Related Experiment Video
Updated: May 2, 2026
08:17
Dynamic Lung Tumor Tracking for Stereotactic Ablative Body Radiation Therapy
Published on: June 7, 2015
15.2K
Understanding and leveraging uncertainties in autosegmentation for radiotherapy
Stine Sofia Korreman1, Jintao Ren1
1Department of Clinical Medicine, Danish Center for Particle Therapy, Aarhus University, DK-8200 Aarhus, Denmark.
BJR Artificial Intelligence
|May 1, 2026
Summary
Understanding uncertainties in autosegmentation for radiotherapy is crucial. Quantifying these uncertainties can improve AI tools, guide manual adjustments, and enhance patient care.
Area of Science:
- Radiotherapy
- Medical Imaging
- Artificial Intelligence
Background:
- Autosegmentation in radiotherapy offers improved consistency, accuracy, and efficiency compared to manual delineation.
- However, uncertainties in autosegmentation, stemming from limited data and models, pose challenges.
Purpose of the Study:
- To examine the role and impact of uncertainties in radiotherapy autosegmentation.
- To explore sources, quantification, and implications of these uncertainties.
Main Methods:
- Review of uncertainty sources in autosegmentation models.
- Exploration of uncertainty quantification techniques.
- Discussion of practical applications and limitations.
Main Results:
- Uncertainties impact clinical decisions and patient outcomes.
- Quantification methods and uncertainty maps can guide manual adjustments and quality assurance.
- Limitations include computational cost and information overload.
Conclusions:
- Meaningful uncertainty quantification in autosegmentation can enhance clinical workflows.
- It builds trust in AI tools for radiotherapy.
- Ultimately, it has the potential to improve patient care.
