Researchers at Mass General Brigham built an AI tool that assigns cancer patients survival-risk scores from a facial photograph, The Straits Times reports. The article, credited to Bloomberg, says patients with the highest scores were more likely to die within 90 days. That finding distinguishes risk groups; it does not give a patient an exact lifespan.
According to the report, the AI was trained on more than 40 million face images and photographs of more than 30,000 cancer patients. It estimated age with a margin of error of about five years and sorted patients into higher- and lower-risk groups that subsequently had different risks of death. The report says the research was published Oct. 19 in The Lancet Digital Health.
Raymond Mak, a radiation oncologist at Mass General Brigham Cancer Institute and a study author, described the tool as “a way for doctors to have one more piece of information.” That is a proposed use, not a demonstrated improvement in treatment decisions.
Miroslaw Bober, a University of Surrey professor who was not involved in the research, said the tool is not ready for clinical use. “Predicting an outcome is not the same as explaining it,” he said. Bober said the study had not shown that its predictions add clinically useful information beyond doctors’ assessments, which the study did not measure. He called for testing against doctors’ judgment and across different hospitals and patient groups.
Comments