Communication Dans Un Congrès Année : 2024

SI-MIL: Taming Deep MIL for Self-Interpretability in Gigapixel Histopathology

Saarthak Kapse
  • Fonction : Auteur
Pushpak Pati
  • Fonction : Auteur
Srijan Das
  • Fonction : Auteur
Jingwei Zhang
  • Fonction : Auteur
Chao Chen
  • Fonction : Auteur
Joel Saltz
  • Fonction : Auteur
Dimitris Samaras
  • Fonction : Auteur
Rajarsi R. Gupta
  • Fonction : Auteur
Prateek Prasanna
  • Fonction : Auteur

Résumé

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-theart methods on WSI-level prediction tasks across three cancer types. In addition, we thoroughly benchmark the localand global-interpretability of SI-MIL in terms of statistical analysis, a domain expert study, and desiderata of interpretability, namely, user-friendliness and faithfulness.

Fichier principal
Vignette du fichier
CVPR_SI_MIL.pdf (26.75 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04901584 , version 1 (20-01-2025)

Identifiants

Citer

Saarthak Kapse, Pushpak Pati, Srijan Das, Jingwei Zhang, Chao Chen, et al.. SI-MIL: Taming Deep MIL for Self-Interpretability in Gigapixel Histopathology. Computer Vision and Pattern Recognition, Jun 2024, Seattle, United States. ⟨hal-04901584⟩
0 Consultations
0 Téléchargements

Altmetric

Partager

More