AI Surpasses Human Level in Breast Cancer Detection
Google Health researchers reported that AI has shown impressive results in detecting breast cancer during screening tests on mammograms, surpassing accuracy level of human doctors, reads the article in Nature:
“Here we present an artificial intelligence (AI) system that is capable of surpassing human experts in breast cancer prediction”
To determine if Google’s AI system can “spot the signs of breast cancer more accurately”, the system was trained on data from more than 28 000 women in the US and UK, thanks to the partnership with Cancer Research U.K. Imperial Centre and Royal Surrey County Hospital.
After the training, AI system has been able to significantly reduce false negative and false positive results, or errors in mammogram analysis, giving the researchers enough hope to suggest proceeding to clinical testing of the system.
“This robust assessment of the AI system paves the way for clinical trials to improve the accuracy and efficiency of breast cancer screening”
Researchers have also highlighted that in order for AI to further improve its accuracy rates, it needs to be trained on more data, suggesting to think of data as a precious resource to mine future positive medical outcomes. Earlier breast cancer detection significantly improves patient chances for successful treatment, and even a little improvement in reducing diagnosing can save lots of life.
As Future Time previously reported, Brandon Fornwalt from Geisinger and his colleagues have also revealed that their AI Model can predict patient death within a year better than practicing cardiologists, relying only on electrocardiogram (ECG) results.