SIB-MIL: Sparsity-Induced Bayesian Neural Network for Robust Multiple Instance Learning on Whole Slide Image Analysis
- 27 November 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 45 (4) , 1638-1650
Abstract
Multiple instance learning (MIL) has shown prominent success in analyzing whole slide histopathology images (WSIs). However, existing MIL methods often suffer from overfitting due to weak supervision and the “needle-in-a-haystack” nature of WSIs. Additionally, most deterministic approaches lack a mechanism for uncertainty quantification. While Bayesian neural networks (BNNs) have emerged as a promising solution to mitigate overfitting and enable uncertainty estimation by imposing prior constraints, commonly used Gaussian BNNs exhibit unstable posterior predictive distributions under weak supervision and suffer from high prediction variance. To tackle these challenges, we propose a sparsity-induced Bayesian Neural Network to be adopted in the MIL scheme, named SIB-MIL, for robust WSI prediction. Instead of using Gaussian prior distributions, we place a sparsity-induced prior, the Horse-shoe prior, on the BNN parameters to address the variance overflowing issue. Such sparsity also filters unimportant noise and highlights salient regions, which only occupy a small proportion in WSIs. Empirical evaluations on cancer classification and subtyping tasks corroborate that not only can our method improve the existing MIL networks, but it also performs well in uncertainty quantification. Codes are available at https://github.com/HKU-MedAI/SIB-MIL.Keywords
Funding Information
- The Research Council (27206123)
- The Research Council (17200125)
- The Research Council (C5055-24G)
- The Research Council (T45-401/22-N)
- Innovation and Technology Fund (ITS/274/22)
- Innovation and Technology Fund (GHP/318/22GD)
- National Natural Science Foundation of China (62201483)
- Guangdong Natural Science Fund (2024A1515011875)
This publication has 33 references indexed in Scilit:
- cDP-MIL: Robust Multiple Instance Learning via Cascaded Dirichlet Process2024
- Exploring Multiple Instance Learning (MIL): A brief surveyExpert Systems with Applications, 2024
- Simultaneous instance pooling and bag representation selection approach for multiple-instance learning (MIL) using vision transformerNeural Computing and Applications, 2024
- Robust bag classification approach for multi-instance learning via subspace fuzzy clusteringExpert Systems with Applications, 2022
- Deep Gaussian mixture model based instance relevance estimation for multiple instance learning applicationsApplied Intelligence, 2022
- Deep Gaussian processes for multiple instance learning: Application to CT intracranial hemorrhage detectionComputer Methods and Programs in Biomedicine, 2022
- Deep multiple instance selectionScience China Information Sciences, 2021
- Diagnostic Assessment of Deep Learning Algorithms for Detection of Lymph Node Metastases in Women With Breast CancerJAMA, 2017
- Defining and Measuring Diagnostic Uncertainty in Medicine: A Systematic ReviewJournal of General Internal Medicine, 2017
- Label Stability in Multiple Instance Learning2015