A research team from KAIST, Sungkyunkwan University, and Korea University Anam Hospital analyzed real-world lifelog data from 1,224 older adults, identifying imminent diagnostic risk of cerebrovascular disease with 96.5% accuracy. The study, published in npj Digital Medicine with KAIST Dr. Jeongyeop Baek as the first author, was announced by KAIST on the 12th of July. The team, led by Professor Lisa Lim from KAIST, developed an AI framework that uses long-term lifelog data collected in homes to identify the prodromal phase of cerebrovascular disease. The study was based on lifelog data from 1,224 older adults collected by LivOn Care Co., Ltd., and the team analyzed a total of 13,362 two-week lifelog samples. The AI technology identifies risk stages by analyzing daily activity, sleep, circadian rhythm, indoor environmental information, age, and chronic disease data. The team also succeeded in assessing whether a cerebrovascular disease diagnosis was approaching by analyzing changes in lifestyle patterns over time, with lifelog data from within four weeks before diagnosis classified.