Pub Date : 2010-04-14DOI: 10.1109/ISBI.2010.5490375
T. Nuzhnaya, M. Barnathan, Haibin Ling, V. Megalooikonomou, P. Bakic, Andrew D. A. Maidment
Probabilistic branching node inference is an important step for analyzing branching patterns involved in many anatomic structures. Based on an approach we have developed previously, we investigate combining machine learning techniques and hybrid image statistics for probabilistic branching node inference, using adaptive boosting as a probabilistic inference framework. Then, we use local image statistics at different image scales for feature representation, including the Harris cornerness, Laplacian, eigenvalues of the Hessian, and Harralick texture features. The proposed approach is applied to a breast imaging dataset consisting of 30 images, 7 of which were previously reported. The use of boosting and the Harralick texture feature further improves upon our previous results, highlighting the role of texture in the analysis of the breast ducts and other branching structures.
{"title":"Probabilistic branching node detection using AdaBoost and hybrid local features","authors":"T. Nuzhnaya, M. Barnathan, Haibin Ling, V. Megalooikonomou, P. Bakic, Andrew D. A. Maidment","doi":"10.1109/ISBI.2010.5490375","DOIUrl":"https://doi.org/10.1109/ISBI.2010.5490375","url":null,"abstract":"Probabilistic branching node inference is an important step for analyzing branching patterns involved in many anatomic structures. Based on an approach we have developed previously, we investigate combining machine learning techniques and hybrid image statistics for probabilistic branching node inference, using adaptive boosting as a probabilistic inference framework. Then, we use local image statistics at different image scales for feature representation, including the Harris cornerness, Laplacian, eigenvalues of the Hessian, and Harralick texture features. The proposed approach is applied to a breast imaging dataset consisting of 30 images, 7 of which were previously reported. The use of boosting and the Harralick texture feature further improves upon our previous results, highlighting the role of texture in the analysis of the breast ducts and other branching structures.","PeriodicalId":250523,"journal":{"name":"2010 IEEE International Symposium on Biomedical Imaging: From Nano to Macro","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114064493","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2010-04-14DOI: 10.1109/ISBI.2010.5490067
G. Tzagkarakis, P. Tsakalides
The increased high-resolution capabilities of modern medical image acquisition systems raise the crucial tasks of effectively storing and interacting with large databases of such data. The ease of image storage and query would be unfeasible without compression, which represents high-resolution images with a relatively small set of significant transform coefficients. Due to the specific content of medical images, compression often results in highly sparse representations in appropriate orthonormal bases. The inherent property of compressed sensing (CS) working simultaneously as a sensing and compression protocol using a small subset of random projection coefficients, enables a potentially significant reduction in storage requirements. In this paper, we introduce a Bayesian CS approach for obtaining highly sparse representations of medical images based on a set of noisy CS measurements, where the prior belief that the vector of transform coefficients should be sparse is exploited by modeling its probability distribution by means of a Gaussian Scale Mixture. The experimental results show that the proposed approach maintains the reconstruction performance of other state-of-the-art CS methods while achieving significantly sparser representations of medical images with distinct content.
{"title":"Sparse representation of medical images via compressed sensing using Gaussian Scale Mixtures","authors":"G. Tzagkarakis, P. Tsakalides","doi":"10.1109/ISBI.2010.5490067","DOIUrl":"https://doi.org/10.1109/ISBI.2010.5490067","url":null,"abstract":"The increased high-resolution capabilities of modern medical image acquisition systems raise the crucial tasks of effectively storing and interacting with large databases of such data. The ease of image storage and query would be unfeasible without compression, which represents high-resolution images with a relatively small set of significant transform coefficients. Due to the specific content of medical images, compression often results in highly sparse representations in appropriate orthonormal bases. The inherent property of compressed sensing (CS) working simultaneously as a sensing and compression protocol using a small subset of random projection coefficients, enables a potentially significant reduction in storage requirements. In this paper, we introduce a Bayesian CS approach for obtaining highly sparse representations of medical images based on a set of noisy CS measurements, where the prior belief that the vector of transform coefficients should be sparse is exploited by modeling its probability distribution by means of a Gaussian Scale Mixture. The experimental results show that the proposed approach maintains the reconstruction performance of other state-of-the-art CS methods while achieving significantly sparser representations of medical images with distinct content.","PeriodicalId":250523,"journal":{"name":"2010 IEEE International Symposium on Biomedical Imaging: From Nano to Macro","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114137582","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2010-04-14DOI: 10.1109/ISBI.2010.5490249
Javed M. Aman, Jianhua Yao, R. Summers
We present a content-based image retrieval (CBIR) paradigm to enhance computed tomographic colonography computer-aided detection (CTCCAD). Our method uses scale-invariant feature transform (SIFT) features in conjunction with the bag-of-words model to describe and differentiate 3D images of CTCCAD detections. We evaluate the performance of our system using both digital colon phantoms and detections form CTCCAD. Our method shows promise in distinguishing common structures found within the colon.
