Related Experiment Video
Updated: Jul 30, 2026
Application of Granger Causality Analysis of the Directed Functional Connection in Alzheimer's Disease and Mild Cognitive Impairment
Published on: August 7, 2017
Brain Microstructural Changes in Patients with Amnestic mild Cognitive Impairment : Detected by Neurite Orientation
Xiuwei Fu1, Xiaonan Wang2, Yu Zhang2
1Department of Radiology, Tianjin Medical University General Hospital, Tianjin, China.
Purpose:
This study investigated brain microstructural changes in patients with amnestic mild cognitive impairment (aMCI) by retrospectively analyzing neurite orientation dispersion and density imaging (NODDI) data with machine learning algorithms.
Methods:
A total of 26 aMCI patients and 24 healthy controls (HC) underwent NODDI magnetic resonance imaging (MRI) examinations. The NODDI parameters including neurite density index (NDI), orientation dispersion index (ODI), and volume fraction of isotropic water molecules (Viso) were estimated. Machine learning algorithms such as K‑nearest neighbor (KNN), logistic regression (LR), random forest (RF), and support vector machine (SVM) were used to evaluate the diagnostic efficacy of NODDI parameters in predicting aMCI. The differences in the NODDI parameter values between the aMCI and HC groups were analyzed using the independent sample t‑test, False discovery rate (FDR) correction was used for multiple testing. After adjusting for age, sex, and educational years, partial correlation analysis was used to evaluate the relationship between NODDI parameters and clinical cognitive status of aMCI patients.
Results:
The NDI, ODI, and Viso values of white matter (WM) and gray matter (GM) structure templates combined with the KNN, LR, RF and SVM machine learning algorithms accomplished the discrimination between aMCI and HC groups. The NDI and ODI values decreased (p value range, < 0.001-0.042) and Viso values increased (p value range, < 0.001-0.043) in the aMCI group compared to the HCs. The NDI, ODI, and Viso values of the WM and GM structure templates with significant differences were significantly correlated with mini-mental state examination (MMSE) and Montreal cognitive assessment (MoCA) scores.
Conclusion:
NODDI combined with machine learning algorithms is a promising strategy for early diagnosis of aMCI. Moreover, NODDI parameters correlated with the clinical cognitive status of aMCI patients.
