Use of AI in mammogram cancer screening can cut radiologist workloads in half

Out of the total, 39,996 patients were randomly assigned AI-powered screenings, from which 28 percent returned a positive sighting. (Representational Image)

A new study has found that AI can detect cancer on par with two radiologists working together.

The study published in the Lancet Oncology Journal said that AI-supported mammography screenings resulted in a detection rate similar to a, "standard double screening".

Also read | OpenAI's GPT-3 performs as well as undergraduates in reasoning tests

A research team from Lund University in Sweden, followed 80,033 Swedish women, with an average age of 54, for a year from 2021 to 2022.

Out of the total, 39,996 patients were randomly assigned AI-powered screenings, from which 28 percent returned a positive sighting. Of the patients that used conventional methods, 25 percent reported positive.

Of the 41 total cancers detected by AI, 19 were invasive. Both the AI screenings and conventional methods had a 1.5 percent false positive rate. What made it interesting is that radiologists using AI ended up looking at fewer readings overall, reducing the "screen-reading workload".

Also read | Meta releases new AI tool that creates music from prompts

Co-author of the report, Dr Kristina Lang, said that the "promising interim safety results should be used to inform new trials and program-based evaluations to address the pronounced radiologist shortage in many countries", but were not enough to confirm that AI was ready to be implemented in medical screenings.

Lang said they "still need to understand the implications on patients’ outcomes, especially whether combining radiologists’ expertise with AI can help detect interval cancers that are often missed by traditional screening, as well as the cost-effectiveness of the technology".