Interventional Bag Multi-Instance Learning On Whole-Slide Pathological Images
This addresses a confounder limiting performance in medical image analysis, offering an incremental but orthogonal enhancement to existing MIL methods.
The paper tackles the problem of spurious correlations in multi-instance learning for whole-slide pathological images by proposing IBMIL, a deconfounded bag-level prediction method based on backdoor adjustment, which achieves new state-of-the-art performance with consistent improvements over existing schemes.
Multi-instance learning (MIL) is an effective paradigm for whole-slide pathological images (WSIs) classification to handle the gigapixel resolution and slide-level label. Prevailing MIL methods primarily focus on improving the feature extractor and aggregator. However, one deficiency of these methods is that the bag contextual prior may trick the model into capturing spurious correlations between bags and labels. This deficiency is a confounder that limits the performance of existing MIL methods. In this paper, we propose a novel scheme, Interventional Bag Multi-Instance Learning (IBMIL), to achieve deconfounded bag-level prediction. Unlike traditional likelihood-based strategies, the proposed scheme is based on the backdoor adjustment to achieve the interventional training, thus is capable of suppressing the bias caused by the bag contextual prior. Note that the principle of IBMIL is orthogonal to existing bag MIL methods. Therefore, IBMIL is able to bring consistent performance boosting to existing schemes, achieving new state-of-the-art performance. Code is available at https://github.com/HHHedo/IBMIL.