The use of Artificial Intelligence (AI) can significantly improve the early detection of breast cancer, according to a study by the Radiological Society of North America (RSNA). The study found that the cancer detection rate increased by 21% with the use of AI. AI was used to examine mammography screening images for suspicious areas, complementing the work of medical professionals.
AI in mammography is suitable as a “second pair of eyes” to support radiologists, as stated in an RSNA press release. The study was presented at their annual meeting. AI can be used as a decision aid and for risk prediction, offering several advantages. When the radiologist and AI disagreed, an additional radiologist was involved to review the images.
Due to AI detections, more women were called in for further examinations, confirming suspicions in many cases and allowing treatments to start quickly. A British study earlier this year also concluded that AI detects more tumors in breast cancer screenings.
The study included mammography screenings from 747,604 women in the US over 12 months. All ten participating clinical practices reported a significantly higher cancer detection rate. Women had to pay for the AI-assisted breast cancer screening themselves, as insurance companies did not cover the costs.
Researchers are hoping for further studies. “These data show that many women want to use AI to improve their screening mammography, and when AI is coupled with a safety review, more cancers are detected,” said lead author Dr. Gregory Sorensen in the press release. Overall, 36% of women with access to AI-assisted screening chose to use it.
The researchers aim to further investigate the benefits of AI in a subsequent study to obtain more reliable data.