{"title":"Content-based image retrieval on CT colonography using rotation and scale invariant features and bag-of-words model","authors":"Javed M. Aman, Jianhua Yao, R. Summers","doi":"10.1109/ISBI.2010.5490249","DOIUrl":"https://doi.org/10.1109/ISBI.2010.5490249","url":null,"abstract":"We present a content-based image retrieval (CBIR) paradigm to enhance computed tomographic colonography computer-aided detection (CTCCAD). Our method uses scale-invariant feature transform (SIFT) features in conjunction with the bag-of-words model to describe and differentiate 3D images of CTCCAD detections. We evaluate the performance of our system using both digital colon phantoms and detections form CTCCAD. Our method shows promise in distinguishing common structures found within the colon.","PeriodicalId":250523,"journal":{"name":"2010 IEEE International Symposium on Biomedical Imaging: From Nano to Macro","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120951292","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2010-04-14DOI: 10.1109/ISBI.2010.5490355
J. Seabra, J. Sanches
The information encoded in ultrasound speckle is often discarded but it is widely recognized that this phenomenon is dependent of the intrinsic acoustic properties of tissues. In this paper we propose a robust method to estimate the de-speckled and speckle components from the ultrasound data with the purpose of tissue characterization. A de-speckling method, which can conveniently work with either Radio Frequency (RF) or B-mode data, contributes to an improvement on the visualization of anatomical details, while providing useful fields from where echogenicity and texture features can be extracted. The adequacy of the RF image retrieval and despeckling methods are tackled using both synthetic and real ultrasonic data.
{"title":"On estimating de-speckled and speckle components from B-mode ultrasound images","authors":"J. Seabra, J. Sanches","doi":"10.1109/ISBI.2010.5490355","DOIUrl":"https://doi.org/10.1109/ISBI.2010.5490355","url":null,"abstract":"The information encoded in ultrasound speckle is often discarded but it is widely recognized that this phenomenon is dependent of the intrinsic acoustic properties of tissues. In this paper we propose a robust method to estimate the de-speckled and speckle components from the ultrasound data with the purpose of tissue characterization. A de-speckling method, which can conveniently work with either Radio Frequency (RF) or B-mode data, contributes to an improvement on the visualization of anatomical details, while providing useful fields from where echogenicity and texture features can be extracted. The adequacy of the RF image retrieval and despeckling methods are tackled using both synthetic and real ultrasonic data.","PeriodicalId":250523,"journal":{"name":"2010 IEEE International Symposium on Biomedical Imaging: From Nano to Macro","volume":"194 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116065324","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2010-04-14DOI: 10.1109/ISBI.2010.5490329
Duhgoon Lee, W. H. Nam, D. Hyun, Jae Young Lee, J. Ra
Synchronization between real-time ultrasound (US) and preoperative images can provide much information for US-guided intervention. For the synchronization, we present a real-time registration system between the two images of the liver without any help of sensors. In this system, we first generate a 4D preoperative image, which is composed of multiple 3D images along the respiration, by considering their local deformation. In the intraoperative stage, we achieve the pose information of a pose-fixed 3D US transducer by using several 3D US images. We then acquire 2D US images and find their corresponding images in real-time from the 4D preoperative image. The related registration is done by comparing a gradient-based similarity measure between a 2D US image and generated 2D preoperative image candidates. By the visual assessment of registration results, we confirm the feasibility of the proposed system for image-guidance.
