In the study, the first randomised controlled trial to investigate the use of AI in mammography screening, 80,033 Swedish women aged 40-80 years who had undergone mammogram were randomly assigned in a 1:1 ratio to either AI-supported analysis, where a commercially available AI-supported system analysed the mammograms before they were also read by one or two radiologists, or standard analysis by two radiologists between April 2021 and July 2022.
Researchers found that AI-supported screening resulted in a cancer detection rate of six per 1,000 screened women compared to five per 1,000 for standard double reading without AI. Also, the use of AI did not increase false positives and reduced the mammogram reading workload by 44%.
“While our AI-supported system requires at least one radiologist in charge of detection, it could potentially do away with the need for double reading of most of the mammograms,” Dr Kristina Lång, lead author of the study, said.
AI has shown encouraging results in retrospective studies using the technology to triage examinations to either single or double reading and by providing radiologists with computer-aided detection marks highlighting suspicious features to reduce false negative results. But, the new study has for the first time provided robust evidence.
Artificial Intelligence predicts the outcome of a diagnostic or prognostic process by using its experience that is based on data of previous patients that were used to train the AI system. It may be considered as a super-calculator that reduces our error rate.
“AI can extract information from natural language processing, helping detect patterns and information that would otherwise be hidden in electronic data. The future of AI would involve all healthcare fields etc,” Dr Prateek Sharma, who was the first chair of US task force of AI, told TOI in a recent interview.