In this study, we present ULS+, an enhanced version of the Universal Lesion Segmentation (ULS) model. The original ULS model segments lesions across the whole body in CT scans given volumes of interest (VOIs) centered around a click-point. Since its release, several new public datasets have become available that can further improve model performance. ULS+ incorporates these additional datasets and uses smaller input image sizes, resulting in higher accuracy and faster inference. We compared ULS and ULS+ using the Dice score and robustness to click-point location on the ULS23 Challenge test data and a subset of the Longitudinal-CT dataset. In all comparisons, ULS+ significantly outperformed ULS. Additionally, ULS+ ranks first on the ULS23 Challenge test-phase leaderboard. By maintaining a cycle of data-driven updates and clinical validation, ULS+ establishes a foundation for robust and clinically relevant lesion segmentation models.
@article{arxiv.2601.02988,
title = {ULS+: Data-driven Model Adaptation Enhances Lesion Segmentation},
author = {Rianne Weber and Niels Rocholl and Max de Grauw and Mathias Prokop and Ewoud Smit and Alessa Hering},
journal= {arXiv preprint arXiv:2601.02988},
year = {2026}
}
Comments
Accepted for publication at BVM 2026 (Bildverarbeitung f\"ur die Medizin), peer-reviewed conference paper