{"title":"Sensorless and real-time registration between 2D ultrasound and preoperative images of the liver","authors":"Duhgoon Lee, W. H. Nam, D. Hyun, Jae Young Lee, J. Ra","doi":"10.1109/ISBI.2010.5490329","DOIUrl":"https://doi.org/10.1109/ISBI.2010.5490329","url":null,"abstract":"Synchronization between real-time ultrasound (US) and preoperative images can provide much information for US-guided intervention. For the synchronization, we present a real-time registration system between the two images of the liver without any help of sensors. In this system, we first generate a 4D preoperative image, which is composed of multiple 3D images along the respiration, by considering their local deformation. In the intraoperative stage, we achieve the pose information of a pose-fixed 3D US transducer by using several 3D US images. We then acquire 2D US images and find their corresponding images in real-time from the 4D preoperative image. The related registration is done by comparing a gradient-based similarity measure between a 2D US image and generated 2D preoperative image candidates. By the visual assessment of registration results, we confirm the feasibility of the proposed system for image-guidance.","PeriodicalId":250523,"journal":{"name":"2010 IEEE International Symposium on Biomedical Imaging: From Nano to Macro","volume":"4 8","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121011229","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2010-04-14DOI: 10.1109/ISBI.2010.5490328
N. Cahill, J. Noble, D. Hawkes
Any similarity measure used for image registration depends in some way on the region Ω describing the overlap between the floating and reference images. In variational registration, where the Gâteaux derivative of the similarity measure drives the registration, most literature implicitly assumes that Ω remains constant. This assumption is valid if homogeneous Dirichlet or sliding boundary conditions are chosen for the displacement field; however, it is invalid if any other type of boundary conditions are chosen, or if the similarity measure is computed over some masked portion of the overlap region. This article illustrates how these more general situations of different boundary conditions and/or masked regions can be accommodated in variational registration by explicitly accounting for the varying Ω in the Gâteaux derivative of the similarity measure.
{"title":"Accounting for changing overlap in variational image registration","authors":"N. Cahill, J. Noble, D. Hawkes","doi":"10.1109/ISBI.2010.5490328","DOIUrl":"https://doi.org/10.1109/ISBI.2010.5490328","url":null,"abstract":"Any similarity measure used for image registration depends in some way on the region Ω describing the overlap between the floating and reference images. In variational registration, where the Gâteaux derivative of the similarity measure drives the registration, most literature implicitly assumes that Ω remains constant. This assumption is valid if homogeneous Dirichlet or sliding boundary conditions are chosen for the displacement field; however, it is invalid if any other type of boundary conditions are chosen, or if the similarity measure is computed over some masked portion of the overlap region. This article illustrates how these more general situations of different boundary conditions and/or masked regions can be accommodated in variational registration by explicitly accounting for the varying Ω in the Gâteaux derivative of the similarity measure.","PeriodicalId":250523,"journal":{"name":"2010 IEEE International Symposium on Biomedical Imaging: From Nano to Macro","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116623003","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2010-04-14DOI: 10.1109/ISBI.2010.5490108
J. Fessler
Motion artifacts in image reconstruction problems can be reduced by performing image motion estimation and image reconstruction jointly using a penalized-likelihood cost function. However, updating the motion parameters by conventional gradient-based iterations can be computationally demanding due to the system model required in inverse problems. This paper describes an optimization transfer approach that leads to minimization steps for the motion parameters that have comparable complexity to those needed in image registration problems. This approach can simplify the implementation of motion-compensated image reconstruction (MCIR) methods when the motion parameters are estimated jointly with the reconstructed image.
{"title":"Optimization transfer approach to joint registration / reconstruction for motion-compensated image reconstruction","authors":"J. Fessler","doi":"10.1109/ISBI.2010.5490108","DOIUrl":"https://doi.org/10.1109/ISBI.2010.5490108","url":null,"abstract":"Motion artifacts in image reconstruction problems can be reduced by performing image motion estimation and image reconstruction jointly using a penalized-likelihood cost function. However, updating the motion parameters by conventional gradient-based iterations can be computationally demanding due to the system model required in inverse problems. This paper describes an optimization transfer approach that leads to minimization steps for the motion parameters that have comparable complexity to those needed in image registration problems. This approach can simplify the implementation of motion-compensated image reconstruction (MCIR) methods when the motion parameters are estimated jointly with the reconstructed image.","PeriodicalId":250523,"journal":{"name":"2010 IEEE International Symposium on Biomedical Imaging: From Nano to Macro","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124996006","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2010-04-14DOI: 10.1109/ISBI.2010.5490227
S. Brinkers, Heidelinde R. C. Dietrich, S. Stallinga, J. Mes, I. T. Young, B. Rieger
Current methods for tuberculosis nucleic acid detection require amplification and labeling before detection is possible. We propose here a method for direct detection using Tethered Particle Motion: gold nanoparticles are tethered to a glass substrate by single-stranded DNA molecules consisting of the complementary sequence to the target. Detection takes place by observing a change in the motion of the nanoparticles. The particles are imaged by a dark field microscope and captured on an EMCCD camera. Single particle tracking is carried out through maximum likelihood estimation of the Poisson noise limited Gaussian image profile using a parallelized algorithm on a GPU. The method is characterized by biophysical modeling and the ability to detect nucleic acids is shown. This single molecule method is suitable for multiplexing and could form the basis of an exquisitely sensitive method of detecting the presence of nucleic acids derived from human pathogens directly from patient material.
