Continual Learning for Tumor Classification in Histopathology Images

TL;DR

This paper explores continual learning methods for histopathology image analysis, addressing model forgetting in evolving data scenarios and demonstrating promising results for clinical tumor classification applications.

Contribution

It introduces systematic CL scenarios for digital pathology, creates a new dataset for colorectal cancer, and evaluates CL methods across multiple tumor types and settings.

Findings

CL methods improve model retention across data shifts

Augmented datasets effectively simulate real-world data variability

Online few-shot CL shows potential in resource-constrained environments

Abstract

Recent years have seen great advancements in the development of deep learning models for histopathology image analysis in digital pathology applications, evidenced by the increasingly common deployment of these models in both research and clinical settings. Although such models have shown unprecedented performance in solving fundamental computational tasks in DP applications, they suffer from catastrophic forgetting when adapted to unseen data with transfer learning. With an increasing need for deep learning models to handle ever changing data distributions, including evolving patient population and new diagnosis assays, continual learning models that alleviate model forgetting need to be introduced in DP based analysis. However, to our best knowledge, there is no systematic study of such models for DP-specific applications. Here, we propose CL scenarios in DP settings, where histopathology image data from different sources/distributions arrive sequentially, the knowledge of which is integrated into a single model without training all the data from scratch. We then established an augmented dataset for colorectal cancer H&E classification to simulate shifts of image appearance and evaluated CL model performance in the proposed CL scenarios. We leveraged a breast tumor H&E dataset along with the colorectal cancer to evaluate CL from different tumor types. In addition, we evaluated CL methods in an online few-shot setting under the constraints of annotation and computational resources. We revealed promising results of CL in DP applications, potentially paving the way for application of these methods in clinical practice.