Opportunistic Breast Cancer Risk Stratification From Low-dose Chest CT Using Multiple Instance Learning

Ni Y, Panambur AB, Liu C, Nguyen TT, Bayer S, Juan H, Jiayu S, Su L, Maier A (2026)

Publication Type: Conference contribution

Publication year: 2026

Journal

Informatik aktuell Gesellschaft für Informatik (GI)

Series: Informatik aktuell

Pages Range: 427-434

Conference Proceedings Title: Bildverarbeitung für die Medizin 2026

Event location: Lübeck DE

ISBN: 9783658510992

DOI: 10.1007/978-3-658-51100-5_84

Abstract

Breast cancer screening in Asian populations faces significant challenges due to high breast density prevalence, which reduces mammographic sensitivity. Low-dose computed tomography (LDCT) scans acquired for lung cancer screening capture breast tissue and present an opportunity for opportunistic breast cancer risk assessment. This study develops and evaluates deep learning frameworks using multiple instance learning (MIL) for breast cancer risk stratification from LDCT scans combined with clinical features. Two complementary approaches were developed: an individual breast model using attention mechanisms for slice-level interpretability, and a bilateral model employing global pooling to capture asymmetry patterns. Evaluated on 60 patients, the individual model achieved 72.7% accuracy with AUC 0.85, while the bilateral model achieved 75.0% accuracy with AUC 0.87. Both models successfully stratified patients into medium-risk (BI-RADS 3–4) and high-risk (BI-RADS 5–6) categories, providing interpretable outputs through attention maps and asymmetry visualizations. This represents the first comprehensive, proof-of-concept study demonstrating the feasibility of automated breast cancer risk assessment from opportunistic LDCT imaging.

Authors with CRIS profile

Chang Liu Lehrstuhl für Informatik 5 (Mustererkennung) Tri-Thien Nguyen Lehrstuhl für Informatik 5 (Mustererkennung) Siming Bayer Lehrstuhl für Informatik 5 (Mustererkennung) Andreas Maier Lehrstuhl für Informatik 5 (Mustererkennung)

Involved external institutions

West China Medical Center of Sichuan University / 四川大学华西医学中心 CN China (CN)

How to cite

APA:

Ni, Y., Panambur, A.B., Liu, C., Nguyen, T.-T., Bayer, S., Juan, H.,... Maier, A. (2026). Opportunistic Breast Cancer Risk Stratification From Low-dose Chest CT Using Multiple Instance Learning. In Bildverarbeitung für die Medizin 2026 (pp. 427-434). Lübeck, DE.

MLA:

Ni, Yaqiong, et al. "Opportunistic Breast Cancer Risk Stratification From Low-dose Chest CT Using Multiple Instance Learning." Proceedings of the Bildverarbeitung für die Medizin 2026 (BVM 2026), Lübeck 2026. 427-434.

BibTeX: Download