Pub Date : 2010-04-14DOI: 10.1109/ISBI.2010.5490399
Zhaozheng Yin, Ryoma Bise, Mei Chen, T. Kanade
Cell segmentation in microscopy imagery is essential for many bioimage applications such as cell tracking. To segment cells from the background accurately, we present a pixel classification approach that is independent of cell type or imaging modality. We train a set of Bayesian classifiers from clustered local training image patches. Each Bayesian classifier is an expert to make decision in its specific domain. The decision from the mixture of experts determines how likely a new pixel is a cell pixel. We demonstrate the effectiveness of this approach on four cell types with diverse morphologies under different microscopy imaging modalities.
{"title":"Cell segmentation in microscopy imagery using a bag of local Bayesian classifiers","authors":"Zhaozheng Yin, Ryoma Bise, Mei Chen, T. Kanade","doi":"10.1109/ISBI.2010.5490399","DOIUrl":"https://doi.org/10.1109/ISBI.2010.5490399","url":null,"abstract":"Cell segmentation in microscopy imagery is essential for many bioimage applications such as cell tracking. To segment cells from the background accurately, we present a pixel classification approach that is independent of cell type or imaging modality. We train a set of Bayesian classifiers from clustered local training image patches. Each Bayesian classifier is an expert to make decision in its specific domain. The decision from the mixture of experts determines how likely a new pixel is a cell pixel. We demonstrate the effectiveness of this approach on four cell types with diverse morphologies under different microscopy imaging modalities.","PeriodicalId":250523,"journal":{"name":"2010 IEEE International Symposium on Biomedical Imaging: From Nano to Macro","volume":"37 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":"127748850","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.2010.5490351
A. Gasnier, R. Ardon, C. Ciofolo-Veit, E. Leen, J. Correas
3D contrast-enhanced ultrasound (CEUS) is a powerful imaging technique for tumour vascularity assessment, which is critical for radio-frequency ablation (RFA) planning or for the assessment of response to antiangiogenic therapies. In this paper, we propose a novel semi-automated method for the quantification of tumour vascularity in 3D CEUS data. We apply a two-step framework combining an interactive segmentation of the tumour necrosis followed by an automatic detection of the vascularity based on implicit representations. Experimental results on 3D CEUS images of renal cell carcinomas (RCC) show that our method is promising in terms of speed and quality.
{"title":"Assessing tumour vascularity with 3D contrast-enhanced ultrasound: A new semi-automated segmentation framework","authors":"A. Gasnier, R. Ardon, C. Ciofolo-Veit, E. Leen, J. Correas","doi":"10.1109/ISBI.2010.5490351","DOIUrl":"https://doi.org/10.1109/ISBI.2010.5490351","url":null,"abstract":"3D contrast-enhanced ultrasound (CEUS) is a powerful imaging technique for tumour vascularity assessment, which is critical for radio-frequency ablation (RFA) planning or for the assessment of response to antiangiogenic therapies. In this paper, we propose a novel semi-automated method for the quantification of tumour vascularity in 3D CEUS data. We apply a two-step framework combining an interactive segmentation of the tumour necrosis followed by an automatic detection of the vascularity based on implicit representations. Experimental results on 3D CEUS images of renal cell carcinomas (RCC) show that our method is promising in terms of speed and quality.","PeriodicalId":250523,"journal":{"name":"2010 IEEE International Symposium on Biomedical Imaging: From Nano to Macro","volume":"114 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":"132751118","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.2010.5490141
D. Sussman, Jianhua Yao, R. Summers
Obesity has become widespread in America and has been identified as a risk factor for many illnesses. Measuring adipose tissue (AT) with traditional means is often unreliable and inaccurate. MRI provides a safe and minimally invasive means to measure AT accurately and segment visceral AT from subcutaneous AT. However, MRI is often corrupted by image artifacts which make manual measurements difficult and time consuming. We present a fully automated method to measure and segment abdominal AT in MRI. Our method uses non-parametric non-uniform intensity normalization (N3) to correct for image artifacts and inhomogeneities, fuzzy c-means to cluster AT regions and active contour models to separate subcutaneous and visceral AT. Our method was able to measure images with severe intensity inhomogeneities and demonstrated agreement with two manual users that was close to the agreement between the manual users.
