HistoKernel: Whole Slide Image Level Maximum Mean Discrepancy Kernels for Pan-Cancer Predictive Modelling

TL;DR

HistoKernel introduces a novel kernel method based on Maximum Mean Discrepancy to improve Whole Slide Image predictive modeling in computational pathology, outperforming existing methods across multiple tasks and enabling explainability.

Contribution

It pioneers the use of kernel-based distributional similarity measures for WSI-level prediction and integrates multi-modal data with explainability features.

Findings

Outperforms existing deep learning methods in multiple tasks

Effective across retrieval, drug sensitivity, mutation classification, and survival analysis

Provides a novel patch-level explainability approach

Abstract

Machine learning in computational pathology (CPath) often aggregates patch-level predictions from multi-gigapixel Whole Slide Images (WSIs) to generate WSI-level prediction scores for crucial tasks such as survival prediction and drug effect prediction. However, current methods do not explicitly characterize distributional differences between patch sets within WSIs. We introduce HistoKernel, a novel Maximum Mean Discrepancy (MMD) kernel that measures distributional similarity between WSIs for enhanced prediction performance on downstream prediction tasks. Our comprehensive analysis demonstrates HistoKernel's effectiveness across various machine learning tasks, including retrieval (n = 9,362), drug sensitivity regression (n = 551), point mutation classification (n = 3,419), and survival analysis (n = 2,291), outperforming existing deep learning methods. Additionally, HistoKernel seamlessly integrates multi-modal data and offers a novel perturbation-based method for patch-level explainability. This work pioneers the use of kernel-based methods for WSI-level predictive modeling, opening new avenues for research. Code is available at https://github.com/pkeller00/HistoKernel.