EUVIMEDEUVIMED entdecken
ValidierungsstudieOpen Access

Deep Learning–driven Segmentation of Breast Cancer MR Imaging Using a Hybrid ResNet–Transformer Architecture and Cross-dataset Validation

Yongjian Sui, Peiru Li, Jiacheng Li, Peng Zhou, Xiaolong Zheng

Magnetic Resonance in Medical Sciences · 2026

DOI: 10.2463/mrms.mp.2026-0099 ↗

Kurz gefasst

Validierungsstudie in der Zeitschrift Magnetic Resonance in Medical Sciences, 2026. Verfasst von Yongjian Sui und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiology, Yongkang First People's Hospital, Yongkang, Zhejiang, China. Hauptthemen laut MEDLINE: Breast Neoplasms, Deep Learning, Magnetic Resonance Imaging, Image Interpretation, Computer-Assisted. Die Arbeit nennt 39 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: Zhejiang Provincial Health Science and Technology Program.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

PURPOSE: To develop and cross-dataset validate a hybrid ResNet50-Transformer U-Net for automated segmentation of breast cancer (BC) lesions on 2D dynamic contrast-enhanced MRI (DCE-MRI). METHODS: A total of 20434 BC MRI images from the BC MRI Segmentation Benchmark (BC-MRI-SEG), comprising 4 independent public datasets, were included. A strict patient-level data splitting strategy was applied to avoid data leakage. The proposed model was based on a U-Net architecture integrating a ResNet50 encoder for local feature extraction and a bottleneck Transformer module for global contextual modeling. Model performance was evaluated using a leave-one-dataset-out cross-dataset external validation strategy. Segmentation performance was assessed using the Dice similarity coefficient (DSC), intersection over union (IoU), 95% Hausdorff distance (HD95), sensitivity, and precision. RESULTS: The proposed hybrid model achieved stable segmentation performance across the evaluated datasets. In leave-one-dataset-out external validation, patient-level DSC ranged from 0.75 ± 0.13 to 0.79 ± 0.10, while IoU ranged from 0.61 ± 0.15 to 0.66 ± 0.12. In the fixed held-out comparison cohort, the proposed model achieved a DSC of 0.82 ± 0.08 and an IoU of 0.70 ± 0.10 and outperformed the evaluated baseline models (Holm-adjusted P ≤ 0.021). Qualitative analysis showed generally close agreement between the predicted masks and reference annotations across lesions with different morphologic appearances. CONCLUSION: The proposed ResNet-Transformer-based U-Net framework provided a robust approach for automated BC MRI segmentation across heterogeneous public datasets. These findings suggest that combining local feature extraction with global contextual modeling may be useful for computer-assisted breast MRI analysis and related clinical workflows.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Breast NeoplasmsDeep LearningMagnetic Resonance ImagingImage Interpretation, Computer-AssistedImage Processing, Computer-AssistedHumansFemaleConvolutional Neural NetworksDynamic Contrast Enhanced Magnetic Resonance ImagingDatasets as Topic

Stichwörter der Autor:innen

Transformer modelautomated detectionbreast cancerdeep learninglesion recognitionmagnetic resonance imaging segmentation

Autor:innen und Institutionen

  • Department of Radiology, Yongkang First People's Hospital, Yongkang, Zhejiang, China.
  • Department of Cardiothoracic and Breast Surgery, Yongkang First People's Hospital, Yongkang, Zhejiang, China.

Förderung

  • Zhejiang Provincial Health Science and Technology Program

Zitieren

APA

Sui, Y., Li, P., Li, J., Zhou, P., & Zheng, X. (2026). Deep Learning–driven Segmentation of Breast Cancer MR Imaging Using a Hybrid ResNet–Transformer Architecture and Cross-dataset Validation. Magnetic Resonance in Medical Sciences. https://doi.org/10.2463/mrms.mp.2026-0099

Vancouver

Sui Y, Li P, Li J, Zhou P, Zheng X. Deep Learning–driven Segmentation of Breast Cancer MR Imaging Using a Hybrid ResNet–Transformer Architecture and Cross-dataset Validation. Magnetic Resonance in Medical Sciences. 2026. doi:10.2463/mrms.mp.2026-0099

BibTeX
@article{Sui2026,
  author = {Sui, Yongjian and Li, Peiru and Li, Jiacheng and Zhou, Peng and Zheng, Xiaolong},
  title = {Deep Learning–driven Segmentation of Breast Cancer MR Imaging Using a Hybrid ResNet–Transformer Architecture and Cross-dataset Validation},
  journal = {Magnetic Resonance in Medical Sciences},
  year = {2026},
  doi = {10.2463/mrms.mp.2026-0099}
}
Kostenlos starten →