EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Research on the establishment of a four-classification model for breast mass ultrasound images based on transfer learning

Jinlao Li iD, Bo Zhao, YongQi Chen, Guang Wang, XueLiang Tian, ShuYu Zhou

PLOS One · 2026

DOI: 10.1371/journal.pone.0340111 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, 2026. Verfasst von Jinlao Li und 5 weiteren Autor:innen. Beteiligte Einrichtung: Diagnostic Ultrasound Center, The First Hospital of Jilin University, Changchun, China. Hauptthemen laut MEDLINE: Breast Neoplasms, Ultrasonography, Mammary, Breast. Die Arbeit nennt 32 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Breast mass classification via ultrasound is critical for early diagnosis but remains challenging due to overlapping morphological features. This study evaluates the efficacy of optimized transfer learning (TL) based on deep convolutional neural networks (DCNNs) in distinguishing four breast mass categories: invasive ductal carcinoma (IDC), fibroadenoma (FA), mucinous carcinoma (MC), and inflammatory mass (IM). Our approach comprises four key steps: (1) applying extensive data augmentation to a retrospective dataset of 346 ultrasound images from 294 patients (November 2021-March 2023) classified by pathological confirmation; (2) implementing transfer learning with five pretrained DCNN models (ResNet18, ResNet50, DenseNet121, MobileNetV2, GoogLeNet); (3) optimizing model hyperparameters for four-class classification; (4) comparing performance metrics (accuracy, precision, recall, F1-score, AUC-ROC) against two senior radiologists (8- and 10-years’ experience) on a held-out test set. Results: The DenseNet121 and GoogLeNet models demonstrated superior overall accuracy (0.912 and 0.926, respectively), significantly outperforming the radiologists’ consensus (0.691) (p < 0.01). In class-specific analysis, the optimized models achieved notably higher accuracy for IDC, MC, and IM. Statistical testing confirmed the significant performance improvement of the top models over human experts. Optimized TL with DCNNs, particularly DenseNet121 and GoogLeNet, enables highly accurate four-class breast mass classification in ultrasound images, surpassing expert radiologist performance with statistical significance. This approach holds promise for clinical decision support. Code is publicly available at https://github.com/jinlao777/BCC .

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Breast NeoplasmsUltrasonography, MammaryBreastHumansFemaleRetrospective StudiesNeural Networks, ComputerMiddle AgedAdultFibroadenomaCarcinoma, Ductal, BreastAdenocarcinoma, MucinousDeep LearningAged

Autor:innen und Institutionen

  • Diagnostic Ultrasound Center, The First Hospital of Jilin University, Changchun, China.
  • Breast Surgery Department, The First Hospital of Jilin University, Changchun, China.
  • Pharmaceutical College, University Of beihua University, Yanji, China.

ORCID-Profile: Jinlao Li

Zitieren

APA

Li, J., Zhao, B., Chen, Y., Wang, G., Tian, X., & Zhou, S. (2026). Research on the establishment of a four-classification model for breast mass ultrasound images based on transfer learning. PLOS One. https://doi.org/10.1371/journal.pone.0340111

Vancouver

Li J, Zhao B, Chen Y, Wang G, Tian X, Zhou S. Research on the establishment of a four-classification model for breast mass ultrasound images based on transfer learning. PLOS One. 2026. doi:10.1371/journal.pone.0340111

BibTeX
@article{Li2026,
  author = {Li, Jinlao and Zhao, Bo and Chen, YongQi and Wang, Guang and Tian, XueLiang and Zhou, ShuYu},
  title = {Research on the establishment of a four-classification model for breast mass ultrasound images based on transfer learning},
  journal = {PLOS One},
  year = {2026},
  doi = {10.1371/journal.pone.0340111}
}
Kostenlos starten →