AI-Supported Mammography Screening Noninferior to Double Reading for Interval Cancer Rate

FRIDAY, Feb. 6, 2026 (HealthDay News) — Artificial intelligence (AI)-supported mammography screening shows favorable outcomes versus standard double reading, with a noninferior interval cancer rate, according to a study published online Jan. 31 in The Lancet.

Jessie Gommers, from Radboud University Medical Centre in the Netherlands, and colleagues conducted a randomized population-based screening accuracy trial to compare the interval cancer rate in AI-supported mammography screening to standard double reading without AI. A total of 105,934 women were randomly assigned to the intervention or control group in a 1:1 ratio; 19 women were excluded from the analysis.

The researchers found that the interval cancer rates were 1.55 and 1.76 per 1,000 participants in the intervention and control groups, respectively, for a noninferior proportion ratio of 0.88 (95 percent confidence interval, 0.65 to 1.18; P = 0.41). Compared with the control group, the intervention group had fewer interval cancers that were invasive (75 versus 89), T2+ (38 versus 48), or non-luminal A (43 versus 59). Higher sensitivity was seen in the intervention group versus the control group (80.5 versus 73.8 percent), which was consistent across age and breast density, and for invasive cancer but not in-situ cancer. For both groups, specificity was 98.5 percent.

“Our results potentially justify using AI to ease the substantial pressure on radiologists” workloads, enabling these experts to focus on other clinical tasks, which might shorten the waiting times for patients,” Gommers said in a statement.

Two authors disclosed ties to the pharmaceutical and medical technology industries.

Abstract/Full Text (subscription or payment may be required)