The results from the Mammography Screening with Artificial Intelligence (MASAI) study, recently published on The Lancet, show that AI-supported mammography screening has more favorable outcomes than standard double screening on interval cancer rate and interval cancers with unfavorable characteristics.
The MASAI trial was a randomized, controlled, non-inferiority, single-blinded, population-based screening accuracy trial embedded within the Swedish breast cancer screening program where participants were allocated to either AI-supported mammography screening or standard double reading without AI. A total of 105.934 women were randomly assigned to the intervention or control group; in the AI-supported group, examinations with low AI risk scores were triaged to single reading, while those with a high AI risk score underwent double reading, with AI used as detection support for radiologists. Interval cancer rates were 1.55 and 1.76 per 1000 participants in the AI-intervention and control group respectively. AI-supported screening group had fewer interval cancers that were invasive, T2+, or non-luminal A than the control group; sensitivity was higher consistently across age and breast density, and for invasive cancer but not for in-situ cancer. Authors concluded that «AI-supported mammography screening showed consistently favorable outcomes compared with standard double reading, with a non-inferior interval cancer rate, fewer interval cancers with unfavorable characteristics, higher sensitivity, and the same specificity, while also reducing screen reading workload. These findings imply that AI-supported mammography screening can efficiently improve screening performance compared with standard double reading and may be considered for implementation in clinical practice».
AI versus standard double reading for mammography screening
Findings from the MASAI Swedish study show consistently more favorable outcomes with AI-supported mammography screening versus standard double reading
The results from the Mammography Screening with Artificial Intelligence (MASAI) study, recently published on The Lancet, show that AI-supported mammography screening has more favorable outcomes than standard double screening on interval cancer rate and interval cancers with unfavorable characteristics.
The MASAI trial was a randomized, controlled, non-inferiority, single-blinded, population-based screening accuracy trial embedded within the Swedish breast cancer screening program where participants were allocated to either AI-supported mammography screening or standard double reading without AI. A total of 105.934 women were randomly assigned to the intervention or control group; in the AI-supported group, examinations with low AI risk scores were triaged to single reading, while those with a high AI risk score underwent double reading, with AI used as detection support for radiologists. Interval cancer rates were 1.55 and 1.76 per 1000 participants in the AI-intervention and control group respectively. AI-supported screening group had fewer interval cancers that were invasive, T2+, or non-luminal A than the control group; sensitivity was higher consistently across age and breast density, and for invasive cancer but not for in-situ cancer. Authors concluded that «AI-supported mammography screening showed consistently favorable outcomes compared with standard double reading, with a non-inferior interval cancer rate, fewer interval cancers with unfavorable characteristics, higher sensitivity, and the same specificity, while also reducing screen reading workload. These findings imply that AI-supported mammography screening can efficiently improve screening performance compared with standard double reading and may be considered for implementation in clinical practice».