AI in Breast Cancer Screening Improves Early Detection: Study

Experts believe AI could help address screening gaps in low-income countries. In India, where over 50% of breast cancer cases are diagnosed at advanced stages, AI-powered screening could significantly improve early detection.

Researchers found that integrating AI into screening processes helped radiologists identify cancers earlier, potentially improving patient outcomes.

A landmark study published in The Lancet Digital Health has found that artificial intelligence (AI)-supported mammography screening improved early breast cancer detection by 29% without increasing false positives. The trial, conducted as part of Sweden’s national screening program, is the world’s first randomized clinical trial to assess AI’s efficacy in breast cancer detection.

The Mammography Screening with Artificial Intelligence (MASAI) trial analyzed data from over 100,000 women screened between April 2021 and December 2022.

Half underwent AI-supported screening, while the rest followed standard procedures. AI-assisted screening detected 6.4 cancers per 1,000 screenings, compared to 5 per 1,000 in the control group. The method also increased the detection of in-situ cancers (stage 0) and small, invasive cancers that had not spread to lymph nodes.

“Use of AI in mammography screening has the potential to reduce screen-reading workload and increase cancer detection, which could affect patient outcomes,” the researchers noted.

The trial used the Transpara AI system by Netherlands-based ScreenPoint Medical to evaluate mammograms. AI assigned malignancy risk scores from 1 to 10, directing high-risk cases to double readings by radiologists. This approach identified more invasive and precancerous lesions while reducing unnecessary readings.

Experts believe AI could help address screening gaps in low-income countries. In India, where over 50% of breast cancer cases are diagnosed at advanced stages, AI-powered screening could significantly improve early detection.

“AI-powered imaging solutions, trained on vast datasets, can identify subtle patterns that may not be visible to the human eye,” said Kabir Mahajan, Chief Operating Officer at Mahajan Imaging & Labs.

The MASAI study suggests AI could be integrated into mobile mammography vans for wider access. Experts believe AI-driven risk assessment models, combined with genetic and lifestyle data, could further personalize screening strategies.

However, specialists stress that AI should complement, not replace, human expertise. Dr Rupa Ananthasivan, a radiology consultant at Manipal Hospitals, Bengaluru, noted, “AI is a powerful tool, but it must be trained on diverse populations and used as an aid, not a substitute, for radiologists.”

The findings mark a significant step in AI’s role in enhancing breast cancer detection, though experts emphasize the need for further research before full-scale implementation.

Stay tuned for more such updates on Digital Health News.