{"title":"Single molecule detection of tuberculosis nucleic acid using dark field Tethered Particle Motion","authors":"S. Brinkers, Heidelinde R. C. Dietrich, S. Stallinga, J. Mes, I. T. Young, B. Rieger","doi":"10.1109/ISBI.2010.5490227","DOIUrl":"https://doi.org/10.1109/ISBI.2010.5490227","url":null,"abstract":"Current methods for tuberculosis nucleic acid detection require amplification and labeling before detection is possible. We propose here a method for direct detection using Tethered Particle Motion: gold nanoparticles are tethered to a glass substrate by single-stranded DNA molecules consisting of the complementary sequence to the target. Detection takes place by observing a change in the motion of the nanoparticles. The particles are imaged by a dark field microscope and captured on an EMCCD camera. Single particle tracking is carried out through maximum likelihood estimation of the Poisson noise limited Gaussian image profile using a parallelized algorithm on a GPU. The method is characterized by biophysical modeling and the ability to detect nucleic acids is shown. This single molecule method is suitable for multiplexing and could form the basis of an exquisitely sensitive method of detecting the presence of nucleic acids derived from human pathogens directly from patient material.","PeriodicalId":250523,"journal":{"name":"2010 IEEE International Symposium on Biomedical Imaging: From Nano to Macro","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121390852","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2010-04-14DOI: 10.1109/ISBI.2010.5490176
R. Irwan
State of the art of CT technology will be presented. It has a z-coverage of 16cm to cover most of the organs in one single, non-helical, rotation. Whole body imaging using different scan methods will be presented, and a comparison in terms of scan time with helical scan mode will be discussed.
{"title":"Whole body imaging with dynamic volume 320-row CT","authors":"R. Irwan","doi":"10.1109/ISBI.2010.5490176","DOIUrl":"https://doi.org/10.1109/ISBI.2010.5490176","url":null,"abstract":"State of the art of CT technology will be presented. It has a z-coverage of 16cm to cover most of the organs in one single, non-helical, rotation. Whole body imaging using different scan methods will be presented, and a comparison in terms of scan time with helical scan mode will be discussed.","PeriodicalId":250523,"journal":{"name":"2010 IEEE International Symposium on Biomedical Imaging: From Nano to Macro","volume":"93 4","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131541039","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2010-04-14DOI: 10.1109/ISBI.2010.5490400
Juan L. P. Soto, D. Pantazis, K. Jerbi, Sylvain Bailler, R. Leahy
We present a multivariate method based on canonical correlation analysis for the study of functional connectivity in the brain with MEG data. We obtain a time-frequency representation of the brain activity on the cortical surface, and use the signal power at specific frequency bands as inputs to our model. Our measure of interaction between two spatial locations is the canonical correlation, and the vectors associated with it indicate the contribution of each individual frequency band to the interaction. The resulting canonical correlation maps are thresholded for significance using false discovery rate. We further provide a novel way to control for linear mixing by testing whether the correlation vectors are collinear. We apply our method to simulations and experimental data from an MEG visuomotor study, and demonstrate that it is able to detect functional interactions across space as well as the frequency bands that contribute to these interactions.
{"title":"Canonical correlation analysis applied to functional connectivity in MEG","authors":"Juan L. P. Soto, D. Pantazis, K. Jerbi, Sylvain Bailler, R. Leahy","doi":"10.1109/ISBI.2010.5490400","DOIUrl":"https://doi.org/10.1109/ISBI.2010.5490400","url":null,"abstract":"We present a multivariate method based on canonical correlation analysis for the study of functional connectivity in the brain with MEG data. We obtain a time-frequency representation of the brain activity on the cortical surface, and use the signal power at specific frequency bands as inputs to our model. Our measure of interaction between two spatial locations is the canonical correlation, and the vectors associated with it indicate the contribution of each individual frequency band to the interaction. The resulting canonical correlation maps are thresholded for significance using false discovery rate. We further provide a novel way to control for linear mixing by testing whether the correlation vectors are collinear. We apply our method to simulations and experimental data from an MEG visuomotor study, and demonstrate that it is able to detect functional interactions across space as well as the frequency bands that contribute to these interactions.","PeriodicalId":250523,"journal":{"name":"2010 IEEE International Symposium on Biomedical Imaging: From Nano to Macro","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121756274","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：