AI can detect breast cancer 20% more accurately

The study is the first randomized controlled trial to look at the use of AI in breast cancer screening, Politico reported. The results of the study, published in the Lancet Oncology on August 1, showed that using AI assistance and a single radiologist to analyze mammograms was as effective as using two radiologists without AI assistance. AI improved cancer detection accuracy by 20%.

AI could significantly reduce the workload of radiologists, who currently spend nearly 44% of their time analyzing mammograms.

The trial is still underway in Sweden, involving about 80,000 women. Half of the participants had their mammograms analyzed by two radiologists without the help of AI, while the other half had their mammograms analyzed by AI and one radiologist.

The study’s lead author, Kristina Lång of Lund University in Sweden, said the interim safety results were not yet sufficient to confirm that AI was ready for deployment in mammography screening. Lång and her colleagues are awaiting further results from the trial.

“The biggest potential of AI today is that it can reduce the burden on radiologists,” she said, adding that AI could help one doctor analyze mammograms so that other doctors can help more patients.

Commenting on the study, Stephen Duffy, professor of cancer screening at Queen Mary University of London, said that reducing the time burden on radiologists was “a significant issue in many breast cancer screening programmes”. But he was concerned that the use of AI could over-detect harmless lesions.

The new study is one of the first to use AI in radiology, according to the scientists. An editorial in the journal European Radiology in January argued that randomized controlled trials are essential to monitor the safety of AI — the system can make “unpredictable, undetectable errors that cannot be explained by human logic.”

According to Tin Tuc Newspaper