints method is better than other graph cut method and gradient vector flow active contour.
Medical Image Synthesis with Improved Cycle-GAN: CT from CECT BoahKim2
Presentation file for "Medical Image Synthesis with Improved Cycle-GAN: CT from CECT" presented at the Workshop on Deep Learning for Biomedical Image Reconstruction of IEEE Internetional Symposium on Biomedical Imaging, ISBI 2020.
Optic Disc and Macula Localization from Retinal Optical Coherence Tomography ...IJECEIAES
This research used images from Optical Coherence Tomography (OCT) examination as well as fundus images to localize the optical disc and macular layer of retina. The researchers utilized the OCT and fundus image to interpret the distance between macular center and optic disc in the image. The distance will express the area of macula that can be employed for further research. This distance could recognize the thickness of macula parameters diameter that will be used in localizing process of optic disc and macula. The parameters are the circle radius, the size of window’s filter, the constant value and the size of optic disc element structure as well as the size of macula. The results of this study are expected to improve the accuracy of macula detection that experience the edema.
Lec8: Medical Image Segmentation (II) (Region Growing/Merging)Ulaş Bağcı
. Region Growing algorithm
• Homogeneity Criteria
• Split/Merge algorithm
• Examples from CT, MRI, PET
• Limitations
Image Filtering, Enhancement, Noise Reduction, and Signal Processing • MedicalImageRegistration • MedicalImageSegmentation • MedicalImageVisualization • Machine Learning in Medical Imaging • Shape Modeling/Analysis of Medical Images Deep Learning in Radiology
Unsupervised Deformable Image Registration Using Cycle-Consistent CNNBoahKim2
Presentation file for "Unsupervised Deformable Image Registration Using Cycle-Consistent CNN" presented at the International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2019.
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.
Brain Tumor Area Calculation in CT-scan image using Morphological Operationsiosrjce
IOSR Journal of Computer Engineering (IOSR-JCE) is a double blind peer reviewed International Journal that provides rapid publication (within a month) of articles in all areas of computer engineering and its applications. The journal welcomes publications of high quality papers on theoretical developments and practical applications in computer technology. Original research papers, state-of-the-art reviews, and high quality technical notes are invited for publications.
Segmentation of cysts in kidney and 3 d volume calculation from ct imagesbioejjournal
Statistics based optimization, Plackett–Burman design (PBD) and response surface methodology
(RSM) were employed to screen and optimize the media components for the production of
clavulanic acid from Streptomyces clavuligerus MTCC 1142, using solid state fermentation. jackfruit
seed powder was used as both the solid support and carbon source for the growth of Streptomyces
clavuligerus MTCC 1142. Based on the positive influence of the Pareto chart obtained from PBD on
clavulanic acid production, five media components – yeast extract, beef extract, sucrose, malt extract
and ferric chloride were screened. Central composite design (CCD) was employed using these five
media components- yeast extract 2.5%, beef extract 0.5%, sucrose 2.5%, malt extract 0.25% and ferric
chloride nutritional factors at three levels, for further optimization, and the second order polynomial
equation was derived, based on the experimental data. Response surface methodology showed that
the concentrations of yeast extract 2.5%, beef extract 0.5%, sucrose 2.5%, malt extract 0.25% and ferric
chloride 2.5% were the optimal levels for maximal clavulanic acid production (19.37 mg /gds) which were validated through experiments.
Apartamento 511 m² a venda no Condomínio 106 Seridó - Jardim Paulistano. 6 vagas de garagem + Depósito.
Pronto, com acabamentos de altísssimo padrão
Piso em madeira Parquet - taco ebanizado Versailles, Composição em porcelanato xadrez na cozinha e áreas de serviço, Armários de Cozinha Kitchens, Equipamentos de Cozinha Vicking, Armários, Closets e espaços inteligentes Ornare, acabamentos diferenciado nos banheiros das suites, fechamento basculante das varandas,portas blindadas da entrada social e de serviço, portas internas em madeira maciça e ferragens italianas, sistema total de ar condicionado Daikin em todo apartamento, automação e muito mais detalhes.
Contato : Humberto (11) 94731-0790
FORMA Itaim
Um empreendimento para você ficar ao lado de tudo o que faz parte do seu mundo: dia e noite. O Forma Itaim está entre dois parques, em uma área de comércio variado, shopping centers, serviços 24 horas, e próximo a empresas de todos os portes. A duas quadras da Av. Brig. Faria Lima, nas imediações da esquina com a Av. Juscelino Kubitschek. Uma região bem servida de transportes, onde é possível chegar a vários lugares a pé ou usando a ciclovia.
1 Torre
25 Andares
123 Unidades
4 Elevadores
131 Vagas
1.886 m² Área do Terreno
1 ou 2 Vagas
Studio 45 m² - 108 Unidades
Cobertura 90 m² - 3 Unidades
Duplex 66 m² - 6 Unidades
Gardens de 49 m² A 140 m² - 6 Unidades
Todas as unidades apresentam vagas independentes*.
