EUVIMEDEUVIMED entdecken
ValidierungsstudieMultizentrische Studie

Deep Learning-Based Model for Breast Implant Classification in Ultrasonography: A Multi-Institutional Model Development and Validation Study

Hyokyung Yoo, Seoi Jeong, Hyoun-Joong Kong, Jeongmok Cho, Hak Chang, Sungwan Kim, Ki Yong Hong

Aesthetic Surgery Journal · 29.01.2026 · 46 (2) · 137-149

DOI: 10.1093/asj/sjaf220 ↗

Kurz gefasst

Validierungsstudie in der Zeitschrift Aesthetic Surgery Journal, online erschienen am 30.10.2025. Verfasst von Hyokyung Yoo und 6 weiteren Autor:innen. Hauptthemen laut MEDLINE: Deep Learning, Breast Implants, Ultrasonography, Mammary, Breast Implantation. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 47 Literaturangaben. Gefördert von: NAVER Digital Bio Innovation Research Fund, NAVER Corporation.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

The increasing prevalence of implant-based breast surgeries highlights a critical gap in patient knowledge regarding implant information, exacerbated by inadequate record-keeping and emerging safety concerns.

Objectives

The authors of this study address the need for reliable implant identification methods by developing a deep learning model capable of classifying breast implants using ultrasound images.

Methods

Retrospective data of 28,712 breast ultrasound PNG files from 4136 breast implants in 2580 patients obtained from multiple institutions were utilized to train and validate this model.

Results

The findings demonstrate that the deep learning model achieved high diagnostic accuracy, with a balanced accuracy of 0.893 for manufacturer classification and 0.971 for implant texture classification in external test datasets. The model's performance was enhanced by employing Gradient-Weighted Class Activation Mapping (Grad-CAM) for interpretability.

Conclusions

By automating the identification process, this tool alleviates the reliance on specialized training among plastic surgeons regarding breast ultrasound, streamlining patient care. Despite limitations, the model shows promise for improving clinical workflows and patient outcomes.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningBreast ImplantsUltrasonography, MammaryBreast ImplantationBreastHumansFemaleRetrospective StudiesAdultMiddle AgedReproducibility of Results

Förderung

  • NAVER Digital Bio Innovation Research Fund
  • NAVER Corporation

Zitieren

APA

Yoo, H., Jeong, S., Kong, H. J., Cho, J., Chang, H., Kim, S., & Hong, K. Y. (2026). Deep Learning-Based Model for Breast Implant Classification in Ultrasonography: A Multi-Institutional Model Development and Validation Study. Aesthetic Surgery Journal, 46(2), 137-149. https://doi.org/10.1093/asj/sjaf220

Vancouver

Yoo H, Jeong S, Kong HJ, Cho J, Chang H, Kim S, et al. Deep Learning-Based Model for Breast Implant Classification in Ultrasonography: A Multi-Institutional Model Development and Validation Study. Aesthetic Surgery Journal. 2026;46(2):137-149. doi:10.1093/asj/sjaf220

BibTeX
@article{Yoo2026,
  author = {Yoo, Hyokyung and Jeong, Seoi and Kong, Hyoun-Joong and Cho, Jeongmok and Chang, Hak and Kim, Sungwan and Hong, Ki Yong},
  title = {Deep Learning-Based Model for Breast Implant Classification in Ultrasonography: A Multi-Institutional Model Development and Validation Study},
  journal = {Aesthetic Surgery Journal},
  year = {2026},
  volume = {46},
  number = {2},
  pages = {137--149},
  doi = {10.1093/asj/sjaf220}
}
Kostenlos starten →