Biological interpretation of morphological patterns in histopathological whole-slide images

TLDR

Image representation is crucial for computer‑aided histopathological cancer diagnosis, where models depend on quantitative features extracted from digitized tissue slides; performance hinges on selecting informative features and regions of interest, yet a semantic gap between human visual patterns and quantitative features hampers this selection. The authors propose a framework for studying visual morphological patterns across histopathological whole‑slide images (WSIs). The framework employs data mining and information‑visualization techniques to analyze spatial patterns of features extracted from sub‑sections of WSIs. Using ovarian serous cystadenocarcinoma WSIs from TCGA, the study demonstrates that individual and multivariate image features align with biologically relevant ROIs, and that supervised feature selection can map histopathology domain knowledge to quantitative image features.

Abstract

We propose a framework for studying visual morphological patterns across histopathological whole-slide images (WSIs). Image representation is an important component of computer-aided decision support systems for histopathological cancer diagnosis. Such systems extract hundreds of quantitative image features from digitized tissue biopsy slides and produce models for prediction. The performance of these models depends on the identification of informative features for selection of appropriate regions-of-interest (ROIs) from heterogeneous WSIs and for development of models. However, identification of informative features is hindered by the semantic gap between human interpretation of visual morphological patterns and quantitative image features. We address this challenge by using data mining and information visualization tools to study spatial patterns formed by features extracted from sub-sections of WSIs. Using ovarian serous cystadenocarcinoma (OvCa) WSIs provided by the cancer genome atlas (TCGA), we show that (1) individual and (2) multivariate image features correspond to biologically relevant ROIs, and (3) supervised image feature selection can map histopathology domain knowledge to quantitative image features.