Para maiores informações e agendamentos de visitas entre em contato:
Corretor Humberto - Lopes - 11 98588-8082
about.me/corretorhumberto
Application of 3D Modeling for Preoperative Planning and Intra Operative Navigation during Procedures on the Organs of Retroperitoneal Space by Alexey A Rozhentsov in Experimental Techniques in Urology & Nephrology
MEDICAL IMAGE PROCESSING METHODOLOGY FOR LIVER TUMOUR DIAGNOSISijsc
Apply the Image processing techniques to analyse the medical images may assist medical professionals as well as patients, especially in this research apply the algorithms to diagnose the liver tumours from the abdominal CT image. This research proposes a software solution to illustrate the automated liver
segmentation and tumour detection using artificial intelligent techniques. Evaluate the results of the liver segmentation and tumour detection, in-cooperation with the radiologists by using the prototype of the proposed system. This research overcomes the challenges in medical image processing. The 100 samples
collected from ten patients and received 90% accuracy rate.
Medical Image Processing Methodology for Liver Tumour Diagnosis ijsc
Apply the Image processing techniques to analyse the medical images may assist medical professionals as well as patients, especially in this research apply the algorithms to diagnose the liver tumours from the abdominal CT image. This research proposes a software solution to illustrate the automated liver segmentation and tumour detection using artificial intelligent techniques. Evaluate the results of the liver segmentation and tumour detection, in-cooperation with the radiologists by using the prototype of the proposed system. This research overcomes the challenges in medical image processing. The 100 samples collected from ten patients and received 90% accuracy rate.
The Indian Dental Academy is the Leader in continuing dental education , training dentists in all aspects of dentistry and
offering a wide range of dental certified courses in different formats.for more details please visit
www.indiandentalacademy.com
Reliability of Three-dimensional Photonic Scanner Anthropometry Performed by ...CSCJournals
This work explored the relative and absolute reliability of three-dimensional (3D) anthropometry performed by skilled and naïve operators using a fast, pose tolerant whole-body 3D scanner device. Upon skin landmarking by an experienced operator (skilled anthropometrist, SA), twelve subjects (six males and six females) underwent a thorough 3D anthropometric evaluation by the SA and two naïve operators (NA). Using the same landmarks, the SA also performed traditional anthropometry measurements. All measurements were taken twice. Relative reliability was tested with the Pearson’s correlation coefficient r and the intraclass correlation coefficient (ICC); absolute reliability was tested calculating the percentage coefficient of variation (%CV), the standard error of measurement (SEM), the percentage technical error of measurement (%TEM), and paired Student’s t test. Results showed that intra-operator relative and absolute reliability was excellent for all and most 3D measurement items, respectively, independently of the operator’s skill. Inter-operator (SA vs. individual NA) relative reliability was excellent as well; inter-operator absolute reliability was not acceptable for about only 30% of measurement items. Results of this work show that 3D anthropometry has strong potential in anthropometry due to high intrinsic reliability and less need for operator training vs. traditional anthropometry.
Statistical Feature-based Neural Network Approach for the Detection of Lung C...CSCJournals
Lung cancer, if successfully detected at early stages, enables many treatment options, reduced risk of invasive surgery and increased survival rate. This paper presents a novel approach to detect lung cancer from raw chest X-ray images. At the first stage, we use a pipeline of image processing routines to remove noise and segment the lung from other anatomical structures in the chest X-ray and extract regions that exhibit shape characteristics of lung nodules. Subsequently, first and second order statistical texture features are considered as the inputs to train a neural network to verify whether a region extracted in the first stage is a nodule or not . The proposed approach detected nodules in the diseased area of the lung with an accuracy of 96% using the pixel-based technique while the feature-based technique produced an accuracy of 88%.
The implementation of MDCT in urological imaging has solved much of the diagnostic dilemma. Thanks to its multiplanar capabilities and post processing techniques.
Liver extraction using histogram and morphologyeSAT Journals
Abstract
Liver is the largest glandular organ important for survival in human body. Computed tomography is generally used to image liver
due to its precision. This paper presents a method to extract liver from computed tomography (CT) abdomen images in axial
orientation. A traditional segmentation method based on histogram and morphology is proposed herein. Histogram is used to
analyze the intensity distribution, morphological operations are used to disconnect liver from the neighboring organs and greatest
connected pixels are extracted. The experimental results of the proposed method when applied to CT abdomen images with
contrast are presented and the effectiveness is discussed in accordance to the manual tracing obtained from the radiologist. Dice
similarity co-efficient amounts to 94% in the proposed method.
Keywords: Segmentation, Extraction, Histogram, Morphology, Connected Component, CT Liver
The International Journal of Engineering and Science (The IJES)theijes
The International Journal of Engineering & Science is aimed at providing a platform for researchers, engineers, scientists, or educators to publish their original research results, to exchange new ideas, to disseminate information in innovative designs, engineering experiences and technological skills. It is also the Journal's objective to promote engineering and technology education. All papers submitted to the Journal will be blind peer-reviewed. Only original articles will be published.
The papers for publication in The International Journal of Engineering& Science are selected through rigorous peer reviews to ensure originality, timeliness, relevance, and readability.
