AutoIHC-analyzer: Computer assisted microscopy for automated evaluation of ER, PR and Ki-67 molecular markers

Abstract

Immunohistochemical (IHC) markers viz., estrogen receptor (ER), progesterone receptor (PR) and proliferation marker Ki-67 are widely used for prognostic evaluation of breast cancer. The goal is to quantify the stained cells which are used to comment on the severity of cancer. In general, the expert pathologist performs the visual assessment task which is obviously tedious, time consuming and prone to inter-observer variability. In order to provide improved prognostic decision, there is an urgent need of developing not only reliable but also a rapid IHC quantifier. In view of this, our study aims at developing an automated IHC profiler for quantitative assessment of ER, PR and Ki-67 molecular expression from stained tissue images. We propose here to use CMYK color space assisted IHC image analytics, whereas most of the available literature suggests color deconvolution method for stain separation followed by quantification of positive and negatively stained cells. The proposed AutoIHC-Analyzer is compared with ImmunoRatio software available in public domain. From the results, it can be observed that our method provides better results for the original IHC images and comparable results for preprocessed IHC images.