A novel system for scoring of hormone receptors in breast cancer histopathology slides

Abstract

Grading of breast cancer is often done by an expert pathologist based on their analysis of micro-level structural features of the cancerous tissue specimen as well as the level of presence of certain protein molecules in the specimen. The process of assessment of the level of presence of estrogen and progesterone receptors molecules is subjective by its very nature and therefore, causes large inter-expert and sometimes even intra-expert variability, potentially adding noise to the process of selecting the treatment regime for the patient. Quantification of immunohistochemical stains is critical for an objective assessment of breast cancer histopathology specimens. We present a fast, compact and inexpensive system for scoring the Estrogen and Progesterone hormone receptors in breast cancer histopathology slides using image analysis algorithm. We describe hardware and software issues in the construction of the system, and present a comparison of scores produced by our system to those produced by many expert pathologists.