Establishing Causal Relationship Between Whole Slide Image Predictions and Diagnostic Evidence Subregions in Deep Learning

TL;DR

This paper introduces CI-MIL, a novel causal inference approach for MIL in medical imaging that improves diagnosis accuracy and interpretability without pixelwise annotations by reducing feature distribution deviations.

Contribution

The paper proposes CI-MIL, which uses out-of-distribution generalization and feature re-weighting in Fourier space to enhance causal relationship detection between diagnoses and evidence regions.

Findings

Achieves 92.25% accuracy on Camelyon16

Attains 95.28% AUC on Camelyon16

Outperforms state-of-the-art methods in accuracy and interpretability

Abstract

Due to the lack of fine-grained annotation guidance, current Multiple Instance Learning (MIL) struggles to establish a robust causal relationship between Whole Slide Image (WSI) diagnosis and evidence sub-images, just like fully supervised learning. So many noisy images can undermine the network's prediction. The proposed Causal Inference Multiple Instance Learning (CI-MIL), uses out-of-distribution generalization to reduce the recognition confusion of sub-images by MIL network, without requiring pixelwise annotations. Specifically, feature distillation is introduced to roughly identify the feature representation of lesion patches. Then, in the random Fourier feature space, these features are re-weighted to minimize the cross-correlation, effectively correcting the feature distribution deviation. These processes reduce the uncertainty when tracing the prediction results back to patches. Predicted diagnoses are more direct and reliable because the causal relationship between them and diagnostic evidence images is more clearly recognized by the network. Experimental results demonstrate that CI-MIL outperforms state-of-the-art methods, achieving 92.25% accuracy and 95.28% AUC on the Camelyon16 dataset (breast cancer), while 94.29% accuracy and 98.07% AUC on the TCGA-NSCLC dataset (non-small cell lung cancer). Additionally, CI-MIL exhibits superior interpretability, as its selected regions demonstrate high consistency with ground truth annotations, promising more reliable diagnostic assistance for pathologists.