Scientists in the United States have created an artificial intelligence system designed to assist physicians in assessing patient health status through facial photography analysis. The tool, trained on millions of images of healthy individuals and 35,000 photographs of cancer patients, examines specific facial regions to generate survival estimates.
The ‘FaceSurvival’ AI tool identifies the area beneath the eyes and around the bridge of the nose as particularly significant when evaluating prognosis. According to the research team, these metrics demonstrated applicability across multiple cancer types of varying severity, including cases where the disease had metastasized to the bones or brain. The study population consisted primarily of patients in their 50s and 60s, though the dataset included younger and older participants as well, with patients recruited from medical centers in the US and the Netherlands.
Lead researcher Dr Raymond Mak from Mass General Brigham explained that the goal is to provide physicians with more reliable assessment methods. He noted that as populations age, doctors require improved tools for evaluating physical reserve and frailty before recommending intensive interventions such as radiation therapy, surgical procedures, or systemic treatments. The AI system is intended to complement rather than replace clinical judgment, with a patent application currently pending.
The research team is also developing a companion tool called ‘FaceAge’ to predict biological age based on facial appearance. This system examines facial characteristics including skin texture below the nose, forehead lines, crow’s feet, and cheek appearance. According to researchers, the tool demonstrated consistent performance across variations in photograph quality, lighting conditions, facial expressions, and makeup application, with only minor changes observed following rhinoplasty procedures.
Experts caution that both tools remain in early development stages and require substantial additional research before clinical implementation. Researchers acknowledge the need for expanded datasets representing diverse ethnic populations to validate the systems’ effectiveness. A professor specializing in machine learning at the University of Surrey emphasized that any claims regarding the tool’s ability to track aging or lifestyle changes would be premature, and that the system requires comparison against physician assessment and testing across multiple healthcare settings before clinical use can be considered appropriate.
Article Attribution | Read More at Article Source
Article summary produced by Claude AI