Uncertainty-Aware Pneumothorax Segmentation and Selective Prediction in Chest Radiographs
Abstract
Overview
This manuscript investigates uncertainty-aware pneumothorax segmentation and selective prediction in chest radiographs using a ResNet34 U-Net, Monte Carlo Dropout, and deep ensembles on the SIIM-ACR dataset.
Evaluation and limitations
The study examines patient-isolated segmentation, calibration, error discrimination, risk-coverage behaviour, and projection subgroups. It reports limitations including checkpoint-selection variance, an unreliable fixed operating threshold, and ensemble uncertainty rankings that can perform worse than random. These findings motivate further validation rather than demonstrating readiness for clinical deployment.
Supporting materials
Code and research documentation
This deposit contains the author manuscript sourced from repository commit 023e3314dea2b247d10b0c82d3c786db43b7cd50. It does not include the SIIM-ACR dataset or model weights. No journal acceptance or peer-review status is asserted by this deposit.
Related
- Uncertainty-Aware Selective Prediction in Brain-Tumour MRI Classification: Selective Risk Reduction and Component-Wise Reliability Findings
- AI-Based Automated Segmentation of the Mandibular Cortex and Mental Foramen for Radiomorphometric Measurement on Panoramic Radiographs: A Feasibility Study
- ROI-based binary caries segmentation in intraoral periapical radiographs using a ResNet-conditioned U-Net with FiLM-modulated decoder
- Deep Learning Analysis of Routine Preoperative Knee Radiographs for Five-Year Prediction of Contralateral Arthroplasty
- Spectral-Distribution Uncertainty Modeling for Robust Cross-Domain Medical Image Segmentation
BibTeX
@misc{sah2026uncertaintyaware,
title = {{Uncertainty-Aware Pneumothorax Segmentation and Selective Prediction in Chest Radiographs}},
author = {Raju Sah},
howpublished = {Zenodo},
year = {2026},
month = oct,
doi = {10.5281/zenodo.23558665},
url = {https://zenodo.org/records/23558665},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Uncertainty-Aware Pneumothorax Segmentation and Selective Prediction in Chest Radiographs
AU - Raju Sah
PY - 2026
DA - 2026/10/22
PB - Zenodo
DO - 10.5281/zenodo.23558665
UR - https://zenodo.org/records/23558665
AB - Overview This manuscript investigates uncertainty-aware pneumothorax segmentation and selective prediction in chest radiographs using a ResNet34 U-Net, Monte Carlo Dropout, and deep ensembles on the SIIM-ACR dataset. Evaluation and limitations The study examines patient-isolated segmentation, calibration, error discrimination, risk-coverage behaviour, and projection subgroups. It reports limitations including checkpoint-selection variance, an unreliable fixed operating threshold, and ensemble uncertainty rankings that can perform worse than random. These findings motivate further validation rather than demonstrating readiness for clinical deployment. Supporting materials Code and research documentation This deposit contains the author manuscript sourced from repository commit 023e3314dea2b247d10b0c82d3c786db43b7cd50. It does not include the SIIM-ACR dataset or model weights. No journal acceptance or peer-review status is asserted by this deposit.
ER -