The invention relates to an early screening
system based on Alzheimer's
disease, which belongs to the technical field of early screening of Alzheimer's
disease and comprises an interactive sensing module, a behavior capture module, a
gait analysis module, a
feature extraction module, an intelligent analysis module and a feedback output module. The interactive
perception module is used for presenting and dynamically adjusting a
work memory backtracking task, a scene memory reconstruction task and a meta-memory task, and the behavior capture module is used for collecting
eye movement data of a user in the testing process. According to the method, behavior indexes and AD core biomarkers are clearly associated, so that the
screening result has a
neuropathology basis, the scientificity and interpretation of evaluation are enhanced, the
system has a self-optimization function, the model can be updated through continuous learning increment, and a visual three-dimensional
cognitive map and personalized intervention suggestions are output, so that the
system has a good application prospect. An efficient and quantifiable decision support tool is provided for
clinical diagnosis and health management, and finally non-invasive, low-cost and efficient early
risk prevention and control of the Alzheimer's
disease are achieved.