Similar to 3D Position Tracking System for Flexible Cystoscopy (20)
Model Attribute Check Company Auto PropertyCeline George
In Odoo, the multi-company feature allows you to manage multiple companies within a single Odoo database instance. Each company can have its own configurations while still sharing common resources such as products, customers, and suppliers.
June 3, 2024 Anti-Semitism Letter Sent to MIT President Kornbluth and MIT Cor...Levi Shapiro
Letter from the Congress of the United States regarding Anti-Semitism sent June 3rd to MIT President Sally Kornbluth, MIT Corp Chair, Mark Gorenberg
Dear Dr. Kornbluth and Mr. Gorenberg,
The US House of Representatives is deeply concerned by ongoing and pervasive acts of antisemitic
harassment and intimidation at the Massachusetts Institute of Technology (MIT). Failing to act decisively to ensure a safe learning environment for all students would be a grave dereliction of your responsibilities as President of MIT and Chair of the MIT Corporation.
This Congress will not stand idly by and allow an environment hostile to Jewish students to persist. The House believes that your institution is in violation of Title VI of the Civil Rights Act, and the inability or
unwillingness to rectify this violation through action requires accountability.
Postsecondary education is a unique opportunity for students to learn and have their ideas and beliefs challenged. However, universities receiving hundreds of millions of federal funds annually have denied
students that opportunity and have been hijacked to become venues for the promotion of terrorism, antisemitic harassment and intimidation, unlawful encampments, and in some cases, assaults and riots.
The House of Representatives will not countenance the use of federal funds to indoctrinate students into hateful, antisemitic, anti-American supporters of terrorism. Investigations into campus antisemitism by the Committee on Education and the Workforce and the Committee on Ways and Means have been expanded into a Congress-wide probe across all relevant jurisdictions to address this national crisis. The undersigned Committees will conduct oversight into the use of federal funds at MIT and its learning environment under authorities granted to each Committee.
• The Committee on Education and the Workforce has been investigating your institution since December 7, 2023. The Committee has broad jurisdiction over postsecondary education, including its compliance with Title VI of the Civil Rights Act, campus safety concerns over disruptions to the learning environment, and the awarding of federal student aid under the Higher Education Act.
• The Committee on Oversight and Accountability is investigating the sources of funding and other support flowing to groups espousing pro-Hamas propaganda and engaged in antisemitic harassment and intimidation of students. The Committee on Oversight and Accountability is the principal oversight committee of the US House of Representatives and has broad authority to investigate “any matter” at “any time” under House Rule X.
• The Committee on Ways and Means has been investigating several universities since November 15, 2023, when the Committee held a hearing entitled From Ivory Towers to Dark Corners: Investigating the Nexus Between Antisemitism, Tax-Exempt Universities, and Terror Financing. The Committee followed the hearing with letters to those institutions on January 10, 202
Honest Reviews of Tim Han LMA Course Program.pptxtimhan337
Personal development courses are widely available today, with each one promising life-changing outcomes. Tim Han’s Life Mastery Achievers (LMA) Course has drawn a lot of interest. In addition to offering my frank assessment of Success Insider’s LMA Course, this piece examines the course’s effects via a variety of Tim Han LMA course reviews and Success Insider comments.
Macroeconomics- Movie Location
This will be used as part of your Personal Professional Portfolio once graded.
Objective:
Prepare a presentation or a paper using research, basic comparative analysis, data organization and application of economic information. You will make an informed assessment of an economic climate outside of the United States to accomplish an entertainment industry objective.
Unit 8 - Information and Communication Technology (Paper I).pdfThiyagu K
This slides describes the basic concepts of ICT, basics of Email, Emerging Technology and Digital Initiatives in Education. This presentations aligns with the UGC Paper I syllabus.
2024.06.01 Introducing a competency framework for languag learning materials ...Sandy Millin
http://sandymillin.wordpress.com/iateflwebinar2024
Published classroom materials form the basis of syllabuses, drive teacher professional development, and have a potentially huge influence on learners, teachers and education systems. All teachers also create their own materials, whether a few sentences on a blackboard, a highly-structured fully-realised online course, or anything in between. Despite this, the knowledge and skills needed to create effective language learning materials are rarely part of teacher training, and are mostly learnt by trial and error.
Knowledge and skills frameworks, generally called competency frameworks, for ELT teachers, trainers and managers have existed for a few years now. However, until I created one for my MA dissertation, there wasn’t one drawing together what we need to know and do to be able to effectively produce language learning materials.
This webinar will introduce you to my framework, highlighting the key competencies I identified from my research. It will also show how anybody involved in language teaching (any language, not just English!), teacher training, managing schools or developing language learning materials can benefit from using the framework.
Introduction to AI for Nonprofits with Tapp NetworkTechSoup
Dive into the world of AI! Experts Jon Hill and Tareq Monaur will guide you through AI's role in enhancing nonprofit websites and basic marketing strategies, making it easy to understand and apply.
