Explainability-informed benchmarking of two deep learning models for organ-at-risk segmentation in MR-guided adaptive radiotherapy

H Sekkat; A Khallouqi; Y Hammouga; A Halimi; O El Mouden; A Bannan; Y Berrada; O El Rhazouani

doi:10.1016/j.jmir.2026.102200

Explainability-informed benchmarking of two deep learning models for organ-at-risk segmentation in MR-guided adaptive radiotherapy

J Med Imaging Radiat Sci. 2026 Feb 13;57(3):102200. doi: 10.1016/j.jmir.2026.102200. Online ahead of print.

Authors

H Sekkat¹, A Khallouqi², Y Hammouga³, A Halimi³, O El Mouden⁴, A Bannan⁴, Y Berrada⁵, O El Rhazouani³

Affiliations

¹ Laboratory of Sciences and Engineering of Biomedicals, Biophysics and Health, Higher Institute of Health Sciences, Hassan 1st University, Settat, Morocco; Higher Institute of Nursing Professions and Health Techniques, Rabat, Morocco. Electronic address: sekkat.isss@uhp.ac.ma.
² Laboratory of Sciences and Engineering of Biomedicals, Biophysics and Health, Higher Institute of Health Sciences, Hassan 1st University, Settat, Morocco; Public Hospital of Mediouna, Department of Radiology, Mediouna, Morocco; Private Hospital of Hay Mouhamadi, Department of Radiology, Casablanca, Morocco.
³ Laboratory of Sciences and Engineering of Biomedicals, Biophysics and Health, Higher Institute of Health Sciences, Hassan 1st University, Settat, Morocco.
⁴ Laboratory of Sciences and Engineering of Biomedicals, Biophysics and Health, Higher Institute of Health Sciences, Hassan 1st University, Settat, Morocco; International Hospital of Settat, Department of Radiotherapy, Settat, Morocco.
⁵ International Hospital of Settat, Department of Radiotherapy, Settat, Morocco; Faculty of Medicine and Pharmacy, Hassan II University of Casablanca, Casablanca, Morocco.

PMID: 41690008
DOI: 10.1016/j.jmir.2026.102200

Abstract

Introduction/background: Segmentation of gastrointestinal (GI) organs-at-risk (OARs) is a critical yet time-consuming step in MR-guided adaptive radiotherapy (MRgRT), with manual delineation prone to inter- and intra-observer variability. While deep learning approaches have shown promise, their clinical adoption requires not only accuracy but also interpretability and reliability. This study benchmarks two widely used convolutional architectures, U-Net and Residual U-Net (ResUNet), for abdominal OAR segmentation, with an emphasis on explainability-oriented quantitative analysis.

Methods: An anonymized abdominal MRI dataset was used to train and evaluate U-Net and ResUNet using a 5-fold stratified group cross-validation strategy. Segmentation performance was assessed using the Dice Similarity Coefficient (DSC), Intersection-over-Union (IoU), and the 95th percentile Hausdorff Distance (HD95). Explainability was investigated using Gradient-weighted Class Activation Mapping (Grad-CAM) computed from the final convolutional layer of each network. To enable objective analysis beyond qualitative visualization, Grad-CAM activation maps were quantified using numerical localization metrics relative to ground-truth organ masks, including in-organ energy ratio, boundary energy ratio, pointing accuracy, activation Dice coefficient, centroid distance and activation entropy. Grad-CAM metrics were aggregated across gastrointestinal organs and averaged over the five validation folds.

Results: Both architectures demonstrated comparable segmentation performance across organs, with no statistically significant differences across evaluated metrics. Grad-CAM analysis showed similar region-level attention patterns, with in-organ activation ratios of 71.4 ± 8.6% for U-Net and 66.2 ± 9.1% for ResUNet, boundary energy ratios of 24.1 ± 4.9% and 21.8 ± 5.2%, respectively, and pointing accuracies exceeding 70% for both models. Uncertainty analysis based on inter-fold variability and boundary error dispersion indicated comparable stability and bounded worst-case behavior.

Discussion/conclusion: By integrating performance, uncertainty and explainability quantitative indicators, this study provides an informed benchmarking of two deep learning models for abdominal OAR segmentation. The results suggest that both U-Net and ResUNet exhibit stable and interpretable behavior under the evaluated configurations, supporting their potential use in MR-guided adaptive radiotherapy workflows where reliability and clinical trust are essential.

Keywords: MR-guided radiotherapy (MRgRT); MRI segmentation; Organs-at-risk (OARs); Resunet; U-net.