There are currently at least 50 million dementia patients worldwide, and this number is expected to reach 152 million by 2050, of which about 60-70% will be Alzheimer’s patients. The companion robot based on deep learning is a product of the development of artificial intelligence technology, which is of great significance to the physical and mental health of the elderly, so it is used in the research on the treatment of Alzheimer’s patients.

100 patients with Alzheimer’s disease in a hospital were selected for the study, and 50 patients were randomly divided into experimental group and control group. In the experiment, 50 patients with Alzheimer in the experimental group used a companion robot based on deep learning for auxiliary treatment while carrying out daily treatment. The control group of 50 patients did not receive any adjuvant therapy in addition to daily treatment. After three months of treatment, the study used the 3D-CAM and the mini–mental state examination (MMSE) to collect the treatment status of all patients, and used the SPSS23.0 statistical software to statistically analyze the collected data.

After statistical analysis, the results of the two groups were obtained. The scores of 3D-CAM and MMSE in the experimental group were significantly higher than those in the control group and the difference was statistically significant.

Companion robots based on deep learning are helpful in the treatment of Alzheimer’s patients. They can improve the therapeutic effect and have certain social value.

The Fundamental Research Funds in Heilongjiang Provincial Universities (No.135309356); Qiqihar University Young Teachers’ Scientific Research Initiation Support Program (No.2012k-M17).