Can Adjunctive AI Enhance DBT Detection of Invasive Lobular Cancer?

Invasive lobular carcinoma (ILC) comprises approximately 10 to 15 percent of breast cancer cases but it is “disproportionately represented” in late-stage cancer presentations, according to Leslie R. Lamb, MD, MSc, an assistant professor of radiology at Harvard Medical School.

“ … (Invasive lobular cancer) can be particularly challenging to detect on screening mammography, even with tomosynthesis, because of its infiltrative growth pattern and subtle imaging appearance. So rather than presenting as a discrete mass, it often presents as a subtle area of architectural distortion or asymmetry, or sometimes there's no visible abnormality at all,” pointed out Dr. Lamb during a recent interview with Diagnostic Imaging.

Accordingly, Dr. Lamb and colleagues recently evaluated use of an artificial intelligence (AI) software (Genius AI Detection 2.0, Hologic) in 224 women diagnosed with ILC who had digital breast tomosynthesis (DBT).

In findings presented at the recent Society of Breast Imaging (SBI) Symposium, the researchers found that unassisted radiologists had a sensitivity rate of 82.1 percent while the use of adjunctive AI demonstrated a sensitivity of 89.3 percent. In addition to identifying and localizing 88.6 percent of true-positive cases of ILC, Dr. Lamb noted that adjunctive AI detected and localized ILC in 16 out of 40 false-negative cases.

“The key finding was that the AI model demonstrated high sensitivity for ILC, and it identified a meaningful proportion of cancers missed by radiologists,” noted Dr. Lamb, who is affiliated with the Department of Radiology at Massachusetts General Hospital.

Noting the need for prospective evaluation of the AI software for ILC detection in real-world clinical practice, Dr. Lamb suggested that future studies may evaluate the impact of the software on recall rates, false positives and patient outcomes.

“ … Another kind of sobering thought was there's still a small subset of cancers, about 10 percent that are missed by both radiologists and AI. I think this really highlights both the challenge of ILC and the fact that there's still room for improvement,” added Dr. Lamb.

(Editor’s note: For related content, see “Emerging Insights on the Use of FES PET for Women with Lobular Breast Cancer,” “Moving Beyond Mammography for Screening and Staging of Invasive Lobular Carcinoma” and “Large Mammography Study Shows AI-Aided DBT Bolsters Cancer Detection Without Increasing Recall Rate.”)

Reference

1. Lamb LR. Artificial intelligence for the detection of invasive lobular carcinoma on screening digital breast tomosynthesis. Presented at the Society of Breast Imaging Symposium, April 16-19, 2026, Seattle. https://2026-sbi-symposium.eventscribe.net/index.asp