Related Experiment Video
Updated: Mar 24, 2026
Diffusion Tensor Magnetic Resonance Imaging in the Analysis of Neurodegenerative Diseases
Published on: July 28, 2013
Prediction of Impaired Performance in Trail Making Test in MCI Patients With Small Vessel Disease Using DTI Data
Abstract:
Mild cognitive impairment (MCI) is a common condition in patients with diffuse hyperintensities of cerebral white matter (WM) in T2-weighted magnetic resonance images and cerebral small vessel disease (SVD). In MCI due to SVD, the most prominent feature of cognitive impairment lies in degradation of executive functions, i.e., of processes that supervise the organization and execution of complex behavior. The trail making test is a widely employed test sensitive to cognitive processing speed and executive functioning. MCI due to SVD has been hypothesized to be the effect of WM damage, and diffusion tensor imaging (DTI) is a well-established technique for in vivo characterization of WM. We propose a machine learning scheme tailored to 1) predicting the impairment in executive functions in patients with MCI and SVD, and 2) examining the brain substrates of this impairment. We employed data from 40 MCI patients with SVD and created feature vectors by averaging mean diffusivity (MD) and fractional anisotropy maps within 50 WM regions of interest. We trained support vector machines (SVMs) with polynomial as well as radial basis function kernels using different DTI-derived features while simultaneously optimizing parameters in leave-one-out nested cross validation. The best performance was obtained using MD features only and linear kernel SVMs, which were able to distinguish an impaired performance with high sensitivity (72.7%-89.5%), specificity (71.4%-83.3%), and accuracy (77.5%-80.0%). While brain substrates of executive functions are still debated, feature ranking confirm that MD in several WM regions, not limited to the frontal lobes, are truly predictive of executive functions.
Insights
Machine learning accurately predicts executive function impairment in mild cognitive impairment (MCI) patients with cerebral small vessel disease (SVD). Mean diffusivity in white matter, not just frontal lobes, is key to understanding this cognitive decline.
Area of Science:
- Neuroimaging
- Neurology
- Artificial Intelligence
Background:
- Mild cognitive impairment (MCI) often co-occurs with cerebral small vessel disease (SVD) and white matter (WM) hyperintensities.
- Executive function deficits are a hallmark of MCI due to SVD, impacting complex behavior organization.
- Diffusion Tensor Imaging (DTI) is crucial for characterizing WM damage, a suspected cause of MCI due to SVD.
Purpose of the Study:
- To develop a machine learning model for predicting executive function impairment in MCI patients with SVD.
- To identify the specific white matter (WM) brain substrates associated with executive function deficits in this population.
Main Methods:
- Utilized DTI data from 40 MCI patients with SVD, extracting mean diffusivity (MD) and fractional anisotropy (FA) from 50 WM regions of interest.
- Trained Support Vector Machines (SVMs) with various kernels and optimized parameters using leave-one-out nested cross-validation.
- Employed feature ranking to identify predictive WM regions for executive function impairment.
Main Results:
- The optimal model used linear kernel SVMs with MD features, achieving high sensitivity (72.7%-89.5%), specificity (71.4%-83.3%), and accuracy (77.5%-80.0%).
- Feature ranking confirmed that MD in multiple WM regions, extending beyond the frontal lobes, significantly predicts executive function performance.
- This suggests a distributed network of WM damage underlies executive dysfunction in MCI due to SVD.
Conclusions:
- Machine learning, particularly SVMs with DTI-derived MD, can effectively predict executive function impairment in MCI due to SVD.
- Mean diffusivity in various white matter tracts is a strong neuroimaging biomarker for executive dysfunction in this condition.
- The findings highlight the importance of widespread white matter integrity for maintaining executive functions in patients with SVD.
