EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Uncertainty quantification of U‐Net based segmentation tool using conformal prediction

Bailey J Borden, John B Graham-Knight, Patricia Lasserre, Sarah Lucas, Rasika D Rajapakshe

Journal of Applied Clinical Medical Physics · 2026-07 · 27 (7)

DOI: 10.1002/acm2.70663 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Applied Clinical Medical Physics, online erschienen am 23.06.2026. Verfasst von Bailey J Borden und 4 weiteren Autor:innen. Beteiligte Einrichtung: The University of British Columbia - Okanagan Campus, Kelowna, British Columbia, Canada. Hauptthemen laut MEDLINE: Tomography, X-Ray Computed, Algorithms, Radiotherapy Planning, Computer-Assisted, Breast Neoplasms. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 37 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: BC Cancer Studentship.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

The radiation therapy treatment process is very labour intensive, and artificial intelligence (AI) based auto contouring tools are increasingly being adopted to improve efficiency. However, current acceptance testing of AI auto contouring algorithms relies primarily on area‐ and distance‐based metrics, with limited assessment of model uncertainty.

Purpose

To demonstrate conformal prediction as a complementary error analysis technique for AI auto contouring algorithms, providing spatially localized uncertainty information that traditional metrics do not capture.

Methods

A U‐Net architecture with a ResNet‐34 encoder was trained on BC Cancer breast data to segment the left lung, right lung, and the heart. Initial testing was performed on a subset of 376 computed tomography (CT) scans using both area‐based (IoU) and distance‐based (HD95) metrics. Conformal prediction using adaptive prediction sets was then performed on 138 CT scans. The change in the derivative of the intersection over union (IoU) between the original predictions and the conformal predictions was observed with respect to the selected confidence level.

Results

U‐Net achieved a mean IoU of 0.924 and a mean HD95 of 11.35. When conformal prediction was applied using a 90% confidence threshold, the percent differences between the conformal prediction IoUs and the U‐Net prediction IoUs were 1.01%, 0.89%, and 1.46% for the left lung, right lung, and the heart, respectively. The IoU derivatives differed significantly between true positive and false positive structure predictions ( P < 0.001), indicating that conformal prediction can distinguish reliable predictions from unreliable ones.

Conclusions

Conformal prediction provides an additional tool for acceptance testing of AI auto contouring algorithms. Beyond traditional area‐ and distance‐based metrics, it spatially localizes uncertain predictions and offers a mechanism for identifying false positives.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Tomography, X-Ray ComputedAlgorithmsRadiotherapy Planning, Computer-AssistedBreast NeoplasmsImage Processing, Computer-AssistedArtificial IntelligenceHumansUncertaintyRadiotherapy DosageRadiotherapy, Intensity-ModulatedFemale

Stichwörter der Autor:innen

CTautomationbreastcomputer visiondeep learningmachine learningmodel uncertaintysegmentation

Autor:innen und Institutionen

  • The University of British Columbia - Okanagan Campus, Kelowna, British Columbia, Canada.
  • BC Cancer - Kelowna, Kelowna, British Columbia, Canada.

Förderung

  • BC Cancer Studentship

Zitieren

APA

Borden, B. J., Graham-Knight, J. B., Lasserre, P., Lucas, S., & Rajapakshe, R. D. (2026). Uncertainty quantification of U‐Net based segmentation tool using conformal prediction. Journal of Applied Clinical Medical Physics, 27(7). https://doi.org/10.1002/acm2.70663

Vancouver

Borden BJ, Graham-Knight JB, Lasserre P, Lucas S, Rajapakshe RD. Uncertainty quantification of U‐Net based segmentation tool using conformal prediction. Journal of Applied Clinical Medical Physics. 2026;27(7). doi:10.1002/acm2.70663

BibTeX
@article{Borden2026,
  author = {Borden, Bailey J and Graham-Knight, John B and Lasserre, Patricia and Lucas, Sarah and Rajapakshe, Rasika D},
  title = {Uncertainty quantification of U‐Net based segmentation tool using conformal prediction},
  journal = {Journal of Applied Clinical Medical Physics},
  year = {2026},
  volume = {27},
  number = {7},
  doi = {10.1002/acm2.70663}
}
Kostenlos starten →