Classification of mild cognitive impairment and Alzheimer’s Disease with machine-learning techniques using 1H Magnetic Resonance Spectroscopy data

作者：

Highlights：

• First application of 1H-MRS data and machine-learning to the classification of AD.

• Classification of individuals affected by different stages of dementia.

• With two spectroscopic voxel volumes in left hippocampus we achieved a 0.866 AUROC.

• Classification results are in agreement with previous studies using MRI data.

• Composition of white and grey matter and cerebrospinal fluid is essential in 1H-MRS.

摘要

•First application of 1H-MRS data and machine-learning to the classification of AD.•Classification of individuals affected by different stages of dementia.•With two spectroscopic voxel volumes in left hippocampus we achieved a 0.866 AUROC.•Classification results are in agreement with previous studies using MRI data.•Composition of white and grey matter and cerebrospinal fluid is essential in 1H-MRS.

论文关键词：Magnetic Resonance Spectroscopy,Metabolite,Alzheimer’s Disease,Machine learning,Single-layer perceptron

论文评审过程：Available online 30 March 2015.

论文官网地址：https://doi.org/10.1016/j.eswa.2015.03.011