A Swedish study has shown that artificial intelligence (AI) can safely detect breast cancer in mammograms – X-rays of the breast.
Researchers put ScreenPoint’s Transpara AI tool head-to-head with the current standard of two expert readers and found that computer-aided detection could spot cancer at a similar rate.
Over 80,000 people from Sweden attended screening clinics and were split into two groups.
Half of the scans were assessed by two radiologists, and the other half by the AI-supported screening tool.
In total, the AI-supported screening tool detected cancer in 244 people, compared to 203 people recalled from standard screening.
In both the AI-supported screening tool group and the standard screening group, the false-positive rate was 1.5%.
Despite the authors of the study noting that the results were not enough to confirm that AI should be used routinely in mammography screening, its use did result in a significant reduction in the workload that is currently faced in screening centres.
The AI technology, which almost halved the screen time of standard care scan assessments, could benefit healthcare systems like the NHS, which is currently experiencing workforce issues.
Lead author of the study, Dr Kristina Lang, from Lund University in Sweden, said: “While our AI-supported screening system requires at least one radiologist in charge of detection, it could potentially do away with the need for double reading of the majority of mammograms easing the pressure on workloads and enabling radiologists to focus on more advanced diagnostics while shortening waiting times for patients.”
Recent figures from the Royal College of Radiologists have revealed that the NHS is facing a 29% shortfall in clinical radiologists, increasing waiting times and worsening health outcomes.
Dr Katharine Halliday, president of the Royal College of Radiologists, said: “While real-life clinical radiologists are essential and irreplaceable, a clinical radiologist with the data, insight and accuracy of AI will increasingly be a formidable force in patient care.”
The primary outcome results from the study are not expected for several years and will establish whether AI reduces interval cancers – cancers diagnosed between screenings – in 100,000 people with at least two years follow-up, and ultimately whether AI’s use in mammography screening is justified.