AI Improves Breast Cancer Detection in Mammograms, Study Shows
A recent study conducted at the Karolinska Institutet in Stockholm has revealed that artificial intelligence (AI) can significantly enhance the detection of breast cancer in mammograms compared to traditional methods involving two radiologists. The ScreenTrustCAD study, which involved over 55,500 women between the ages of 40 and 74, demonstrated that a single radiologist supported by AI software could detect more cancer cases, improve efficiency, and reduce the rate of false positives, when compared to the conventional dual-radiologist approach.
Traditionally, breast cancer screening through mammography involves two radiologists examining each exam. In this study, the researchers compared the conventional method to a new approach where a single radiologist is assisted by AI technology. The findings revealed that the conventional dual-radiologist approach detected 250 cancers, whereas a single radiologist aided by AI detected 261 cases, and AI alone identified 246 cases. The study concluded that AI’s performance was statistically non-inferior to the dual-radiologist method.
One of the primary advantages of AI in breast cancer detection is its ability to reduce image-reading time for radiologists by half and decrease the occurrence of false positives. This leads to less unnecessary distress and cost for patients. The study’s first author, Karin Dembrower, emphasized that the collaboration between AI and humans provides a synergistic effect that enhances the chances of detecting cancer. Fredrik Strand, the study’s principal investigator, stated that for screening mammography, one AI-supported radiologist is a better alternative than two radiologists without AI.
The implications of this study are significant as it supports the integration of AI into breast cancer screening. AI has the potential to revolutionize the way mammograms are conducted, making the process faster, more accurate, and potentially more accessible to patients. However, experts caution that the chosen AI system must undergo rigorous testing on images from the same type of mammography equipment and continuous monitoring after implementation.
To further validate the effectiveness of AI in breast cancer detection, AI-supported radiologists have been assessing screening mammograms at Capio St Göran’s Hospital since June 2023. This implementation allows the medical staff to focus more on cancer patients, freeing up their time with the support of AI.
The ScreenTrustCAD study adds to a growing body of evidence supporting the use of AI in healthcare, particularly in cancer detection. The improved efficiency and effectiveness offered by AI have the potential to have a transformative impact on breast cancer screening. Not only can it speed up the process, but it can also improve accuracy, leading to better outcomes for patients.
In conclusion, the study conducted at the Karolinska Institutet highlights the value of AI in breast cancer detection. The findings demonstrate that AI-supported radiologists can detect more cases of cancer, enhance efficiency, and decrease false positives. With further testing and continuous monitoring, AI has the potential to enhance breast cancer screenings, benefiting patients worldwide.