{"title":"Automated measurement and segmentation of abdominal adipose tissue in MRI","authors":"D. Sussman, Jianhua Yao, R. Summers","doi":"10.1109/ISBI.2010.5490141","DOIUrl":"https://doi.org/10.1109/ISBI.2010.5490141","url":null,"abstract":"Obesity has become widespread in America and has been identified as a risk factor for many illnesses. Measuring adipose tissue (AT) with traditional means is often unreliable and inaccurate. MRI provides a safe and minimally invasive means to measure AT accurately and segment visceral AT from subcutaneous AT. However, MRI is often corrupted by image artifacts which make manual measurements difficult and time consuming. We present a fully automated method to measure and segment abdominal AT in MRI. Our method uses non-parametric non-uniform intensity normalization (N3) to correct for image artifacts and inhomogeneities, fuzzy c-means to cluster AT regions and active contour models to separate subcutaneous and visceral AT. Our method was able to measure images with severe intensity inhomogeneities and demonstrated agreement with two manual users that was close to the agreement between the manual users.","PeriodicalId":250523,"journal":{"name":"2010 IEEE International Symposium on Biomedical Imaging: From Nano to Macro","volume":"13 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":"133278747","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.2010.5490300
C. Constantinides, N. Aristokleous, G. Johnson, Dimitris Perperides
Magnetic Resonance Imaging (MRI) has exhibited significant potential for quantifying cardiac function and dysfunction in the mouse. Recent advances in high-resolution cardiac MR imaging techniques have contributed to the development of acquisition approaches that allow fast and accurate description of anatomic structures, and accurate surface and finite element (FE) mesh model constructions for study of global mechanical function in normal and transgenic mice. This study presents work in progress for construction of quantitatively accurate three-dimensional (3D) and 4D dynamic surface and FE models of murine left ventricular (LV) muscle in C57BL/6J (n=10) mice. Constructed models are subsequently imported into commercial software packages for the solution of the constitutive equations that characterize mechanical function, including computation of the stress and strain fields. They are further used with solid-free form fabrication processes to construct model-based material renditions of the human and mouse hearts.
{"title":"Static and dynamic cardiac modelling: Initial strides and results towards a quantitatively accurate mechanical heart model","authors":"C. Constantinides, N. Aristokleous, G. Johnson, Dimitris Perperides","doi":"10.1109/ISBI.2010.5490300","DOIUrl":"https://doi.org/10.1109/ISBI.2010.5490300","url":null,"abstract":"Magnetic Resonance Imaging (MRI) has exhibited significant potential for quantifying cardiac function and dysfunction in the mouse. Recent advances in high-resolution cardiac MR imaging techniques have contributed to the development of acquisition approaches that allow fast and accurate description of anatomic structures, and accurate surface and finite element (FE) mesh model constructions for study of global mechanical function in normal and transgenic mice. This study presents work in progress for construction of quantitatively accurate three-dimensional (3D) and 4D dynamic surface and FE models of murine left ventricular (LV) muscle in C57BL/6J (n=10) mice. Constructed models are subsequently imported into commercial software packages for the solution of the constitutive equations that characterize mechanical function, including computation of the stress and strain fields. They are further used with solid-free form fabrication processes to construct model-based material renditions of the human and mouse hearts.","PeriodicalId":250523,"journal":{"name":"2010 IEEE International Symposium on Biomedical Imaging: From Nano to Macro","volume":"16 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":"131929169","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.2010.5490188
F. Deligianni, E. Robinson, C. Beckmann, D. Sharp, A. Edwards, D. Rueckert
Studies that examine the relationship of functional and structural connectivity are tremendously important in interpreting neurophysiological data. Although, the relationship between functional and structural connectivity has been explored with a number of statistical tools [1, 2], there is no explicit attempt to quantitatively measure how well functional data can be predicted from structural data. Here, we predict functional connectivity from structural connectivity, explicitly, by utilizing a predictive model based on PCA and CCA. The combination of these techniques allowed the reduction of dimensionality and modeling of inter-correlations, successfully. We provide both qualitative and quantitative results based on a leave-one-out validation.
