647 Genome-wide scans identified genetic variants associated with facial aging traits quantified by deep learning methods

Abstract

As life expectancy and aging population increase worldwide, facial aging is becoming progressively important. The occurrence and progression of facial aging are strongly influenced by genetic factors. Previous studies have identified a number of genetic variants associated with facial aging phenotypes using traditional image analysis and human assessments. In this study, we developed a deep learning framework to precisely measure 4 facial aging phenotypes (lacrimal sulcus, pigmentation spots, wrinkle forehead, and wrinkle under eyes).