Embracing GenAI - A Strategic ImperativePeter Windle
Artificial Intelligence (AI) technologies such as Generative AI, Image Generators and Large Language Models have had a dramatic impact on teaching, learning and assessment over the past 18 months. The most immediate threat AI posed was to Academic Integrity with Higher Education Institutes (HEIs) focusing their efforts on combating the use of GenAI in assessment. Guidelines were developed for staff and students, policies put in place too. Innovative educators have forged paths in the use of Generative AI for teaching, learning and assessments leading to pockets of transformation springing up across HEIs, often with little or no top-down guidance, support or direction.
This Gasta posits a strategic approach to integrating AI into HEIs to prepare staff, students and the curriculum for an evolving world and workplace. We will highlight the advantages of working with these technologies beyond the realm of teaching, learning and assessment by considering prompt engineering skills, industry impact, curriculum changes, and the need for staff upskilling. In contrast, not engaging strategically with Generative AI poses risks, including falling behind peers, missed opportunities and failing to ensure our graduates remain employable. The rapid evolution of AI technologies necessitates a proactive and strategic approach if we are to remain relevant.
How to Make a Field invisible in Odoo 17Celine George
It is possible to hide or invisible some fields in odoo. Commonly using “invisible” attribute in the field definition to invisible the fields. This slide will show how to make a field invisible in odoo 17.
3D Position Tracking System for Flexible Cystoscopy
1. Munehiro Nakamura, Yusuke Kajiwara, Tatsuhito Hasegawa & Haruhiko Kimura
International Journal of Image Processing (IJIP), Volume (7) : Issue (4) : 2013 418
3D Position Tracking System for Flexible Cystoscopy
Munehiro Nakamura m-nakamura@blitz.ec.t.kanazawa-u.ac.jp
Department of Natural Science and Engineering
Kanazawa University
Kanazawa, 9201192, Japan
Yusuke Kajiwara kajiwara@de.is.ritsumei.ac.jp
Department of Information Science
Ritsumeikan University
Kusatsu, 525877, Japan
Tatsuhito Hasegawa t-hasegawa@blitz.ec.t.kanazawa-u.ac.jp
Department of Natural Science and Engineering
Kanazawa University
Kanazawa, 9201192, Japan
Haruhiko Kimura kimura@blitz.ec.t.kanazawa-u.ac.jp
Department of Natural Science and Engineering
Kanazawa University
Kanazawa, 9201192, Japan
Abstract
Flexible cystoscopy is an examination that allows physicians to look inside the bladder. In flexible
cystoscopy, beginner physicians tend to lose track of the observation due to complex handling
patterns of a flexible cystoscope and poor characteristics of the bladder. In this paper, as a
diagnostic support tool for beginner physicians in flexible cystoscopy, we propose a system for
tracking the observation using cystoscopic images. Our system discriminates three handling
patterns of a flexible cystoscope, namely bending, rotation, or insertion. To discriminate the
handling patterns accurately, we propose to use the degree of bending, rotation, or insertion as
features for the discrimination as well as ZNCC-based optical flows. These features are learned
by a Random Forest classifier. The classifier discriminates sequential handling patterns of the
cystoscope by a time-series analysis. Experimental results on ten videos obtained in flexible
cystoscopy show that each of the three handling patterns were correctly discriminated over 90%
in average. In addition, we reproduced the observation in a virtual bladder we propose.
Keywords: Flexible Cystoscopy, Position Tracking, Optical Flow, Zero-mean Normalized Cross-
Correlation, Handling Pattern.
1. INTRODUCTION
With increase of aged people in the world, incidents of bladder disease are gradually
increasing[1]. Bladder disease can be detected in cystoscopy or non-invasive examinations such
as blood test, MRI, CT, PET, and ultrasonography. The non-invasive examinations are painless
and less stressful. However, it is still difficult to detect tiny legions in a non-invasive imaging
examination[2]. Cystoscopy is conducted when a lesion found in a non-invasive imaging
examination or sever symptoms are appeared in a patient. Cystoscopy enables physicians to look
inside the bladder to confirm a patient's legion directly. There are two types of cystoscopes,
namely rigid and flexible. The examination with the latter one is less painful and is used widely. In
this paper, we deal the examination with flexible cystoscope.
In flexible cystoscopy, images of the bladder are obtained from a camera embedded in the tip of
flexible cystoscope and displayed in the monitor. However, there are three major difficulties for
physicians to check the whole inner of the bladder completely. First, since the bladder has similar
2. Munehiro Nakamura, Yusuke Kajiwara, Tatsuhito Hasegawa & Haruhiko Kimura
International Journal of Image Processing (IJIP), Volume (7) : Issue (4) : 2013 419
shape and color, sometimes beginner physicians lose track of the observation. Second,
cystoscopic images are sometimes unclear due to halation. Third, it requires some experiments
to control flexible cystoscope properly due to complex handling patterns of the equipment.