{"title":"Inference of functional connectivity from structural brain connectivity","authors":"F. Deligianni, E. Robinson, C. Beckmann, D. Sharp, A. Edwards, D. Rueckert","doi":"10.1109/ISBI.2010.5490188","DOIUrl":"https://doi.org/10.1109/ISBI.2010.5490188","url":null,"abstract":"Studies that examine the relationship of functional and structural connectivity are tremendously important in interpreting neurophysiological data. Although, the relationship between functional and structural connectivity has been explored with a number of statistical tools [1, 2], there is no explicit attempt to quantitatively measure how well functional data can be predicted from structural data. Here, we predict functional connectivity from structural connectivity, explicitly, by utilizing a predictive model based on PCA and CCA. The combination of these techniques allowed the reduction of dimensionality and modeling of inter-correlations, successfully. We provide both qualitative and quantitative results based on a leave-one-out validation.","PeriodicalId":250523,"journal":{"name":"2010 IEEE International Symposium on Biomedical Imaging: From Nano to Macro","volume":"8 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":"134100696","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.2010.5490261
P. Coupé, P. Hellier, X. Morandi, C. Barillot
In this paper, a global approach based on 3D freehand ultrasound imaging is proposed to (a) correct the error of the neuronavigation system in image-patient registration and (b) compensate for the deformations of the cerebral structures occurring during a neurosurgical procedure. The rigid and non rigid multimodal registrations are achieved by matching the hyperechogenic structures of brain. The quantitative evaluation of the non rigid registration was performed within a framework based on synthetic deformation. Finally, experiments were carried out on real data sets of 4 patients with lesions such as cavernoma and low-grade glioma. Qualitative and quantitative results on the estimated error performed by neuronavigation system and the estimated brain deformations are given.
{"title":"Intraoperative ultrasonography for the correction of brainshift based on the matching of hyperechogenic structures","authors":"P. Coupé, P. Hellier, X. Morandi, C. Barillot","doi":"10.1109/ISBI.2010.5490261","DOIUrl":"https://doi.org/10.1109/ISBI.2010.5490261","url":null,"abstract":"In this paper, a global approach based on 3D freehand ultrasound imaging is proposed to (a) correct the error of the neuronavigation system in image-patient registration and (b) compensate for the deformations of the cerebral structures occurring during a neurosurgical procedure. The rigid and non rigid multimodal registrations are achieved by matching the hyperechogenic structures of brain. The quantitative evaluation of the non rigid registration was performed within a framework based on synthetic deformation. Finally, experiments were carried out on real data sets of 4 patients with lesions such as cavernoma and low-grade glioma. Qualitative and quantitative results on the estimated error performed by neuronavigation system and the estimated brain deformations are given.","PeriodicalId":250523,"journal":{"name":"2010 IEEE International Symposium on Biomedical Imaging: From Nano to Macro","volume":"1 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":"114538345","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.2010.5490264
A. Madabhushi, A. Basavanhally, Scott Doyle, S. Agner, George Lee
Computer-aided prognosis (CAP) is a new and exciting complement to the field of computer-aided diagnosis (CAD) and involves developing computerized image analysis and multi-modal data fusion algorithms for helping physicians predict disease outcome and patient survival. At the Laboratory for Computational Imaging and Bioinformatics (LCIB)1 at Rutgers University we have been developing computerized algorithms for high dimensional data and image analysis for predicting disease outcome from multiple modalities includng MRI, digital pathology, and protein expression. Additionally, we have been developing novel data fusion algorithms based on nonlinear dimensionality reduction methods (such as Graph Embedding) to quantitatively integrate prognostic information from multiple data sources and modalities. In this paper, we briefly describe 5 representative and ongoing CAP projects at LCIB. These projects include (1) an Image-based Risk Score (IbRiS) algorithm for predicting outcome of ER+ breast cancer patients based on quantitative image analysis of digitized breast cancer biopsy specimens alone, (2) segmenting and determining extent of lymphocytic infiltration (identified as a possible prognostic marker for outcome in Her2+ breast cancers) from digitized histopathology, (3) segmenting and diagnosing highly agressive triple-negative breast cancers on dynamic contrast enhanced (DCE) MRI, (4) distinguishing patients with different Gleason grades of prostate cancer (grade being known to be correlated to outcome) from digitzed needle biopsy specimens, and (5) integrating protein expression measurements obtained from mass spectrometry with quantitative image features derived from digitized histopathology for distinguishing between prostate cancer patients at low and high risk of disease recurrence.
