Related Experiment Video
Updated: Aug 24, 2025
Author Spotlight: Using Motor Imagery Brain-Computer Interface to Improve Motor and Cognitive Function in Stroke Patients
Published on: September 1, 2023
Brain volumetric correlates of remotely versus in-person administered symbol digit modalities test in multiple
Korhan Buyukturkoglu1, Jordan D Dworkin2, Victor Leiva3
1Department of Neurology, Columbia University Irving Medical Center, 630 W. 168th Street, PH 18-324, New York, NY 10032, USA; The Center for Translational and Computational Neuroimmunology, NY, USA; Columbia University MS Center, NY, USA.
Background:
Prior studies in multiple sclerosis (MS) support reliability of telehealth-delivered cognitive batteries, although, to date, none have reported relationships of cognitive test performance to neural correlates across administration modalities. In this study we aimed to compare brain-behavior relationships, using the Symbol Digit Modalities Test (SDMT), the most reliable and sensitive cognitive measure in MS, measured from patients seen via telehealth versus in-person.
Methods:
SDMT was administered to individuals with MS either in-person (N=60, mean age=39.7) or remotely via video conference (N=51, mean age=47.4). Magnetic resonance imaging (MRI) data was collected in 3-Tesla scanners. Using 3-dimensional T1 images cerebral, cortical, deep gray, cerebral white matter and thalamic nuclei volumes were calculated. Using a meta-analysis approach with an interaction term for participant group, individual regression models were run for each MRI measure having SDMT scores as the outcome variable in each model. In addition, the correlation and average difference between In-person and Remote group associations across the MRI measures were calculated. Finally, for each MRI variable I2 score was quantified to test the heterogeneity between the groups.
Results:
Administration modality did not affect the association of SDMT performance with MRI measures. Brain tissue volumes showing high associations with the SDMT scores in one group also showed high associations in the other (r = 0.83; 95% CI = [0.07, 0.86]). The average difference between the In-person and the Remote group associations was not significant (βRemote - βIn-person = 0.14, 95% CI = [-0.04, 0.34]). Across MRI measures, the average I2 value was 14%, reflecting very little heterogeneity in the relationship of SDMT performance to brain volume.
Conclusion:
We found consistent relationships to neural correlates across in-person and remote SDMT administration modalities. Hence, our study extended the findings of the previous studies demonstrating the feasibility of remote administration of the SDMT.
