Artificial Intelligence Could Halve Breast Cancer Scan Reading Time

(Patria) - Artificial intelligence (AI) could help nearly halve the workload of radiologists when it comes to searching for signs of breast cancer in routine scans, a Swedish study has shown.

Interim results from the trial have been hailed as promising, but the authors cautioned that more research is needed before AI can be used for breast cancer screening on a wider scale, AFP reported.

While increasingly convincing chatbots such as ChatGPT have sparked speculation about future potential applications of AI, one area where the technology has already shown expertise is reading medical scans. With many countries suffering from a shortage of radiologists, there are hopes that artificial intelligence could make the time-consuming job of analyzing routine scans faster and more accurate.

This could have a particularly large impact on breast cancer. More than 2.3 million women were diagnosed with cancer in 2020 alone, according to the World Health Organization, and it caused 685,000 deaths. Regular screening is vital for identifying early signs of cancer. In Europe, women aged 50 to 69 are advised to have a mammogram every two years, and the resulting image is analyzed by two radiologists.

The study in Sweden involved scanning 80,000 women who had mammograms at four locations in southwestern Sweden between April 2021 and July last year.

Their scans were randomly divided for analysis by an AI-supported system or two human radiologists, who served as a control group. The AI algorithm read the images and predicted the risk of cancer. Its predictions were then checked by a radiologist. The AI-supported system detected 20 percent more cancers, the study says, which amounted to an additional case per thousand women screened.

When it came to false positive results, the AI-supported system and the two humans detected the same rate: 1.5 percent. The radiologists' workload was reduced by 44 percent for the AI group, because only one person was needed to read the scans instead of two.

"The greatest potential of AI right now is that it could allow radiologists to be less burdened by an excessive amount of reading," said Kristina Lang, a radiologist at Sweden's Lund University and lead author of the study.

But Lang said that "promising interim safety results" "are not enough on their own to confirm that AI is ready for implementation in mammography screening."

It will be another two years before the trial can say whether using AI leads to a reduction in so-called interval cancers, which are detected between routine screenings, the researchers cautioned.