Out-of-sample extrapolation using semi-supervised manifold learning (OSE-SSL): Content-based image retrieval for prostate histology grading

TY - CHAP

T1 - Out-of-sample extrapolation using semi-supervised manifold learning (OSE-SSL): Content-based image retrieval for prostate histology grading

AU - Sparks, Rachel

AU - Madabhushi, Anant

PY - 2011

Y1 - 2011

N2 - Content-based image retrieval (CBIR) retrieves database images most similar to the query image by (1) extracting quantitative image descriptors and (2) calculating similarity between database and query image descriptors. Recently, manifold learning (ML) has been used to perform CBIR in a low dimensional representation of the high dimensional image descriptor space to avoid the curse of dimensionality. ML schemes are computationally expensive, requiring an eigenvalue decomposition (EVD) for every new query image to learn its low dimensional representation. We present out-of-sample extrapolation utilizing semi-supervised ML (OSE-SSL) to learn the low dimensional representation without recomputing the EVD for each query image. OSE-SSL incorporates semantic information, partial class label, into a ML scheme such that the low dimensional representation co-localizes semantically similar images. In the context of prostate histopathology, gland morphology is an integral component of the Gleason score which enables discrimination between prostate cancer aggressiveness. Images are represented by shape features extracted from the prostate gland. CBIR with OSE-SSL for prostate histology obtained from 58 patient studies, yielded an area under the precision recall curve (AUPRC) of 0.53 ± 0.03 comparatively a CBIR with Principal Component Analysis (PCA) to learn a low dimensional space yielded an AUPRC of 0.44 ± 0.01. Manual examination of prostate histopathology by an expert pathologist is the current gold standard for prostate cancer diagnosis, with roughly 242,000 new cases ever year

AB - Content-based image retrieval (CBIR) retrieves database images most similar to the query image by (1) extracting quantitative image descriptors and (2) calculating similarity between database and query image descriptors. Recently, manifold learning (ML) has been used to perform CBIR in a low dimensional representation of the high dimensional image descriptor space to avoid the curse of dimensionality. ML schemes are computationally expensive, requiring an eigenvalue decomposition (EVD) for every new query image to learn its low dimensional representation. We present out-of-sample extrapolation utilizing semi-supervised ML (OSE-SSL) to learn the low dimensional representation without recomputing the EVD for each query image. OSE-SSL incorporates semantic information, partial class label, into a ML scheme such that the low dimensional representation co-localizes semantically similar images. In the context of prostate histopathology, gland morphology is an integral component of the Gleason score which enables discrimination between prostate cancer aggressiveness. Images are represented by shape features extracted from the prostate gland. CBIR with OSE-SSL for prostate histology obtained from 58 patient studies, yielded an area under the precision recall curve (AUPRC) of 0.53 ± 0.03 comparatively a CBIR with Principal Component Analysis (PCA) to learn a low dimensional space yielded an AUPRC of 0.44 ± 0.01. Manual examination of prostate histopathology by an expert pathologist is the current gold standard for prostate cancer diagnosis, with roughly 242,000 new cases ever year

KW - Content-Based Image Retrieval

KW - Graph Embedding

KW - Nonlinear Dimensionality Reduction

KW - Nystrm Method

KW - Out-of-Sample Extrapolation

KW - Semi-Supervised Manifold Learning

U2 - 10.1109/ISBI.2011.5872510

DO - 10.1109/ISBI.2011.5872510

M3 - Chapter

C2 - 27264985

SN - 9781424441280

T3 - Proceedings - International Symposium on Biomedical Imaging

SP - 734

EP - 737

BT - Proceedings - International Symposium on Biomedical Imaging

ER -