{"title":"Computer-aided prognosis: Predicting patient and disease outcome via multi-modal image analysis","authors":"A. Madabhushi, A. Basavanhally, Scott Doyle, S. Agner, George Lee","doi":"10.1109/ISBI.2010.5490264","DOIUrl":"https://doi.org/10.1109/ISBI.2010.5490264","url":null,"abstract":"Computer-aided prognosis (CAP) is a new and exciting complement to the field of computer-aided diagnosis (CAD) and involves developing computerized image analysis and multi-modal data fusion algorithms for helping physicians predict disease outcome and patient survival. At the Laboratory for Computational Imaging and Bioinformatics (LCIB)1 at Rutgers University we have been developing computerized algorithms for high dimensional data and image analysis for predicting disease outcome from multiple modalities includng MRI, digital pathology, and protein expression. Additionally, we have been developing novel data fusion algorithms based on nonlinear dimensionality reduction methods (such as Graph Embedding) to quantitatively integrate prognostic information from multiple data sources and modalities. In this paper, we briefly describe 5 representative and ongoing CAP projects at LCIB. These projects include (1) an Image-based Risk Score (IbRiS) algorithm for predicting outcome of ER+ breast cancer patients based on quantitative image analysis of digitized breast cancer biopsy specimens alone, (2) segmenting and determining extent of lymphocytic infiltration (identified as a possible prognostic marker for outcome in Her2+ breast cancers) from digitized histopathology, (3) segmenting and diagnosing highly agressive triple-negative breast cancers on dynamic contrast enhanced (DCE) MRI, (4) distinguishing patients with different Gleason grades of prostate cancer (grade being known to be correlated to outcome) from digitzed needle biopsy specimens, and (5) integrating protein expression measurements obtained from mass spectrometry with quantitative image features derived from digitized histopathology for distinguishing between prostate cancer patients at low and high risk of disease recurrence.","PeriodicalId":250523,"journal":{"name":"2010 IEEE International Symposium on Biomedical Imaging: From Nano to Macro","volume":"112 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":"124841417","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.2010.5490114
Damon E. Hyde, S. Warfield
We present a new spatio-temporal regularization approach for EEG source localization. Using separable spatial and temporal smoothing constraints, we are able to construct a computationally feasible maximum a posteriori (MAP) solution. The smoothing is achieved using a Helmholtz-type functional which allows explicit control over the distance at which correlation between voxels is present. Temporal variation in signal to noise ratio is incorporated as a column-wise of the temporal regularization matrix. Using both simulated and experimental EEG data, we show that this approach allows for improvements in both the spatial and temporal accuracy of the resulting solutions.
