A Multi-Level Probabilistic Deep Learning Network Augmented With Normalizing Flow for Ambiguous Medical Image Segmentation
Open Access
- 31 December 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 4063-4079
Abstract
Medical image segmentation often involves inherent uncertainty due to inter observer variability. In this case, a single deterministic mask obtained by conventional segmentation networks, such as U-Net, cannot capture the distribution of plausible expert annotations, risking missed clinically relevant variants. In order to enable uncertainty quantification and reflect inter expert variability, probabilistic models like Probabilistic U-Net are used, which perform aleatoric or ambiguous segmentation where a latent space is sampled to generate multiple segmentation masks. However, the common use of a conditioned unimodal posterior in these models fails to represent true multimodality, leading to mode bias and limited diversity. To address these limitations, a multi-level Probabilistic U-Net augmented with normalizing flows is proposed to enhance the expressiveness of the latent distribution. The multi-level design induces multiple latent distributions in separate levels of U-Net, enabling more diverse sampling, while the flow module transforms the posterior to add data required modes and expand representational capacity, thereby enriching the expressiveness of the distributions. The proposed flow incorporated multi-level network enables a more flexible and powerful distribution, thereby enhancing the model’s ability to generate high fidelity segmentation masks. Extensive experiments on some publicly available datasets with multiple expert annotations per image demonstrate that the proposed model reduces generalized energy distance (GED), preserves clinically meaningful diversity and sharpens boundary fidelity, with latent grid analyses indicating fuller mode coverage and fewer artifacts. Collectively, these results indicate that the proposed framework advances accuracy, robustness, and clinical reliability for aleatoric, uncertainty aware medical image segmentation.Keywords
This publication has 36 references indexed in Scilit:
- Grounding DINO: Marrying DINO with Grounded Pre-training for Open-Set Object Detection2024
- EndoSparse: Real-Time Sparse View Synthesis of Endoscopic Scenes using Gaussian Splatting2024
- Calibrating ensembles for scalable uncertainty quantification in deep learning-based medical image segmentationComputers in Biology and Medicine, 2023
- Intervention & Interaction Federated Abnormality Detection with Noisy Clients2022
- Improving Aleatoric Uncertainty Quantification in Multi-annotated Medical Image Segmentation with Normalizing Flows2021
- Is Segmentation Uncertainty Useful?2021
- Uncertainty Quantification in Medical Image Segmentation with Normalizing Flows2020
- PHiSeg: Capturing Uncertainty in Medical Image Segmentation2019
- Brain Tumor Segmentation and Radiomics Survival Prediction: Contribution to the BRATS 2017 Challenge2018
- U-Net: Convolutional Networks for Biomedical Image Segmentation2015