As a diagnostic support tool for beginner physicians in flexible cystoscopy, this paper presents a
system for tracking the observation. To achieve the objective, it is considered to attach
acceleration sensors or location sensors to a flexible cystoscope. In that case, it is necessary to
obtain an approval for usage of the cystoscope according to the pharmaceutical low, as well as to
buy the cystoscope which would be at least 10,000$. Another approach to track the observation
in cystoscopy is mapping observed regions in a 3D space. As a representative 3D tracking
system, Choi et al. proposed a robust segment-based object tracking system that uses the
backside of a car image[3]. Their system measures depth information by calculating the
enlargement factor of a target region. Ramisa et al. proposed to measure the distance between a
single camera and a person[4]. However, since cystoscopic images are significantly unclear than
the car image and human image, existing algorithms for 3D tracking[3, 4] could not be applied to
cystoscopic images.
In this paper, we propose to discriminate the handling patterns of a flexible cystoscope. First, the
proposed system extracts ZNCC-based optical flows[5] from cystoscopic images as features for
estimating the handling patterns. Next, various features including the ZNCC-based optical flows
are learned by a Random Forest classifier[6]. The classifier discriminates sequential handling
patterns of the cystoscope by a time-series analysis. Finally, the observation in flexible
cystoscopy is reproduced in a virtual 3D bladder we propose.
In section 2, we will introduce the process of flexible cystoscopy. In section 3, we will explain the
proposed system. In section 4, we will examine the performance of the proposed system
regarding the accuracy in estimating the handling patterns and tracking the observation in the
virtual bladder.
2. FLEXIBLE CYSTOSCOPY
The human bladder is a hollow and balloon shaped organ that is broadly distinguished into seven
regions; trigone, neck, left side wall, right side wall, posterior wall, dome, and anterior wall. Figure
1 shows images of the bladder. From the figure, we could perceive the images except neck are
similar to each other. In addition, sometimes cystoscopic images are noisy due to halation which
often occurs when the cystoscope is close to the bladder wall.
The flexible cystoscopy is conducted by a physician as below.
(1) The physician inserts a flexible cystoscope into a patient's urethra.
(2) The physician pushes the cystoscope slowly to the bladder.
(3) By adjusting the position of the cystoscope, the physician observes the whole inner of the
bladder.
(4) The physician pulls the cystoscope after checking all the regions.
Figure 2 shows three handling patterns of the cystoscope to adjust the position of the cystoscope
in step (3). The problem in this examination is that a beginner physician in the step (4) is
sometimes unsure that all the regions were completely observed. Since oversight of severe
legion would cause fatal case, diagnostic support tools for beginner physicians in flexible
cystoscopy have been required.
3. Munehiro Nakamura, Yusuke Kajiwara, Tatsuhito Hasegawa & Haruhiko Kimura
International Journal of Image Processing (IJIP), Volume (7) : Issue (4) : 2013 420
3. PROPOSED SYSTEM
3.1 Overview
As mentioned in Sec. 2, the bladder can be distinguished into seven regions. However, the
definitive region of each part is not defined. Considering a normal bladder, we construct a sphere
bladder model which has seven regions. Figure 3 shows the virtual bladder we propose. And,
Table 1 shows definitions of each region for the virtual bladder, which determined by an expert
physician in flexible cystoscopy.
3.2 Preprocessing
Figure 4 shows the interface for a flexible cystoscopy. In the proposed system, first, ROI (Region
of Interest) is set on the rectangle in the interface. The size of ROI is 300 pixel × 300 pixel. Next,
Figure 5 (b) shows the cystoscopic image applied 8-bit gray scale transformation to Figure 5 (a).
And,
trigone neck
left side wall right side wall
posterior wall dome anterior wall
FIGURE 1: Example of Images Obtained from a Flexible Cystoscope.
4. Munehiro Nakamura, Yusuke Kajiwara, Tatsuhito Hasegawa & Haruhiko Kimura
International Journal of Image Processing (IJIP), Volume (7) : Issue (4) : 2013 421
FIGURE 2: Three Handling Patterns of a Flexible Cystoscope.
Figure 5 (c) shows the image applied the histogram stretching to Figure 5 (b). Finally, Figure 5 (d)
shows the image applied the selective local averaging to Figure 5 (c). Compared with Figure 5
(b), we could perceive that blood vessels are enhanced in the images of Figure 5 (d).
3.3 Extraction of Optical Flows
Approaches of well-known object tracking can be distinguished into gradient method and block
matching method. Gradient method is effective on videos where target objects move slowly[7].
FIGURE 3: 3D Virtual Bladder We Propose.
5. Munehiro Nakamura, Yusuke Kajiwara, Tatsuhito Hasegawa & Haruhiko Kimura
International Journal of Image Processing (IJIP), Volume (7) : Issue (4) : 2013 422
TABLE 1: Definition of the Virtual Bladder Map.
Meanwhile, block matching method is robust to quick movement of target objects, sudden
illumination changes, and noises. Since cystoscopic images are often unclear due to noises such
as halation and focus error, the proposed method applies the block matching[8]. As a robust
method to measure the movements of a flexible cystoscope, the proposed method extracts
ZNCC-based optical flows from consecutive cystoscopic images. In ZNCC (Zero-mean
Normalized Cross-Correlation), an image is divided into A × B blocks and optical flows are
extracted from each block by the following formula
1 1
2 2
1 1 1 1
( ( , ) )( ( , ) )
(1)
( ( , ) ) ( ( , ) )
N M
i j
ZNCC N M N M
i j i j
I i j I T i j T
R
I i j I T i j T
− −
= =
− −
= = = =
− −
=
− × −
∑∑
∑∑ ∑∑
where I (i, j) is the pixel value at ith row and jth column in the template image, T (i, j) is the pixel
value at ith row and jth column in the target image, N is the search range towards x axis, and M is
FIGURE 4: Interface for the Examination with a Flexible Cystoscope.
