Propagating Uncertainty in Multi-Stage Bayesian Convolutional Neural Networks with Application to Pulmonary Nodule Detection

TL;DR

This paper introduces methods to propagate and fuse uncertainty in multi-stage Bayesian CNNs for pulmonary nodule detection, demonstrating improved accuracy and confidence in predictions.

Contribution

It presents novel techniques for propagating uncertainty across multiple CNN stages, enhancing detection performance in medical imaging applications.

Findings

Uncertainty propagation improves overall detection accuracy.

Fusing uncertainty enhances model confidence.

Multi-stage Bayesian CNNs outperform non-Bayesian counterparts.

Abstract

Motivated by the problem of computer-aided detection (CAD) of pulmonary nodules, we introduce methods to propagate and fuse uncertainty information in a multi-stage Bayesian convolutional neural network (CNN) architecture. The question we seek to answer is "can we take advantage of the model uncertainty provided by one deep learning model to improve the performance of the subsequent deep learning models and ultimately of the overall performance in a multi-stage Bayesian deep learning architecture?". Our experiments show that propagating uncertainty through the pipeline enables us to improve the overall performance in terms of both final prediction accuracy and model confidence.