Related Experiment Video
Updated: Jul 21, 2026
12:21
Probing the Brain in Autism Using fMRI and Diffusion Tensor Imaging
Published on: September 12, 2011
25.7K
Voxel-based analysis and multivariate pattern analysis of diffusion tensor imaging study in anti-NMDA receptor
Yanli Liang1, Luhui Cai1, Xia Zhou1
1Department of Neurology, The First Affiliated Hospital of Guangxi Medical University, Nanning, China.
Neuroradiology
|December 1, 2019
Summary
Brain white matter changes in anti-N-methyl-D-aspartate receptor encephalitis patients are linked to cognitive deficits. Diffusion tensor imaging and machine learning accurately classify patients based on these white matter alterations.
Area of Science:
- Neuroscience
- Neurology
- Radiology
Background:
- Anti-N-methyl-D-aspartate (anti-NMDA) receptor encephalitis is an autoimmune disorder that can cause significant neurological and psychiatric symptoms.
- Cognitive deficits, particularly in memory, executive function, and perception, are common residual symptoms following treatment.
- Understanding the underlying brain changes, especially in white matter (WM), is crucial for diagnosing and managing these deficits.
Purpose of the Study:
- To investigate white matter (WM) alterations in patients with anti-NMDA receptor encephalitis.
- To explore the relationship between WM changes and cognitive function in these patients.
- To assess the utility of multivariate pattern analysis (MVPA) using diffusion tensor imaging (DTI) metrics for classifying patients.
Main Methods:
- Fifteen patients with anti-NMDA receptor encephalitis and fifteen healthy controls underwent DTI and the Montreal Cognitive Assessment (MoCA).
- Fractional anisotropy (FA) and mean diffusivity (MD) values were extracted from various brain regions.
- MVPA was employed to classify patients based on FA and MD, calculating regional weights.
Main Results:
- Patients exhibited reduced FA in the right middle temporal gyrus and right precuneus, and increased MD in the left medial temporal gyrus and left frontal lobe compared to controls.
- Cognitive impairments were noted in executive function, fluency, delayed recall, and visual perception.
- FA in the right precuneus correlated positively with MoCA and fluency scores; MD in the left frontal lobe correlated negatively with MoCA scores. MD in the left medial temporal gyrus correlated positively with delayed recall. MVPA achieved 70-80% accuracy in classification.
- Key brain regions identified by MVPA (temporal lobe, cerebellum, hippocampus) showed overlapping importance for both FA and MD.
Conclusions:
- White matter microstructural changes, detectable by DTI, are significantly associated with cognitive impairments in anti-NMDA receptor encephalitis.
- MVPA using FA and MD demonstrates robust classification capabilities, aiding in understanding disease-specific patterns.
- These findings offer insights into the pathophysiology of persistent cognitive deficits and highlight DTI as a valuable tool for assessment.