{"title":"Controllable spatio-temporal smoothness constraints for EEG source localization","authors":"Damon E. Hyde, S. Warfield","doi":"10.1109/ISBI.2010.5490114","DOIUrl":"https://doi.org/10.1109/ISBI.2010.5490114","url":null,"abstract":"We present a new spatio-temporal regularization approach for EEG source localization. Using separable spatial and temporal smoothing constraints, we are able to construct a computationally feasible maximum a posteriori (MAP) solution. The smoothing is achieved using a Helmholtz-type functional which allows explicit control over the distance at which correlation between voxels is present. Temporal variation in signal to noise ratio is incorporated as a column-wise of the temporal regularization matrix. Using both simulated and experimental EEG data, we show that this approach allows for improvements in both the spatial and temporal accuracy of the resulting solutions.","PeriodicalId":250523,"journal":{"name":"2010 IEEE International Symposium on Biomedical Imaging: From Nano to Macro","volume":"296 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":"134455479","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.2010.5490094
Liu Lukai, M. Haider, D. Langer, I. Yetik
Imaging methods to localize prostate cancer with sufficient accuracy are extremely useful in guiding biopsy, radiotherapy and surgery as well as to monitor disease progression. Imaging prostate cancer with multispectral magnetic resonance imaging (MRI) has shown a superior performance when compared to classical imaging modality transrectal ultrasound (TRUS). An important component of multispectral MRI is dynamic contrast-enhanced magnetic resonance imaging (DCE MRI). However, parametric images based on DCE MRI suffer from low signal-to-noise ratio (SNR). In this study, we propose a kinetic parametric imaging method with DCE MRI to overcome this problem using spatial regularization for improved prostate cancer localization. We demonstrate that the proposed method outperforms pixel-wise parametric imaging method, and that the performance of resulting tumor localization has a considerable improvement. Both visual and quantitative evaluations based on a task-based approach focused on tumor localization are provided.
{"title":"Improved prostate cancer localization with spatially regularized dynamic contrast-enhanced magnetic resonance imaging","authors":"Liu Lukai, M. Haider, D. Langer, I. Yetik","doi":"10.1109/ISBI.2010.5490094","DOIUrl":"https://doi.org/10.1109/ISBI.2010.5490094","url":null,"abstract":"Imaging methods to localize prostate cancer with sufficient accuracy are extremely useful in guiding biopsy, radiotherapy and surgery as well as to monitor disease progression. Imaging prostate cancer with multispectral magnetic resonance imaging (MRI) has shown a superior performance when compared to classical imaging modality transrectal ultrasound (TRUS). An important component of multispectral MRI is dynamic contrast-enhanced magnetic resonance imaging (DCE MRI). However, parametric images based on DCE MRI suffer from low signal-to-noise ratio (SNR). In this study, we propose a kinetic parametric imaging method with DCE MRI to overcome this problem using spatial regularization for improved prostate cancer localization. We demonstrate that the proposed method outperforms pixel-wise parametric imaging method, and that the performance of resulting tumor localization has a considerable improvement. Both visual and quantitative evaluations based on a task-based approach focused on tumor localization are provided.","PeriodicalId":250523,"journal":{"name":"2010 IEEE International Symposium on Biomedical Imaging: From Nano to Macro","volume":"2 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":"131995482","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.2010.5490107
Xuqing Wu, S. Shah
Cell segmentation is a challenging problem in histology and cytology that can benefit from additional information obtained in using multispectral imaging. Unique transmission spectra of biological tissues are potentially useful for better classification and segmentation of sub-cellular structures. In this paper, we propose a conditional random field (CRF) model that interprets high-dimensional spectral data during inference and pixel labeling. High quality segmentations are computed by combining low-level cues and high-level contextual information extracted by unsupervised topic discovery. Comparative analysis of the proposed model against the commonly used 2-D CRF model in color space is also performed. Results of this evaluation show the benefits of our proposed model.
{"title":"A bottom-up and top-down model for cell segmentation using multispectral data","authors":"Xuqing Wu, S. Shah","doi":"10.1109/ISBI.2010.5490107","DOIUrl":"https://doi.org/10.1109/ISBI.2010.5490107","url":null,"abstract":"Cell segmentation is a challenging problem in histology and cytology that can benefit from additional information obtained in using multispectral imaging. Unique transmission spectra of biological tissues are potentially useful for better classification and segmentation of sub-cellular structures. In this paper, we propose a conditional random field (CRF) model that interprets high-dimensional spectral data during inference and pixel labeling. High quality segmentations are computed by combining low-level cues and high-level contextual information extracted by unsupervised topic discovery. Comparative analysis of the proposed model against the commonly used 2-D CRF model in color space is also performed. Results of this evaluation show the benefits of our proposed model.","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":"133887507","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：