Bladder region Definition
Dome More than lat. 70 degrees S, excluding the trigone defined below.
Bladder neck More than lat. 70 degrees S, excluding the trigone defined below.
Trigone Triangle part is surrounded with lat. 45 degrees S, a parallel of lines of
longitude of long. 60 degrees E / long. 60 degrees W.
Posterior wall The quadrangle surrounded with a parallel of lat. 60 degrees N/ lat. 70
degrees S, a line of longitude of long. 60 degrees E/ long. 60 degrees W.
Anterior wall The quadrangle surrounded with a parallel of lat. 60 degrees N/ lat. 70
degrees S, a line of longitude of long. 120 degrees E/ long. 120 degrees
W.
Right side wall The reminded region in the east side.
Left side wall The reminded region in the west side.
6. Munehiro Nakamura, Yusuke Kajiwara, Tatsuhito Hasegawa & Haruhiko Kimura
International Journal of Image Processing (IJIP), Volume (7) : Issue (4) : 2013 423
(a) (b) (c) (d)
FIGURE 5: Preprocessing for the Enhancement of Blood Vessels.
the search range towards y axis. As an algorithm of ZNCC, the proposed system uses the one
proposed by Yoo and Han[5].
3.4 Discrimination of Handling Patterns for a Flexible Cystoscope
Table 2 shows handling techniques of the flexible cystoscope. From the table, we can find that
the combination of the handling techniques is up to 27 patterns. And, figure 6 shows an example
of optical flows obtained by rotation (a) and insertion (b) when the cystoscope is 90 degrees down
or zero degrees or 90 degrees up. From figure 6, we can find that the ZNCC-based optical flows
are depended on the degree of rotation and insertion.
We define three features that improve the discrimination of the handling patterns as below
1
(2)
t
t i
i
Rot R
=
= ∑
1
(3)
t
t i
i
Bend B
=
= ∑
Handling pattern
rotation left right neutral
bending up down neutral
insertion push pull neutral
TABLE 2: Handling Patterns of Flexible Cystoscope.
7. Munehiro Nakamura, Yusuke Kajiwara, Tatsuhito Hasegawa & Haruhiko Kimura
International Journal of Image Processing (IJIP), Volume (7) : Issue (4) : 2013 424
FIGURE 6: Example of Optical Flows Obtained by Rotation (a) and Insertion (b).
1
(4)
t
t i
i
Ins I
=
= ∑
where Rott is the degree of rotation at tth frame, Ri is the handling pattern of rotation at ith frame
and returns 1 when Ri=left, -1 when Ri=right, and 0 when Ri=neutral. Similarly, Bendt is the
degree of bending at tth frame, Bi is the handling pattern of bending at ith frame and returns 1
when Bi=up, -1 when Bi=down, and 0 when Bi=neutral, and Inst is the degree of insertion at tth
frame, Ii is the handling pattern of insertion at ith frame and returns 1 when Ii=push, -1 when
Ii=pull, and 0 when Ii=neutral.
Thus, the proposed system extracts features as A × B optical flows, Rott, Bendt, and Inst from a
cystoscopic image. By feeding all the features, a RF (random forest) classifier discriminates the
27 handling patterns frame by frame. RF is a noise-robust classification algorithm proposed by
Breiman.
8. Munehiro Nakamura, Yusuke Kajiwara, Tatsuhito Hasegawa & Haruhiko Kimura
International Journal of Image Processing (IJIP), Volume (7) : Issue (4) : 2013 425
FIGURE 7: Selection of consecutive k frames whose prediction probabilities are similar to those of ft (k=5).
3.5 Discrimination of Sequential Handling Patterns for a Flexible Cystoscope
In flexible cystoscopy, the same handling pattern would last for several frames. Hence, in a case
that the discriminate handling for a frame ft is different from the one for ft-1 and ft+1, ft is expected
to be another handling pattern. Considering the case, we propose to correct the handling pattern
for ft. Figure 7 shows a process to discriminate sequential k frames, where k = 5 is configured in
this case. The sequential k frames are determined by the following steps.
(1) Select the sequential k frames from ft-k+α to ft-1+α (Initially, α = 0).
(2) Calculate a similarity of the class prediction probabilities for each frame.
(3) α = α + 1 if α < k and go back to step (1).
(4) Determine the k frames when the similarity in step (2) is maximum.
And then, the similarity Simt is calculated by the following formula
27
1
[ ] | [ ] [ ]| (5)t t t
j
Sim aprob j prob jα
=
= −∑
where probt[j] is the class prediction probability of ft for jth handling pattern, aprobt[j] is the
average class prediction probability of selected k frames for jth handling pattern. Thus, the
proposed method selects k frames whose average class prediction probability are similar to that
of ft. And then, the discriminated handling pattern for ft is replaced by the jth handling pattern
where aprobt[j] becomes maximal in k frames.
