AI Mammograms Detect 20% More Breast Cancer Cases, Study Reveals
Artificial intelligence (AI) could play a key role in improving breast cancer detection rates, according to a recent study. Preliminary analysis of a long-term trial involving 80,000 women in Sweden has shown that AI readings of mammograms identified 20 percent more cases of breast cancer compared to the standard reading by two radiologists. This breakthrough has the potential to reduce patient waiting times and alleviate the pressure on radiologists amid a global workforce shortage.
Although it may take some time before mammograms are routinely read by AI, the study’s findings are being hailed as breathtaking by experts in the field. Integrating AI into screening procedures could lead to earlier identification of breast cancer, which in turn enhances the treatability of the disease. Breast cancer is the most prevalent cancer worldwide, making this potential improvement in detection a significant win for patients.
The study, the first randomized controlled trial to investigate the use of AI in mammography screening, recruited 80,020 women between the ages of 40 and 80 who had mammograms in Sweden between April 2021 and July 2022. Half of the participants had their mammograms read by a commercially available AI model and one or two radiologists, depending on their identified risk score. The other half had their mammograms assessed by two radiologists, considered the standard practice in Europe.
In addition to detecting breast cancer at a higher rate, AI-supported screenings did not lead to a higher number of false positives. The AI model provided radiologists with information from the initial screening to help them more accurately interpret mammograms. If a mammogram was flagged as suspicious, patients were asked to undergo further tests.
Given that early-stage breast cancers are increasingly treatable, improving detection rates is crucial. Breast cancer claimed the lives of at least 685,000 women worldwide in 2020, according to the World Health Organization. In the United States, the average woman has a 13 percent chance of developing breast cancer in her lifetime and roughly a 2.5 percent chance of dying from it.
The implementation of AI in breast cancer screenings could have a significant impact on the day-to-day work of healthcare professionals. James O’Connor, a professor of radiology at the Institute of Cancer Research in London, believes that widespread use of AI-supported screenings could save a tremendous amount of time and help address workflow shortages in the healthcare industry. However, challenges exist around implementation due to varying regulations regarding AI diagnostics across different jurisdictions, as well as patient acceptance of AI in medical care.
While AI has shown promise in assisting with medical diagnostics, including cancer identification, further training and testing of AI models are necessary before wider deployment in healthcare settings. Nonetheless, the study’s findings provide valuable insights into the potential benefits of AI in mammography screening. By leveraging machine learning technologies, healthcare professionals can aim to detect breast cancer earlier and more accurately, ultimately improving patient outcomes and reducing breast cancer mortality rates.
