Abstract: Objective To explore the effects of artificial intelligence (AI)-assisted multi-cognitive domain training on community elderly with mild cognitive impairment (MCI). Methods A randomized controlled trial was conducted. A total of 100 community-dwelling older adults diagnosed with MCI and visiting the outpatient clinic of the Third People’s Hospital of Ganzhou between January 2024 and February 2025 were enrolled and randomly assigned in a 1∶1 ratio to either a control group (n=50) or an experimental group (n=50). The control group received conventional cognitive intervention, while the experimental group received AI-assisted multi-cognitive domain training. Both interventions lasted for 12 weeks. Cognitive function was assessed using the Montreal Cognitive Assessment (MoCA) and the Mini-Mental State Examination (MMSE); Neuropsychiatric symptoms were evaluated with the Neuropsychiatric Inventory (NPI); And activities of daily living were measured by the modified Barthel Index (MBI) at baseline, the 8th week, and the 12th week of the intervention. Results There were no statistically significant differences between the two groups in terms of age, gender, years of education, marital status, or use of Alzheimer’s disease medication (all P>0.05). Repeated-measures ANONA showed that, at the 8th and 12th week of the intervention, both groups demonstrated significant improvements in the scores of MoCA, MMSE, NPI, and MBI compared to baseline, especially in the experimental group (all P<0.05). Conclusions AI-assisted multi-cognitive domain training can effectively improve cognitive function, alleviate neuropsychiatric symptoms, and enhance activities of daily living among community elderly with MCI, demonstrating favorable intervention efficacy and potential for broader application.    

Key words: artificial intelligence, mild cognitive impairment, cognitive training, randomized controlled trial, aged