Next, regarding fnum as the number of replaced handling patterns (fnum=1,2,3...,k-1), fnum is
increased in 1 by 1 (fnum is 1 in the initial step above). Then, regarding k as k', k' is set as k-fnum+1.
Note that the replacement of k' frames is conducted when the following term is fulfilled.
(6)B AP > P
where PB is the average class prediction probability for a handling pattern B in k' frames and PA is
the average class prediction probability for a handling pattern A in k' frames. These terms can
prevent a case that the handling pattern of the correct sequential k' frames is replaced to
incorrect one. The method above can be applied to frames before ft when the discrimination of
the handling pattern for t+k frames is finished.
4. PROPOSED SYSTEM
4.1 Experimental Environment
We applied the proposed system to ten videos of flexible cystoscopy at the Kanazawa University
Hospital in Japan. All the examinations were conducted by an expect physician. The flexible
cystoscope used in the examinations is OLYMPUS CYF TYPE VA2. Regarding the videos, the
format is AVI (24-bit color), the frame rate is 29.97 fps, and the average length is 115 seconds.
Each of the videos was cut manually as they start from the scene that the image of the bladder
wall appears at the first time to the scene that the image of the urethra appears at the first time
after the start scene. T evaluate the performance of the proposed system, by observing the
9. Munehiro Nakamura, Yusuke Kajiwara, Tatsuhito Hasegawa & Haruhiko Kimura
International Journal of Image Processing (IJIP), Volume (7) : Issue (4) : 2013 426
videos, the expert physician judged the handling patterns frame by frame. Then, a RF (Random
Forest) classifier discriminated the handling patterns. The accuracy of estimating the handling
patterns is evaluated by 10-fold cross validation[9] as below.
(1) Choose the first 1000 frames in each of the ten videos.
(2) Extract ZNCC-based optical flows and Rott, Bendt, and Inst (refer to Sec. 3.4) from each
of the 1000 frames.
(3) Choose the 1000 frames in one of the videos as test data and the other 9000 frames as
training data.
(4) RF learns the training data.
(5) RF discriminates the handling patterns frame by frame in the test data.
(6) Sequential handling patterns of the cystoscope are discriminated by the time series
analysis described in Sec. 3.5.
(7) Repeat the procedure from (3) until all the 1000 frames are selected as test data.
4.2 Results
First, we discriminated the handling patterns of the flexible cystoscope in each of the 1000 frames.
Table 3 shows the average correct rate in the discrimination. In this experiment, parameters for
the Random Forest classifier were optimized, namely the number of the trees was configured as
525. As the table shows, the proposed method outperforms base line.
Next, we reproduced each of the cystoscopic examinations in the virtual bladder defined in Sec.
3.1. In each examination, observed regions were painted on the virtual bladder. To evaluate the
performance of reproducing the observation, we assume that the physician could not observe one
of the whole regions. Such a region is called as target region in the rest of this paper. For
example, suppose that the physician could not observe trigone, the region of trigone in the virtual
Handling pattern Number of frames Base line proposed method
neutral 877 70.6% 93.2%
left 1049 81.0% 94.5%
right 932 82.1% 93.4%
up 741 83.3% 95.0%
down 1214 82.0% 96.2%
push 801 76.0% 94.6%
right 932 82.1% 93.4%
TABLE 3: Average correct rate for each of the handling patterns. base line is the case that only
ZNCC-based optical flows were learned by the Random Forest classifier.
No. 1 2 3 4 5
Cr 54.9% 81.9% 91.6% 91.2% 62.5%
Fr 6.4% 2.6% 1.2% 1.3% 5.4%
No. 6 7 8 9 10
Cr 83.5% 84.1% 82.9% 92.1% 90.4%
Fr 2.4% 2.3% 2.4% 1.1% 1.4%
TABLE 4: Parameter Cr and Fr obtained in the Experiment.
10. Munehiro Nakamura, Yusuke Kajiwara, Tatsuhito Hasegawa & Haruhiko Kimura
International Journal of Image Processing (IJIP), Volume (7) : Issue (4) : 2013 427
bladder is unpainted ideally and the other regions are painted. Here, we define Cr as the
percentage of the unpainted area in the target region and Fr as the percentage of the painted
area in the other region. Therefore, the ideal Cr is 100% and the ideal Fr is 0%. Table 4 shows
the average Cr and Fr for each of the video. From the table, we can find that Cr has been more
than 80% except the video No.1 and No.5.
4.3 Failure Cases
In Table 4, both of Cr and Fr for the video No.1 has been the worst among all the videos.
Observing the video, we could find that bladder stones were floating around trigone. Figure 8
shows the stones marked in circle. In 114 frames, although the camera was being stopped, the
block-matching method detected movement of the stones and the proposed system incorrectly
judged the handling patterns in the 114 frames as left or insertion. The bladder stones are
sometimes observed in cystoscopy. Therefore, we need to detect the white stones as noise. In
Figure 8, average density of the stones in areas of each circle has been 223.4 (standard
deviation is 15.9) while average density of all the 64072 frames in ten videos is 124.9. The
densities above were
FIGURE 8: Bladder Stones Appeared Around Trigone in the Video No.1.
