AUTOMATED IMAGE BASED TOOLS FOR DIGITAL PATHOLOGY

Abstract

Cancer is a group of diseases involving abnormal cell growth affecting many body organs. General treatment of cancer involves assessment of cancerous tissues by the pathologist, followed by therapy according to the pathology report. Pathologists look for various heterogeneous cellular and cytological patterns during histopathological tissue slide assessment. This manual and tedious observation task impedes intra-observer agreement among pathologists.

Prominent nucleoli are important cytological patterns for cancer of kidney, breast, and prostate. Architecture, size, and shape of gland are also important for prostate cancer (PCa) assessment. Cribriform pattern in glands is in particular crucial for PCa assessment. Nuclear pleomorphic patterns along with prominent nucleoli are used for renal cancer grading.

This thesis proposes algorithms for prominent nucleoli detection, automated renal cancer grading, gland segmentation, and cribriform pattern detection while aiming towards effective automated grading of cancers. These algorithms aid cancer care by enhancing the pathologists' efficiency and reducing their errors.