FaceAge AI Predicts Cancer Survival with 81% Accuracy in US, Netherlands

Researchers at Mass General Brigham have developed FaceAge, an AI tool that estimates a person's biological age from facial photos to predict cancer outcomes with 81% accuracy, outperforming traditional clinical methods. FaceAge analyzes subtle facial features linked to physiological aging, providing more precise survival estimates and aiding treatment decisions, especially in palliative radiotherapy cases. However, concerns remain about its applicability across diverse populations, as the model was primarily trained on white faces, and its performance on people of color or those with facial alterations like plastic surgery is still uncertain. Separately, a Weill Cornell Medicine-led study introduced a machine learning method that groups cancer patients based on shared traits to better predict treatment responses, enhancing personalized therapy and clinical trial selection. Additionally, University College Dublin researchers are developing AI models to analyze genetic alterations in childhood cancers to improve drug treatments. Together, these advancements showcase AI's growing role in refining cancer prognosis and personalized treatment strategies, though challenges in inclusivity and validation remain.