FIGURE 9: One of the Images where Halation Observed in the Video No.5.
measured from the original images obtained from the cystoscope. Hence, areas of the bladder
stones in each image would be extracted using the density distributions of the whole image.
11. Munehiro Nakamura, Yusuke Kajiwara, Tatsuhito Hasegawa & Haruhiko Kimura
International Journal of Image Processing (IJIP), Volume (7) : Issue (4) : 2013 428
In Table 4, Cr and Fr for the video No.5 has been the second worst among all the videos.
Observing the video No.5, we could find that halation lasted for 125 consecutive frames. Figure 9
shows an image of the halation. Since the halation covered overall area in each image, the block-
matching method could not extract movement of the flexible cystoscope. Hence, in the case when
handling patterns are judged incorrectly due to the halation, it is necessary to discriminate the
handling patterns from handling patterns in other frames.
4.4 Future Works
From the experimental results shown in Sec. 4.2, this paper has indicated that our proposed
system can be used as a trainer for beginner physicians. Although cystoscopic images are
significantly unclear compared with images used in related works [3, 4, 10], it is seen that the
proposed system works well on tracking the observation in a flexible cystoscopy. However,
sometimes the tracking would fail due to the failure cases described in Sec. 4.3. In such case, it is
necessary to estimate the actual position according to landmarks placed in the virtual bladder.
One of the ways for generating landmarks is the use of HOG (Histogram of Oriented Gradients)
which is robust against rotation and size difference. In addition, we would like to correct turbulent
flows obtained from noisy images.
5. CONCLUSION
This paper has presented a system of tracking the observation in flexible cystoscopy. The
proposed system discriminates three handling patterns of flexible cystoscope. To achieve this
objective accurately, we proposed to extract ZNCC-based optical flows and three features that nd
the represent the degree of the handling patterns from cystoscopic images. In addition, we
proposed to discriminate sequential handling patterns of the flexible cystoscope. Experimental
results using ten videos have shown the average correct ratio of the three handling patterns has
been at least 90%. We also reproduced the observation in a flexible cystoscopy in a virtual 3D
bladder we constructed.
Considering the failure cases for tracking the observation, we need to estimate the actual position
in case of the tracking failure and correct turbulent flows obtained from cystoscopic images where
bladder stones are floating and halation is happened. Besides, we would like to estimate the
shape and size of the bladder using CT or MRI images.
6. REFERENCES
[1] M. Froehner, M. A. Brausi, H. W. Herr, G. Muto, U E. Studer. “Complications following radical
cystectomy for bladder cancer in the elderly” European Urology, vol.56, no.3, pp.443-454,
2009.
[2] J. Key, D. Dhawan, D. K. Knapp, K. Kim, I. C. Kwon, K. Choi, J. F. Leary. “Method and
apparatus for estimating the velocity vector of multiple vehicles on non-level and curved roads
using a single camera” in Proc. SPIE 8225, Imaging, Manipulation, and Analysis of
Biomolecules, Cells, and Tissues X, 82251F, 2012, pp.1-8.
[3] H C. Choi, S. Y Oh. “Robust segment-based object tracking using generalized hyperplane
approximation” Pattern Recognition, vol.45, no.8, pp.2980-2991, 2012.
[4] A. Ramisa, G. Alenya, C. Torras. “Single image 3D human pose estimation from noisy
observations” in Proc. 2012 IEEE Conference on Computer Vision and Pattern Recognition,
2012, pp.2673-2680.
[5] J. C. Yoo and T. Han. “Fast normalized cross-correlation” Circ. Syst. Signal Process, vol.28,
no.6, pp.819-843, 2009.
[6] L. Breiman. “Random Forests” Machine Learning, vol.45, no.1, pp.5-32, 2001.
12. Munehiro Nakamura, Yusuke Kajiwara, Tatsuhito Hasegawa & Haruhiko Kimura
International Journal of Image Processing (IJIP), Volume (7) : Issue (4) : 2013 429
[7] S.L. Tanimoto. “Template Matching in Pyramids” Computer Graphics and Image Processing,
vol.16, no.4, pp.356-369, 1981.
[8] B.K.P Horn and B.G. Schunck. “Determining optical flow” Artif. Intell., vol.17, pp.185-203,
1981.
[9] M. Fosteller. “A k-sample slippage test for an extreme population” Annals of Mathematical
Statistics, vol.19, no.1, pp.58-65, 1948.
[10] L. Beyang, S. Gould, and D. Koller. “Single image depth estimation from predicted semantic
labels” in Proc. 2010 IEEE Conference on Computer Vision and Pattern Recognition, 2010,
pp.1253-1260.
[11] N. Dalal and B. Triggs. “Histograms of Oriented Gradients for Human Detection” in Proc.2005
IEEE Conference on Computer Vision and Pattern Recognition, 2005, pp.886-893.