Google AI Platform Beats Experts in Spotting Breast Cancer
According to a recent publication in Nature magazine, Google’s UK-based AI subsidiary DeepMind bested experts in spotting breast cancer in de-identified screening mammograms, providing fewer false positives and false negatives compared to results demonstrated by human professionals.
DeepMind team jointly with Cancer Research UK Imperial Centre, Northwestern University and Royal Surrey Country Hospital trained its AI-powered cancer detection model by feeding it scanned data from more than 90,000 women in the UK and the US. Notably, unlike human doctors who were able to analyze patient stories and prior mammograms, the AI was relying solely on X-ray images but still demonstrated better diagnostic accuracy.
The AI training was based on deep learning algorithms developed by Google to help doctors spot cancer and other diseases more quickly and accurately in pathology slides.
“Looking forward to future applications, there are some promising signs that the model could potentially increase the accuracy and efficiency of screening programs, as well as reduce wait times and stress for patients,” – said Shravya Shetty, technical lead at Google Health, in a blog post describing the mammography findings, also adding that “getting there will require continued research, prospective clinical studies and regulatory approval to understand and prove how software systems inspired by this research could improve patient care”.