Attention-based Interpretable Regression of Gene Expression in Histology

TL;DR

This paper introduces an interpretable deep learning model that links histology image features to gene expression levels, aiding in understanding cancer tissue morphology and potentially improving patient stratification.

Contribution

The study presents an attention-based model that estimates gene expression from histology images, revealing meaningful tissue patterns associated with gene activity in colorectal cancer.

Findings

Successfully identified hotspots of high gene expression in tissue images

Revealed connections between tissue morphology and gene expression profiles

Potential to improve patient stratification in pathology

Abstract

Interpretability of deep learning is widely used to evaluate the reliability of medical imaging models and reduce the risks of inaccurate patient recommendations. For models exceeding human performance, e.g. predicting RNA structure from microscopy images, interpretable modelling can be further used to uncover highly non-trivial patterns which are otherwise imperceptible to the human eye. We show that interpretability can reveal connections between the microscopic appearance of cancer tissue and its gene expression profiling. While exhaustive profiling of all genes from the histology images is still challenging, we estimate the expression values of a well-known subset of genes that is indicative of cancer molecular subtype, survival, and treatment response in colorectal cancer. Our approach successfully identifies meaningful information from the image slides, highlighting hotspots of high gene expression. Our method can help characterise how gene expression shapes tissue morphology and this may be beneficial for patient stratification in the pathology unit. The code is available on GitHub.