DTI based Alzheimer’s disease classification with rank modulated fusion of CNNs and random forest

• Published in: Expert systems with applications Vol. 169; p. 114338
• Main Authors: De, Arijit, Chowdhury, Ananda S.
• Format: Journal Article
• Language: English
• Published: New York Elsevier Ltd 01.05.2021; Elsevier BV
• Subjects: Alzheimer's disease; Alzheimer’s disease classification; Anisotropy; Brain; Classification; Cognitive ability; Decision fusion; Decision trees; DTI; Experimentation; Impairment; Medical imaging; Random forest; Tensors; VoxCNN; VoxCNN; DTI; Random forest; Alzheimer’s disease classification; Decision fusion
• ISSN: 0957-4174; 1873-6793
• DOI: 10.1016/j.eswa.2020.114338

Automated classification of Alzheimer’s disease (AD) plays a key role in the diagnosis of dementia. In this paper, we solve for the first time a direct four-class classification problem, namely, AD, Normal Control (CN), Early Mild Cognitive Impairment (EMCI) and Late Mild Cognitive Impairment (LMCI) by processing Diffusion Tensor Imaging (DTI) in 3D. DTI provides information on brain anatomy in form of Fractional Anisotropy (FA) and Mean Diffusivity (MD) along with Echo Planar Imaging (EPI) intensities. We separately train CNNs, more specifically, VoxCNNs on FA values, MD values, and EPI intensities on 3D DTI scan volumes. In addition, we feed average FA and MD values for each brain region, derived according to the Colin27 brain atlas, into a random forest classifier (RFC). These four (three separately trained VoxCNNs and one RFC) models are first applied in isolation for the above four-class classification problem. Individual classification results are then fused at the decision level using a modulated rank averaging strategy leading to a classification accuracy of 92.6%. Comprehensive experimentation on publicly available ADNI database clearly demonstrates the effectiveness of the proposed solution. •Four-class classification of Alzheimer’s disease addressed for the first time.•Effective DTI processing in 3D with CNN(s) and Random Forest as classifiers.•Rank modulated decision fusion for combining CNNs and Random Forest outputs.•State-of-the-art performance on publicly available ADNI database.