Sparse Bayesian Networks: Efficient Uncertainty Quantification in Medical Image Analysis

Abstract

Efficiently quantifying predictive uncertainty in medical images remains a challenge. While Bayesian neural networks (BNN) offer predictive uncertainty, they require substantial computational resources to train. Although Bayesian approximations such as ensembles have shown promise, they still suffer from high training and inference costs. Existing approaches mainly address the costs of BNN inference post-training, with little focus on improving training efficiency and reducing parameter complexity. This study introduces a training procedure for a sparse (partial) Bayesian network. Our method selectively assigns a subset of parameters as Bayesian by assessing their deterministic saliency through gradient sensitivity analysis. The resulting network combines deterministic and Bayesian parameters, exploiting the advantages of both representations to achieve high task-specific performance and minimize predictive uncertainty. Demonstrated on multi-label ChestMNIST for classification and ISIC, LIDC-IDRI for segmentation, our approach achieves competitive performance and predictive uncertainty estimation by reducing Bayesian parameters by over 95\%, significantly reducing computational expenses compared to fully Bayesian and ensemble methods.

References in corpus (6)

• The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions
• ChestX-ray8: Hospital-scale Chest X-ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseases
• MedMNIST v2 -- A large-scale lightweight benchmark for 2D and 3D biomedical image classification
• Hands-on Bayesian Neural Networks -- a Tutorial for Deep Learning Users
• Pre-Defined Sparse Neural Networks with Hardware Acceleration
• Bayesian Neural Networks at Scale: A Performance Analysis and Pruning Study