An artificial intelligence-derived metabolic network predicts psychosis in Alzheimer’s disease

Abstract Alzheimer’s Disease psychosis (AD + P) is characterized by accelerated cognitive decline and tau pathology. Through exploring the AD + P network (ADPN), the aim is to predict psychosis in AD and understand its mechanisms. Utilizing FDG PET scans from ADNI control and AD groups, we employed a convolutional neural network to identify and validate the ADPN. We analyzed network progression, clinical correlations, and psychosis prediction using expression scores, and network organization using graph theory. The ADPN accurately distinguishes AD + P from controls (97%), with increasing scores correlating with cognitive decline. ADPN-based approach predicts psychosis with 77% accuracy and identifies specific brain regions and connections associated with psychosis. Deep learning identified ADPN, linked to cognitive and functional decline. The increased metabolic connectivity between motor and language/social cognition regions in AD + P may drive delusions and agitated behavior. ADPN holds promise as a biomarker for AD + P, aiding in treatment development and patient stratification.