Ensemble based system for whole-slide prostate cancer probability mapping using color texture features.
dc.contributor.author | DiFranco, Matthew D | |
dc.contributor.author | O'Hurley, Gillian | |
dc.contributor.author | Kay, Elaine W | |
dc.contributor.author | Watson, R William G | |
dc.contributor.author | Cunningham, Padraig | |
dc.date.accessioned | 2012-06-15T15:34:04Z | |
dc.date.available | 2012-06-15T15:34:04Z | |
dc.date.issued | 2011 | |
dc.identifier.citation | Ensemble based system for whole-slide prostate cancer probability mapping using color texture features., 35 (7-8):629-45 Comput Med Imaging Graph | en_GB |
dc.identifier.issn | 1879-0771 | |
dc.identifier.pmid | 21269807 | |
dc.identifier.doi | 10.1016/j.compmedimag.2010.12.005 | |
dc.identifier.uri | http://hdl.handle.net/10147/229196 | |
dc.description.abstract | We present a tile-based approach for producing clinically relevant probability maps of prostatic carcinoma in histological sections from radical prostatectomy. Our methodology incorporates ensemble learning for feature selection and classification on expert-annotated images. Random forest feature selection performed over varying training sets provides a subset of generalized CIEL*a*b* co-occurrence texture features, while sample selection strategies with minimal constraints reduce training data requirements to achieve reliable results. Ensembles of classifiers are built using expert-annotated tiles from training images, and scores for the probability of cancer presence are calculated from the responses of each classifier in the ensemble. Spatial filtering of tile-based texture features prior to classification results in increased heat-map coherence as well as AUC values of 95% using ensembles of either random forests or support vector machines. Our approach is designed for adaptation to different imaging modalities, image features, and histological decision domains. | |
dc.language.iso | en | en |
dc.rights | Archived with thanks to Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society | en_GB |
dc.subject.mesh | Algorithms | |
dc.subject.mesh | Color | |
dc.subject.mesh | Histological Techniques | |
dc.subject.mesh | Humans | |
dc.subject.mesh | Image Interpretation, Computer-Assisted | |
dc.subject.mesh | Male | |
dc.subject.mesh | Pattern Recognition, Automated | |
dc.subject.mesh | Prostatic Neoplasms | |
dc.title | Ensemble based system for whole-slide prostate cancer probability mapping using color texture features. | en_GB |
dc.type | Article | en |
dc.contributor.department | School of Computer Science and Informatics, University College Dublin, Ireland. | en_GB |
dc.identifier.journal | Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society | en_GB |
dc.description.province | Leinster | en |
html.description.abstract | We present a tile-based approach for producing clinically relevant probability maps of prostatic carcinoma in histological sections from radical prostatectomy. Our methodology incorporates ensemble learning for feature selection and classification on expert-annotated images. Random forest feature selection performed over varying training sets provides a subset of generalized CIEL*a*b* co-occurrence texture features, while sample selection strategies with minimal constraints reduce training data requirements to achieve reliable results. Ensembles of classifiers are built using expert-annotated tiles from training images, and scores for the probability of cancer presence are calculated from the responses of each classifier in the ensemble. Spatial filtering of tile-based texture features prior to classification results in increased heat-map coherence as well as AUC values of 95% using ensembles of either random forests or support vector machines. Our approach is designed for adaptation to different imaging modalities, image features, and histological decision domains. |