Classification of Alzheimer Disease Based on Structural Magnetic Resonance Imaging by Kernel Support Vector Machine Decision Tree

Abstract

Abstract-In this paper we proposed a novel classification system to distinguish among elderly subjects with Alzheimer's disease (AD), mild cognitive impairment (MCI), and normal controls (NC). The method employed the magnetic resonance imaging (MRI) data of 178 subjects consisting of 97 NCs, 57 MCIs, and 24 ADs. First, all these three dimensional (3D) MRI images were preprocessed with atlas-registered normalization. Then, gray matter images were extracted and the 3D images were under-sampled. Afterwards, principle component analysis was applied for feature extraction. In total, 20 principal components (PC) were extracted from 3D MRI data using singular value decomposition (SVD) algorithm, and 2 PCs were extracted from additional information (consisting of demographics, clinical examination, and derived anatomic volumes) using alternating least squares (ALS). On the basic of the 22 features, we constructed a kernel support vector machine decision tree (kSVM-DT). The error penalty parameter C and kernel parameter σ were determined by Particle Swarm Optimization (PSO). The weights ω and biases b were still obtained by quadratic programming method. 5-fold cross validation was employed to obtain the out-of-sample estimate. The results show that the proposed kSVM-DT achieves 80% classification accuracy, better than 74% of the method without kernel. Besides, the PSO exceeds the random selection method in choosing the parameters of the classifier. The computation time to predict a new patient is only 0.022 s.

HRR Profiles Time-Frequency Non-Negative Sparse Coding for Sar Target Classification

“…When λ equals zero, the objective function reduces to the squared error version of non-negative matrix factorization (NMF) [25]. The form of ξ defines how sparseness is measured, and it is suggested that a typical choice for ξ is ξ(H kj ) = |H kj | in [11].…”

“…Thus, completely non-negative sparse coding have been investigated in [23]. Recently, NNSC has emerged as a useful feature extraction method in areas related to face recognition and image denoising [24,25]. In this work, non-negative T -F representations has been applied to target HRR profiles to obtain non-negative T -F data matrix, so NNSC is naturally considered for non-negative T -F feature extraction.…”

Triangulation Reconstruction for 3D Surface Based on Information Model

“…Up to date there is no perfect resolution of 3D triangulation in theory and method [3,8,9]. There are four triangulation methods including triangulation growth method, point by point interpolation, merge segmentation method and 3D to 2D projection.…”

Neuroimaging and Pattern Recognition Techniques for Automatic Detection of Alzheimer’s Disease: A Review

“…Zhang [43] proposed a novel classification system to distinguish among elderly subjects with AD, MCI, and NC. First, all these three dimensional (3D) MRI images were pre-processed with atlas-registered normalization.…”