Multimodal feature fusion model for breast mass malignant risk stratification - Takeaways - MDSpire

Multimodal feature fusion model for breast mass malignant risk stratification

  • By

  • Shengxin Pei

  • Xiumei Tang

  • Hongxia Su

  • Jingyan Liu

  • Zihan Lan

  • Siyu Wang

  • Yulan Peng

  • June 3, 2026

  • 0 min

Share

  • 1

    This study developed machine learning models integrating multimodal features to enhance breast mass malignancy risk stratification.

  • 2

    The dataset included 2,685 patients with 3,703 ultrasound images, comprising 2,069 benign and 616 malignant cases.

  • 3

    The Random Forest model using combined multimodal features achieved the highest AUC of 0.850 for overall performance.

  • 4

    Performance was excellent for BI-RADS categories 2 and 3, but limitations were noted in higher-risk categories 4b and 4c.

  • 5

    The findings suggest potential clinical utility for reducing unnecessary biopsies and improving diagnostic confidence in breast cancer.

Original Source(s)

Multimodal feature fusion model for breast mass malignant risk stratification

  • by Shengxin Pei, Xiumei Tang, Hongxia Su, Jingyan Liu, Zihan Lan, Siyu Wang, Yulan Peng

  • June 3, 2026

Related Content

Statin Use and Survival in Early Breast Cancer According to Different Intrinsic Subtypes

A case report of thyroid-like low-grade nasopharyngeal papillary adenocarcinoma with literature review

Current status and trends of immune-related adverse events in lung cancer treated with immune checkpoint inhibitors: a bibliometric analysis of the past decade (2016–2025)