Efficient Epistemic Uncertainty Estimation in Cerebrovascular Segmentation

2025-03-28

AI Summary

Deep learning models for cerebral vascular MR image segmentation suffer from low trustworthiness and lack of calibrated uncertainty quantification. Method: This work introduces cognitive uncertainty modeling—novel in this domain—via a lightweight ensemble framework that synergistically integrates Bayesian approximation and deep ensembles to jointly optimize segmentation accuracy, inference efficiency, and uncertainty calibration. It further incorporates uncertainty-driven post-processing, out-of-distribution (OOD) detection, and robustness evaluation modules. Results: Evaluated on both in-distribution and OOD data, the approach improves Dice scores by 1.8–3.2% after excluding high-uncertainty regions. The model explicitly characterizes prediction limitations, yielding clinically interpretable and deployable uncertainty estimates. This constitutes the first systematic, cognition-aware uncertainty solution for trustworthy cerebral vascular segmentation.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Computer Vision: Segmentation
• Reasoning under Uncertainty: Uncertainty Representations

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Brain vessel segmentation of MR scans is a critical step in the diagnosis of cerebrovascular diseases. Due to the fine vessel structure, manual vessel segmentation is time consuming. Therefore, automatic deep learning (DL) based segmentation techniques are intensively investigated. As conventional DL models yield a high complexity and lack an indication of decision reliability, they are often considered as not trustworthy. This work aims to increase trust in DL based models by incorporating epistemic uncertainty quantification into cerebrovascular segmentation models for the first time. By implementing an efficient ensemble model combining the advantages of Bayesian Approximation and Deep Ensembles, we aim to overcome the high computational costs of conventional probabilistic networks. Areas of high model uncertainty and erroneous predictions are aligned which demonstrates the effectiveness and reliability of the approach. We perform extensive experiments applying the ensemble model on out-of-distribution (OOD) data. We demonstrate that for OOD-images, the estimated uncertainty increases. Additionally, omitting highly uncertain areas improves the segmentation quality, both for in- and out-of-distribution data. The ensemble model explains its limitations in a reliable manner and can maintain trustworthiness also for OOD data and could be considered in clinical applications

Problem

• Estimating epistemic uncertainty in brain vessel segmentation
• Reducing computational costs in probabilistic segmentation models
• Improving trustworthiness for out-of-distribution clinical data

Innovation

• Incorporates epistemic uncertainty quantification in segmentation
• Efficient ensemble model combining Bayesian and Deep Ensembles
• Improves segmentation by omitting highly uncertain areas