Multi-modal deep learning model for auxiliary diagnosis of Alzheimer’s disease

Abstract

Alzheimer’s disease (AD) is one of the most difficult to cure diseases. Alzheimer’s disease seriously affects the normal lives of the elderly and their families. The mild cognitive impairment (MCI) is a transitional state between the normal aging and Alzheimer’s disease, and MCI is most likely to converted to AD later. MCI is often misdiagnosed as the symptoms of normal aging, which results to miss the best opportunity of treatment. Therefore, the accurate diagnosis of MCI is essential for the early diagnosis and treatment of AD. This paper presents a deep learning model for the auxiliary diagnosis of AD, which simulates the clinician’s diagnostic process. During the diagnosis of AD, clinician usually refers to the results of various neuroimaging, as well as the results of neuropsychological diagnosis. In this paper, the multi-modal medical images are trained by two independent convolutional neural networks. Then the consistency of the output of two convolutional neural networks is judged by the correlation analysis. Finally, the results of multi-modal neuroimaging diagnosis are combined with the results of clinical neuropsychological diagnosis. The proposed model provides a comprehensive analysis about patient’s pathology and psychology at the same time, therefore, it improves the accuracy of auxiliary diagnosis. The diagnosis process is closer to the process of clinician’s diagnosis and easy to implement. Experiments on the public database of ADNI (Alzheimer’s disease neuroimaging initiative) show that the proposed method has better performance, and can achieve an excellent diagnostic efficiency in the auxiliary diagnosis of AD.