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

What this paper does

Low-dose chest CT is already acquired at scale for lung screening, and it happens to include breast tissue. This proof-of-concept study applies multiple instance learning to that existing data for opportunistic breast cancer risk stratification, stratified against BI-RADS 3–6 categories on a cohort of 60 patients. The individual-breast model reached 72.7% accuracy with AUC 0.85, and the bilateral model 75.0% accuracy with AUC 0.87. The authors describe it as the first comprehensive proof-of-concept of its kind. I co-supervised the master's thesis this paper grew out of and co-authored the result.

How

Multiple instance learning applied to low-dose chest CT volumes, with individual-breast and bilateral model variants, evaluated against BI-RADS 3–6 risk categories.

Reported results

• Individual-breast model: 72.7% accuracy, AUC 0.85.
• Bilateral model: 75.0% accuracy, AUC 0.87.
• Cohort of 60 patients stratified into BI-RADS 3–6 risk categories.
• Described by the authors as the first comprehensive proof-of-concept study of its kind.

Relevance

Opportunistic screening extracts preventive value from imaging that has already been acquired and paid for, which is a rare thing in radiology AI.

What it does not show

A 60-patient proof-of-concept. These numbers establish feasibility, not clinical performance, and would need a substantially larger multi-centre cohort to mean anything more.

Citation