EPIC-Survival: End-to-end Part Inferred Clustering for Survival Analysis, Featuring Prognostic Stratification Boosting

TL;DR

EPIC-Survival introduces an end-to-end deep learning approach that leverages full histopathology image data to improve survival prediction and risk stratification in cancer, outperforming previous methods.

Contribution

It presents a novel integrated model that combines encoding and aggregation for survival analysis and incorporates stratification boosting to enhance risk group discrimination.

Findings

Achieved a concordance-index of 0.880 on test data.

Outperformed existing models in intrahepatic cholangiocarcinoma survival prediction.

Identified histologic features distinguishing risk groups.

Abstract

Histopathology-based survival modelling has two major hurdles. Firstly, a well-performing survival model has minimal clinical application if it does not contribute to the stratification of a cancer patient cohort into different risk groups, preferably driven by histologic morphologies. In the clinical setting, individuals are not given specific prognostic predictions, but are rather predicted to lie within a risk group which has a general survival trend. Thus, It is imperative that a survival model produces well-stratified risk groups. Secondly, until now, survival modelling was done in a two-stage approach (encoding and aggregation). The massive amount of pixels in digitized whole slide images were never utilized to their fullest extent due to technological constraints on data processing, forcing decoupled learning. EPIC-Survival bridges encoding and aggregation into an end-to-end survival modelling approach, while introducing stratification boosting to encourage the model to not only optimize ranking, but also to discriminate between risk groups. In this study we show that EPIC-Survival performs better than other approaches in modelling intrahepatic cholangiocarcinoma, a historically difficult cancer to model. Further, we show that stratification boosting improves further improves model performance, resulting in a concordance-index of 0.880 on a held-out test set. Finally, we were able to identify specific histologic differences, not commonly sought out in ICC, between low and high risk groups.