AI improves breast cancer detection rate by 20%

Artificial intelligence is able to accurately detect 20 percent more breast cancers from mammograms than traditional screening tests performed by radiologists, according to early results from a Swedish trial published overnight.

He study is the first randomized controlled trial to look at the use of AI in breast cancer detection and comes amid a dramatically changing landscape for the technology and how it is regulated.

The interim results, published in Lancet Oncology late Tuesday, found that using AI-supported mammography screening, along with one or two radiologists, was as good as using two radiologists without AI, and led to 20 detections. percent more cancers.

There was also a significant reduction in the workload for radiologists, as physicians had to spend 44 percent less time reading mammograms.

The trial, which is still ongoing, took place in Sweden and looked at more than 80,000 women. Half of the participants had their mammograms reviewed without AI by two radiologists, while the other half had their mammograms reviewed by AI and a radiographer, except in cases where AI generated the highest risk score, in which case two radiologists they evaluated the evaluation.

The results come amid a significant increase in interest around the potential opportunities and risks that AI poses for medicine and the world at large. AI is increasingly deployed in medical settings, but there are also concerns about how algorithms are trained and validated in these spaces, as well as the potential for bias and overdiagnosis.

Meanwhile, the EU is planning strict regulation around the use of AI and the European Medicines Agency is working on evaluating the risks and benefits technology for drug development.

more to do

Despite the positive findings, lead author Kristina Lång, from Lund University in Sweden, said the interim safety results “are not enough on their own to confirm that AI is ready for deployment in mammography.” Lång and her colleagues are awaiting further trial results indicating whether the use of AI reduces the number of cancers detected between screenings, and therefore whether it is worth implementing.

“The biggest potential for AI right now is that it could allow radiologists to be less overwhelmed by excessive amounts of reading,” Lång said. He said AI could eliminate the need for a second radiologist to read the mammogram, allowing radiologists to help more patients.

Commenting on the results, Stephen Duffy, professor of cancer screening at Queen Mary University of London, said reducing the time load on radiologists was “an issue of considerable importance in many breast screening programmes.” But he said there could be concerns that the use of AI could over-detect harmless injuries.

In addition to being the first randomized trial to look at the use of AI to help read mammograms, the study was one of the first to look at AI in the field of radiology. A editorial in European Radiology in January argued that randomized controlled trials were important to monitor the safety of AI systems, which can produce “unpredictable and undetectable errors, not explainable by human logic.”