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

Clinical Snapshot

Key Findings

• P Value: Not reported - stated as 'no statistically significant differences'

• Effect Size: In-organ activation ratios: U-Net 71.4 ± 8.6% vs ResUNet 66.2 ± 9.1%

• Primary Outcome: Comparable segmentation performance between U-Net and ResUNet with no statistically significant differences

• Nnt Or Sensitivity: Pointing accuracy >70% for both models; boundary energy ratios U-Net 24.1 ± 4.9% vs ResUNet 21.8 ± 5.2%

• Confidence Interval: Not reported

Clinical Application

Moderate - requires specialized MR-guided radiotherapy equipment and deep learning infrastructure Relevant to major Australian cancer centers with MR-guided radiotherapy capabilities; aligns with RANZCR imaging standards and TGA medical device regulations for AI-assisted medical imaging Patients requiring abdominal radiotherapy with MR-guided adaptive treatment planning

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.

References

1. 1.Sekkat, H., Khallouqi, A., Hammouga, Y., Halimi, A., El Mouden, O., Bannan, A., Berrada, Y., & El Rhazouani, O. (2026). Explainability-informed benchmarking of two deep learning models for organ-at-risk segmentation in MR-guided adaptive radiotherapy. Journal of Medical Imaging and Radiation Sciences. https://doi.org/10.1016/j.jmir.2026.102200