Multi-model and multi-slice ensemble learning architecture based on 2D convolutional neural networks for Alzheimer's disease diagnosis

• Published in: Computers in biology and medicine Vol. 136; p. 104678
• Main Authors: Kang, Wenjie, Lin, Lan, Zhang, Baiwen, Shen, Xiaoqi, Wu, Shuicai
• Format: Journal Article
• Language: English
• Published: Oxford Elsevier Ltd 01.09.2021; Elsevier Limited
• Subjects: Accuracy; Alzheimer's disease; Artificial neural networks; Brain mapping; Classification; Cognitive ability; Convolutional neural networks; Datasets; Deep learning; Diagnosis; Ensemble learning; Generative adversarial networks; Internal Medicine; Machine learning; Magnetic resonance imaging; Medical diagnosis; Medical imaging; Medical research; Mild cognitive impairment; Neural networks; Neurodegenerative diseases; Neuroimaging; Other; Software; State-of-the-art reviews; Structural MRI; Substantia grisea; Transfer learning; Two dimensional models; Deep learning; Structural MRI; Generative adversarial networks; Convolutional neural networks; Mild cognitive impairment; Alzheimer's disease
• ISSN: 0010-4825; 1879-0534; 1879-0534
• DOI: 10.1016/j.compbiomed.2021.104678

Alzheimer's Disease (AD) is a chronic neurodegenerative disease without effective medications or supplemental treatments. Thus, predicting AD progression is crucial for clinical practice and medical research. Due to limited neuroimaging data, two-dimensional convolutional neural networks (2D CNNs) have been commonly adopted to differentiate among cognitively normal subjects (CN), people with mild cognitive impairment (MCI), and AD patients. Therefore, this paper proposes an ensemble learning (EL) architecture based on 2D CNNs, using a multi-model and multi-slice ensemble. First, the top 11 coronal slices of grey matter density maps for AD versus CN classifications were selected. Second, the discriminator of a generative adversarial network, VGG16, and ResNet50 were trained with the selected slices, and the majority voting scheme was used to merge the multi-slice decisions of each model. Afterwards, those three classifiers were used to construct an ensemble model. Multi-slice ensemble learning was designed to obtain spatial features, while multi-model integration reduced the prediction error rate. Finally, transfer learning was used in domain adaptation to refine those CNNs, moving them from working solely with AD versus CN classifications to being applicable to other tasks. This ensemble approach achieved accuracy values of 90.36%, 77.19%, and 72.36% when classifying AD versus CN, AD versus MCI, and MCI versus CN, respectively. Compared with other state-of-the-art 2D studies, the proposed approach provides an effective, accurate, automatic diagnosis along the AD continuum. This technique may enhance AD diagnostics when the sample size is limited. •Multi-slice and multi-model ensemble.•Ensemble learning classifies AD patients with high accuracy based on structural MRI.•Generative adversarial network is more effective for task adaptation transfer in weak supervision task.•Utilizing domain adaptation transfer learning and task adaptation transfer learning for re-purposing models.