What question did this study set out to answer?

To evaluate deep learning models for automatic segmentation of clinical target volume in breast radiotherapy.

May 4, 2026SHILAP Revista de lepidopterología0 citationsOpen Access

Deep ensemble optimized models for probabilistic CTV breast segmentation

CRCecilia RianiVita-Salute San Raffaele University CRCecilia RianiVita-Salute San Raffaele University MUMaria Giulia Ubeira-GabelliniVita-Salute San Raffaele University

Key Points

To evaluate deep learning models for automatic segmentation of clinical target volume in breast radiotherapy.
Evaluated six deep learning models for CTV segmentation using planning CTs from 861 training cases and 100 test cases.
Used standard metrics like Dice Similarity Coefficient and Hausdorff Distance for model performance assessment.
Applied Friedman and Conover tests to analyze differences in model accuracy.
All models achieved satisfactory performance (Dice = 0.90) with UNet, DynUNet, nnU-Net, and MedSAM2-A showing highest accuracy (ASD = 1.5 mm, HD95 = 3.8 mm).
Removed two models (SegResNetDS, MS-B) based on lower performance from probability map construction.
Average difference in isoprobability volumes between 100% and 25% was 123 ± 81 cm3, highlighting uncertainty areas in CTV borders.

Abstract

IntroductionTo optimize radiotherapy treatment and minimize toxicities, effective segmentation of organs-at-risk (OARs) and clinical target volume (CTV) is essential. Deep learning (DL) models can achieve high segmentation accuracy through careful tuning. However, their reliability also hinges on addressing uncertainties stemming from variability in clinical contouring practices. This study systematically evaluates six advanced DL models for automatic CTV segmentation in whole-breast radiotherapy. It further leverages the top-performing models to construct a probability map, aiming to improve consistency and mitigate bias in clinical predictions.MethodsPlanning CTs from a single institute (861: training, 100: “temporal” test) were used to train six models for right/left breast CTV segmentation simultaneously: UNet, SegResNetDS, DynUNet (MONAI), nnU-Net (Total Segmentator), MedSAM2 (MS-A: CT-specific weights, MS-B: general medical-image weights). MedSAM2 employed a strong caveat: it needs a spatial prompt to make the prediction. Standard metric such as Dice Similarity Coefficient, Average Surface Distance and Hausdorff Distance were used to evaluate prediction compared to ground truth. A Friedman test followed by post-hoc pairwise comparisons through Conover test were conducted on the temporal test set. In addition, the best models derived (in line with inter-observer variability IOV) were used to build model-based probability maps and quantify the differences between high concordance (100%) and lower concordance (25%) isoprobabilities to clinician CTV.ResultsAll models attained overall satisfactory performance comparable to IOV (Dice = 0.90). Among them, UNet, DynUNet, nnU-Net, and MedSAM2-A demonstrated the highest equivalent accuracy, with average ASD = 1.5 mm and HD95 = 3.8 mm. Conover test displayed more subtle differences with lower performance of two models out of six (SegResNetDS, MS-B), which were therefore removed for the subsequent built of a probability map. The analysis of the 100 vs. 25% isoprobability volumes on the temporal test dataset (average difference = 123 ± 81 cm3) highlighted areas of greater uncertainty at the lateral and cranio-caudal CTV borders.DiscussionDifferent DL architectures were optimized, trained and validated on a large cohort of patients, successfully predicting the CTV for both right/left breast. In addition, the first attempt to provide uncertainty maps was achieved through successful computation of DL model-based probability maps. Supported by CCM 2024 (Ministry of Health).

AI से पूछें

Bookmark

View Full Paper