Cancer screening with artificial intelligence could halve radiologists' workload - study

The use of artificial intelligence in mammography-based cancer screening could safely reduce the radiologist's workload by almost half without risking an increase in false positives.

Here's What We Know

The study, conducted by a team of scientists at Lund University, Sweden, involved follow-up of 80,033 women for just over one year. Of the 39,996 patients who were randomly assigned to breast cancer screenings using AI, 28% were found to have cancerous tumours.

Of the remaining 40,024 patients who underwent conventional screening, cancer was detected in only 25% of cases.

The false-positive rate for both AI and conventional screening was 1.5 per cent. Crucially, radiologists working with AI had to review 36,886 fewer screenings than their colleagues. This reduced their workload by 44%.

"AI-supported mammography screening resulted in a similar cancer detection rate compared with standard double reading, with a substantially lower screen-reading workload, indicating that the use of AI in mammography screening is safe", the study concluded.

Source: Engadget