SI-MIL: Taming Deep MIL for Self-Interpretability in Gigapixel Histopathology
This addresses the problem of limited interpretability for pathologists in whole slide image analysis, offering an incremental improvement by integrating interpretability without sacrificing performance.
The paper tackles the challenge of interpretability in Multiple Instance Learning (MIL) for gigapixel histopathology analysis by proposing SI-MIL, a self-interpretable method that provides feature-level insights and achieves competitive results on WSI-level prediction tasks across three cancer types.
Introducing interpretability and reasoning into Multiple Instance Learning (MIL) methods for Whole Slide Image (WSI) analysis is challenging, given the complexity of gigapixel slides. Traditionally, MIL interpretability is limited to identifying salient regions deemed pertinent for downstream tasks, offering little insight to the end-user (pathologist) regarding the rationale behind these selections. To address this, we propose Self-Interpretable MIL (SI-MIL), a method intrinsically designed for interpretability from the very outset. SI-MIL employs a deep MIL framework to guide an interpretable branch grounded on handcrafted pathological features, facilitating linear predictions. Beyond identifying salient regions, SI-MIL uniquely provides feature-level interpretations rooted in pathological insights for WSIs. Notably, SI-MIL, with its linear prediction constraints, challenges the prevalent myth of an inevitable trade-off between model interpretability and performance, demonstrating competitive results compared to state-of-the-art methods on WSI-level prediction tasks across three cancer types. In addition, we thoroughly benchmark the local and global-interpretability of SI-MIL in terms of statistical analysis, a domain expert study, and desiderata of interpretability, namely, user-friendliness and faithfulness.