Automatic Diagnosis of Alzheimer’s Disease and Mild Cognitive Impairment Based on CNN+SVM Networks with End-to-end Training

Alzheimer's disease (AD) is an irreversible neurodegenerative disease and, at present, once it has been diagnosed, there is no effective curative treatment. Accurate and early diagnosis of Alzheimer's disease is crucial for improving the condition of patients since effective preventive measures can be taken in advance to delay the onset time of the disease. Fluorodeoxyglucose positron emission tomography (FDG-PET) is an effective biomarker of the symptom of AD's, and has been used as medical imaging data for diagnosing AD's. Mild cognitive impairment (MCI) is regarded as an early symptom of AD's, and it has been shown that MCI also has a certain biomedical correlation with FDG-PET. In this paper, we explore how to use 3D FDG-PET images to realize the effective recognition of MCI's, and thus achieve the early prediction of AD's. This problem is then taken as the classification of three categories of FDG-PET images, including MCI, AD and NC (normal controls). In order to get better classification performance, a novel network model is proposed in the paper based on 3D convolution neural networks (CNN) and support vector machines (SVM) by utilizing both the excellent abilities of CNN in feature extraction and SVM in classification. In order to make full use of the optimal property of SVM in solving binary classification problems, the three-category classification problem is divided into three binary classifications, each binary classification being realized with a CNN+SVM network. Then the outputs of the three CNN+SVM networks are fused into a final three-category classification result. An end-to-end learning algorithm is developed to train the CNN+SVM networks and a decision fusion strategy is exploited to realize the fusion of the outputs of three CNN+SVM networks. Experimental results obtained in the work with comparative analyses confirm the effectiveness of the proposed method.