Incredible but true! An artificial intelligence developed by the famous search engine has achieved the feat of reducing the percentage of errors in the diagnosis of breast cancer. From mammography images, AI has indeed been shown to be more precise than specialists, radiologist experts, which promises to provide invaluable help in the early detection of the disease.
The breast cancer, one of the most common and the second type of cancer the most deadly in women needs to be diagnosed in time to be treated at best. And if until now, artificial intelligence (AI) was a somewhat vague term that had little or no impact on our daily lives, it could well become a major tool to avoid misdiagnosis. Indeed, a computer program developed by Google has shown better accuracy than radiologist experts to identify breast cancer.
Published in the scientific journal Nature , the study was based on an artificial intelligence technique developed by Google, which was based on an algorithm composed of almost 29,000 images of mammograms from Great Britain and the States -United. While specialists had access to the patient’s history when interpreting x – rays, the RN only had access to the last mammogram. The results then showed that artificial intelligence was more precise, reducing the proportion of cases where cancer was incorrectly detected: -5.7% for American cases and -1.2% for the British. But AI also decreased the percentage of missed diagnoses: -9.4% among American images and also -2.7% among those from Great Britain.
If a human diagnosis remains and will remain the primary means of screening for breast cancer, AI could act as a “second opinion”. In the United States, for example, only one reading of screening images is performed. In the UK or France, mammograms offered to women are reviewed by two radiologists. Such a device could therefore make it possible to reduce the workload of the second radiologist, while promising a more precise diagnosis. “Further trials, clinical validation and regulatory approvals are needed before this can start to make a difference for patients, but we are committed to working with our partners to achieve this goal.“, explains Dr Dominic King, co-author of this innovative study.
Image source: Getty Images