摘要:SummaryAlzheimer's disease (AD) is a worldwide burden. Diagnosis is complicated by the fact that AD is asymptomatic at an early stage. Studies using AD-modeled animals offer important and useful insights. Here, we classified mice with a high risk of AD at a preclinical stage by using only their behaviors. Wild-type and knock-in AD-modeled (AppNL-G-F/NL-G-F) mice were raised, and their cognitive behaviors were assessed in an automated monitoring system. The classification utilized a machine learning method, i.e., a deep neural network, together with optimized stepwise feature selection and cross-validation. The AD risk could be identified on the basis of compulsive and learning behaviors (89.3% ± 9.8% accuracy) shown by AD-modeled mice in the early age (i.e., 8–12 months old) when the AD symptomatic cognitions were relatively underdeveloped. This finding reveals the advantage of machine learning in unveiling the importance of compulsive and learning behaviors for early AD diagnosis in mice.Graphical abstractDisplay OmittedHighlights•Cognitive-related behaviors were monitored to identify Alzheimer's disease (AD) mice•Those behaviors were used as inputs of the deep neural network algorithm•Best-performing inputs were related to age-specific mild decreases of cognitions•Despite underdeveloped symptoms, AD mice could be distinguished from the early ageSystems neuroscience ; cognitive neuroscience ; systems biology ; model organism