EUVIMEDEUVIMED entdecken
Open Access

A hybrid radiomics framework integrating genetic algorithm-optimized random forest for preoperative identification of Luminal B breast cancer and Ki-67 prediction: A multicenter study

Zimei Lin, Yi-Jie Chen, Yong-Yuan Xu, Qing Wen, Ling-Ling Chen, Li-Ming Shao, Xiao-Yan Niu, Li-Na Tang, Pintong Huang

The Ultrasound Journal · 20.05.2026 · 18 (1) · 18667

DOI: 10.5826/tuj.2026.18667 ↗

Kurz gefasst

Fachartikel in der Zeitschrift The Ultrasound Journal, online erschienen am 20.05.2026. Verfasst von Zimei Lin und 8 weiteren Autor:innen. Beteiligte Einrichtung: The Second Affiliated Hospital Zhejiang University School of Medicine. Die Arbeit nennt 23 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Preoperative identification of Luminal B breast cancer remains a clinical challenge. This study aimed to develop an ultrasound radiomics framework integrating tumoral and peritumoral information for preoperative identification of Luminal B subtype and prediction of Ki-67 status.

Methods

We retrospectively analyzed 1,944 patients from three centers. The development cohort from Centers One and Two was divided by stratified sampling into a training set (n = 1,434) and an internal test set (n = 253), and an independent cohort from Center Three (n = 257) was used for external validation. Lesion-containing ROIs were processed using deep learning-assisted segmentation and standardized for downstream analysis. Radiomic features were extracted, and a genetic algorithm (GA) was coupled with a random forest (RF) classifier to construct two models: one for Luminal B classification and another for predicting Ki-67 expression.

Results

The combined tumor-peritumoral model achieved the highest performance, with the Luminal B classifier showing AUCs of 0.876 (training), 0.693 (test), and 0.786 (external validation). The Ki-67 prediction model yielded AUCs of 0.890 (training) and 0.858 (test), though external validation (AUC=0.661) was limited by dataset distribution. The Delong test confirmed that combined ROIs significantly outperformed tumor-only models, with NRI and IDI tests further validating the added value of peritumoral features.

Conclusions

Ultrasound radiomics integrating tumoral and peritumoral regions can support the preoperative identification of Luminal B breast cancer, and peritumoral region analysis significantly enhances predictive performance. The framework also shows potential for predicting Ki-67 status within this subtype.

Autor:innen und Institutionen

  • The Second Affiliated Hospital Zhejiang University School of Medicine.
  • Fujian Cancer Hospital.
  • Cixi Sixth People's Hospital.
  • The Affiliated Hospital of Qingdao University.

Zitieren

APA

Lin, Z., Chen, Y. J., Xu, Y. Y., Wen, Q., Chen, L. L., Shao, L. M., Niu, X. Y., Tang, L. N., & Huang, P. (2026). A hybrid radiomics framework integrating genetic algorithm-optimized random forest for preoperative identification of Luminal B breast cancer and Ki-67 prediction: A multicenter study. The Ultrasound Journal, 18(1), 18667. https://doi.org/10.5826/tuj.2026.18667

Vancouver

Lin Z, Chen YJ, Xu YY, Wen Q, Chen LL, Shao LM, et al. A hybrid radiomics framework integrating genetic algorithm-optimized random forest for preoperative identification of Luminal B breast cancer and Ki-67 prediction: A multicenter study. The Ultrasound Journal. 2026;18(1):18667. doi:10.5826/tuj.2026.18667

BibTeX
@article{Lin2026,
  author = {Lin, Zimei and Chen, Yi-Jie and Xu, Yong-Yuan and Wen, Qing and Chen, Ling-Ling and Shao, Li-Ming and Niu, Xiao-Yan and Tang, Li-Na and Huang, Pintong},
  title = {A hybrid radiomics framework integrating genetic algorithm-optimized random forest for preoperative identification of Luminal B breast cancer and Ki-67 prediction: A multicenter study},
  journal = {The Ultrasound Journal},
  year = {2026},
  volume = {18},
  number = {1},
  pages = {18667},
  doi = {10.5826/tuj.2026.18667}
}
Kostenlos starten →