EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

MSF-Swin-ICFNet: An Image-Derived Multi-Scale Swin Transformer Framework with Cross-Scale Feature Fusion for Breast Lesion Classification and Segmentation in Mammograms

Narayanam R S Lakshmi Prasanthi, N Thirupathi Rao iD, S Deva Kumar iD

Tomography · 2026

DOI: 10.3390/tomography12090130 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Tomography, 2026. Verfasst von Narayanam R S Lakshmi Prasanthi und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Science and Engineering, Vignan's Foundation for Science, Technology & Research (Deemed to be University), Vadlamudi, Guntur 5…. Hauptthemen laut MEDLINE: Mammography, Breast Neoplasms, Radiographic Image Interpretation, Computer-Assisted. Die Arbeit nennt 28 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Department of Science and Technology, Science and Engineering Research Board (SERB), Department of Science and Technology (DST), Government of India.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Accurate breast lesion classification and localization from mammograms remain challenging because lesions vary in size, morphology, density, contrast, and boundary clarity. This study proposes MSF-Swin-ICFNet, an image-derived Multi-Scale Swin Transformer framework with Image-Derived Cross-Scale Fusion (ICF) for joint benign–malignant classification and lightweight lesion localization.

Methods

The proposed framework extracts hierarchical representations from four Swin Transformer stages and aligns, adaptively recalibrates, and fuses these multi-scale features into a shared representation for classification and lesion-mask prediction. The model was evaluated on CBIS-DDSM, comprising 891 patients and 1592 mammograms, and RTM, comprising 573 patients and 10,070 mammograms. Patient-level stratified partitioning was used to ensure that mammograms and associated annotations from the same patient remained within a single data partition. Classification was evaluated using accuracy, precision, recall, F1-score, and AUC-ROC, while lesion localization was assessed using the Dice coefficient.

Results

MSF-Swin-ICFNet achieved strong mammogram-level classification performance on the patient-disjoint test cohorts. On CBIS-DDSM, the model achieved 0.949 accuracy, 0.931 F1-score, and 0.985 AUC-ROC. On RTM, it achieved 0.968 accuracy, 0.956 F1-score, and 0.992 AUC-ROC. The corresponding Dice coefficients for lesion localization were 0.786 on CBIS-DDSM and 0.878 on RTM. Ablation experiments further supported the contribution of cross-scale fusion, scale recalibration, feature refinement, and the boundary-aware localization head.

Conclusions

MSF-Swin-ICFNet demonstrates promising retrospective performance for image-based breast lesion classification with complementary lesion localization. The results support the technical potential of adaptive hierarchical feature fusion for mammographic analysis. However, the model is not intended as a standalone diagnostic system, and independent multicentre validation and radiologist reader studies are required before clinical deployment.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

MammographyBreast NeoplasmsRadiographic Image Interpretation, Computer-AssistedHumansFemaleAlgorithmsBreast

Stichwörter der Autor:innen

Swin Transformerbreast cancerbreast lesion localizationcross-scale fusionimage-derived feature fusionlesion classificationlesion segmentationmammographymedical image analysis

Autor:innen und Institutionen

  • Department of Computer Science and Engineering, Vignan's Foundation for Science, Technology & Research (Deemed to be University), Vadlamudi, Guntur 522 213, AP, India.
  • Department of Computer Science and Engineering, Vignan's Institute of Information Technology(A), Visakhapatnam 530 049, AP, India.

ORCID-Profile: N Thirupathi Rao, S Deva Kumar

Förderung

  • Department of Science and Technology
  • Science and Engineering Research Board (SERB), Department of Science and Technology (DST), Government of India

Zitieren

APA

Lakshmi Prasanthi, N. R. S., Rao, N. T., & Kumar, S. D. (2026). MSF-Swin-ICFNet: An Image-Derived Multi-Scale Swin Transformer Framework with Cross-Scale Feature Fusion for Breast Lesion Classification and Segmentation in Mammograms. Tomography. https://doi.org/10.3390/tomography12090130

Vancouver

Lakshmi Prasanthi NRS, Rao NT, Kumar SD. MSF-Swin-ICFNet: An Image-Derived Multi-Scale Swin Transformer Framework with Cross-Scale Feature Fusion for Breast Lesion Classification and Segmentation in Mammograms. Tomography. 2026. doi:10.3390/tomography12090130

BibTeX
@article{LakshmiPrasanthi2026,
  author = {Lakshmi Prasanthi, Narayanam R S and Rao, N Thirupathi and Kumar, S Deva},
  title = {MSF-Swin-ICFNet: An Image-Derived Multi-Scale Swin Transformer Framework with Cross-Scale Feature Fusion for Breast Lesion Classification and Segmentation in Mammograms},
  journal = {Tomography},
  year = {2026},
  doi = {10.3390/tomography12090130}
}
Kostenlos starten →