Related Experiment Video
Updated: Sep 20, 2025
09:47
Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
Published on: December 15, 2023
1.3K
Multimodal in vivo staging in amyotrophic lateral sclerosis using artificial intelligence
Anna Behler1, Hans-Peter Müller1, Kelly Del Tredici1
1Department of Neurology, University of Ulm, Germany.
Annals of Clinical and Translational Neurology
|June 10, 2022
Summary
This study used artificial intelligence to link brain imaging (DTI) and functional tests in ALS patients. It found distinct clusters correlating with disease progression, offering new ways to stage amyotrophic lateral sclerosis (ALS).
Area of Science:
- Neuroscience
- Neurology
- Medical Imaging
Background:
- Amyotrophic lateral sclerosis (ALS) neuropathology follows a four-stage pTDP-43 cerebral propagation scheme.
- In vivo imaging using diffusion tensor imaging (DTI) can visualize these stages in specific brain tracts.
- Cognitive and oculomotor dysfunctions in ALS are linked to brain microstructural changes, leading to separate staging classifications.
Purpose of the Study:
- To investigate the association between in vivo staging schemes of neuropathological spreading in ALS.
- To correlate diffusion tensor imaging (DTI) metrics with cognitive and oculomotor performance in ALS patients.
- To explore the utility of artificial intelligence in mapping in vivo correlates of ALS neuropathology.
Main Methods:
- 245 ALS patients underwent DTI, video-oculography, and cognitive testing (Edinburgh Cognitive and Behavioral ALS Screen - ECAS).
- Tract-related diffusion metrics, cognitive, and oculomotor parameters were analyzed.
- Hierarchical and k-means clustering algorithms were employed to identify patient subgroups.
Main Results:
- Cluster analysis differentiated ALS patients into four distinct clusters based on DTI metrics, cognitive, and oculomotor performance.
- Cluster A exhibited the highest fractional anisotropy (FA) and best cognitive/oculomotor function.
- Cluster D showed the lowest FA, lowest ECAS scores, and worst oculomotor performance, indicating advanced disease stages.
Conclusions:
- AI-based cluster analysis of multimodal data reveals high congruence between DTI, executive oculomotor function, and neuropsychological performance in ALS.
- This approach effectively maps in vivo correlates of neuropathological spreading.
- The findings support a multimodal staging system for ALS that integrates neuroimaging and functional assessments.
