Transformed domain convolutional neural network for Alzheimer's disease diagnosis using structural MRI

作者：

Highlights：

• Investigated Jacobian transformation to identify distinctive features from structural magnetic resonance imaging (sMRI) data.

• Fused Jacobian map with deep learning, which provided a quantitative measure for localized brain volume change and eventually built strong transformed domain classifier.

• Proposed a whole brain JD-CNN framework that neither required identification of discriminative landmark (LM) locations nor any region of interests (ROIs).

• Superior AD classification performance has been achieved as compared with previously reported state-of-the-art techniques.

摘要

•Investigated Jacobian transformation to identify distinctive features from structural magnetic resonance imaging (sMRI) data.•Fused Jacobian map with deep learning, which provided a quantitative measure for localized brain volume change and eventually built strong transformed domain classifier.•Proposed a whole brain JD-CNN framework that neither required identification of discriminative landmark (LM) locations nor any region of interests (ROIs).•Superior AD classification performance has been achieved as compared with previously reported state-of-the-art techniques.

论文关键词：Alzheimer disease (AD) detection,Brain disease,Convolutional neural network (CNN),Supervised learning,Structural magnetic resonance imaging (sMRI),Transform domain AD classification,AD diagnosis

论文评审过程：Received 11 April 2022, Revised 19 August 2022, Accepted 5 September 2022, Available online 7 September 2022, Version of Record 16 September 2022.

论文官网地址：https://doi.org/10.1016/j.patcog.2022.109031