Alzheimer's disease (AD) is a progressive and irreversible neurodegenerative disease. To distinguish the stage of the disease, AD classification technology challenge has been proposed in Pattern Recognition and Computer Vision 2021 (PRCV 2021) which provides the gray volume and average cortical thickness data extracted in multiple atlases from magnetic resonance imaging (MRI). Traditional methods either train with convolutional neural network (CNN) by MRI data to adapt the spatial features of images or train with recurrent neural network (RNN) by temporal features to predict the next stage. However, the morphological features from the challenge have been extracted into discrete values. We present a multi-atlases multi-layer perceptron (MAMLP) approach to deal with the relationship between morphological features and the stage of the disease. The model consists of multiple multi-layer perceptron (MLP) modules, and morphological features extracted from different atlases will be classified by different MLP modules. The final vote of all classification results obtains the predicted disease stage. Firstly, to preserve the diversity of brain features, the most representative atlases are chosen from groups of similar atlases, and one atlas is selected in each group. Secondly, each atlas is fed into one MLP to fetch the score of the classification. Thirdly, to obtain more stable results, scores from different atlases are combined to vote the result of the classification. Based on this approach, we rank 10th among 373 teams in the challenge. The results of the experiment indicate as follows: (1) Group selection of atlas reduces the number of features required without reducing the accuracy of the model; (2) The MLP architecture achieves better performance than CNN and RNN networks in morphological features; and (3) Compared with other networks, the combination of multiple MLP networks has faster convergence of about 40% and makes the classification more stable.
Cortical and subcortical structural alteration has been extensively reported in schizophrenia, including the unusual expansion of gray matter volumes (GMVs) of basal ganglia (BG), especially putamen. Previous genome-wide association studies pinpointed kinectin 1 gene (KTN1) as the most significant gene regulating the GMV of putamen. In this study, the role of KTN1 variants in risk and pathogenesis of schizophrenia was explored. A dense set of SNPs (n = 849) covering entire KTN1 was analyzed in three independent European- or African-American samples (n = 6704) and one mixed European and Asian Psychiatric Genomics Consortium sample (n = 56,418 cases vs. 78,818 controls), to identify replicable SNP-schizophrenia associations. The regulatory effects of schizophrenia-associated variants on the KTN1 mRNA expression in 16 cortical or subcortical regions in two European cohorts (n = 138 and 210, respectively), the total intracranial volume (ICV) in 46 European cohorts (n = 18,713), the GMVs of seven subcortical structures in 50 European cohorts (n = 38,258), and the surface areas (SA) and thickness (TH) of whole cortex and 34 cortical regions in 50 European cohorts (n = 33,992) and eight non-European cohorts (n = 2944) were carefully explored. We found that across entire KTN1, only 26 SNPs within the same block (r2 > 0.85) were associated with schizophrenia across ≥ 2 independent samples (7.5 × 10–5 ≤ p ≤ 0.048). The schizophrenia-risk alleles, which increased significantly risk for schizophrenia in Europeans (q < 0.05), were all minor alleles (f < 0.5), consistently increased (1) the KTN1 mRNA expression in 12 brain regions significantly (5.9 × 10–12 ≤ p ≤ 0.050; q < 0.05), (2) the ICV significantly (6.1 × 10–4 ≤ p ≤ 0.008; q < 0.05), (3) the SA of whole (9.6 × 10–3 ≤ p ≤ 0.047) and two regional cortices potentially (2.5 × 10–3 ≤ p ≤ 0.042; q > 0.05), and (4) the TH of eight regional cortices potentially (0.006 ≤ p ≤ 0.050; q > 0.05), and consistently decreased (1) the BG GMVs significantly (1.8 × 10–19 ≤ p ≤ 0.050; q < 0.05), especially putamen GMV (1.8 × 10–19 ≤ p ≤ 1.0 × 10–4; q < 0.05, (2) the SA of four regional cortices potentially (0.010 ≤ p ≤ 0.048), and (3) the TH of four regional cortices potentially (0.015 ≤ p ≤ 0.049) in Europeans. We concluded that we identified a significant, functional, and robust risk variant block covering entire KTN1 that might play a critical role in the risk and pathogenesis of schizophrenia.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.