IMR Press / FBL / Volume 23 / Issue 3 / DOI: 10.2741/4606

Frontiers in Bioscience-Landmark (FBL) is published by IMR Press from Volume 26 Issue 5 (2021). Previous articles were published by another publisher on a subscription basis, and they are hosted by IMR Press on imrpress.com as a courtesy and upon agreement with Frontiers in Bioscience.

Open Access Article

Alzheimer's disease diagnostics by a 3D deeply supervised adaptable convolutional network

Show Less
1 Department of Electrical and Computer Engineering, University of Louisville, Louisville, KY, USA
2 Department of Bioengineering, University of Louisville, Louisville, KY, USA
3 Department of Electrical and Computer Engineering, Abu Dhabi University, USA
4 Department of Pediatrics, University of South Carolina, SC, USA
5 Department of Neurology, University of Louisville, USA
6 Department of Computer Science, University of Auckland, New Zealand
Front. Biosci. (Landmark Ed) 2018, 23(3), 584–596; https://doi.org/10.2741/4606
Published: 1 January 2018
Abstract

Early diagnosis is playing an important role in preventing progress of the Alzheimer’s disease (AD). This paper proposes to improve the prediction of AD with a deep 3D Convolutional Neural Network (3D-CNN), which can show generic features capturing AD biomarkers extracted from brain images, adapt to different domain datasets, and accurately classify subjects with improved fine-tuning method. The 3D-CNN is built upon a convolutional autoencoder, which is pre-trained to capture anatomical shape variations in structural brain MRI scans for source domain. Fully connected upper layers of the 3D-CNN are then fine-tuned for each task-specific AD classification in target domain. In this paper, deep supervision algorithm is used to improve the performance of already proposed 3D Adaptive CNN. Experiments on the ADNI MRI dataset without skull-stripping preprocessing have shown that the proposed 3D Deeply Supervised Adaptable CNN outperforms several proposed approaches, including 3D-CNN model, other CNN-based methods and conventional classifiers by accuracy and robustness. Abilities of the proposed network to generalize the features learnt and adapt to other domains have been validated on the CADDementia dataset.

Keywords
Alzheimer’s disease
deep learning
3D convolutional network
Autoencoder
brain MRI
Share
Back to top