Using machine learning to classify temporal lobe epilepsy based on diffusion MRI. Issue 10 (30th August 2017)

• Record Type: Journal Article
• Authors: Del Gaizo, John; Mofrad, Neda; Jensen, Jens H.; Clark, David; Glenn, Russell; Helpern, Joseph; Bonilha, Leonardo
• Is Part Of: Brain and behavior. Volume 7:Issue 10(2017)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2017-08-30
• Subjects: diffusion kurtosis imaging -- epilepsy -- machine learning -- Magnetic Resonance Imaging (MRI) -- support vector machines; Neurology -- Periodicals; Neurosciences -- Periodicals; Psychology -- Periodicals; Psychiatry -- Periodicals; 616.8005
• DOI: 10.1002/brb3.801
• Languages: English
• ISSNs: 2162-3279

Abstract

Abstract: Background: It is common for patients diagnosed with medial temporal lobe epilepsy (TLE) to have extrahippocampal damage. However, it is unclear whether microstructural extrahippocampal abnormalities are consistent enough to enable classification using diffusion MRI imaging. Therefore, we implemented a support vector machine (SVM)‐based method to predict TLE from three different imaging modalities: mean kurtosis (MK), mean diffusivity (MD), and fractional anisotropy (FA). While MD and FA can be calculated from traditional diffusion tensor imaging (DTI), MK requires diffusion kurtosis imaging (DKI). Methods: Thirty‐two TLE patients and 36 healthy controls underwent DKI imaging. To measure predictive capability, a fivefold cross‐validation (CV) was repeated for 1000 iterations. An ensemble of SVM models, each with a different regularization value, was trained with the subject images in the training set, and had performance assessed on the test set. The different regularization values were determined using a Bayesian‐based method. Results: Mean kurtosis achieved higher accuracy than both FA and MD on every iteration, and had far superior average accuracy: 0.82 (MK), 0.68 (FA), and 0.51 (MD). Finally, the MK voxels with the highest coefficients in the predictive models were distributed within the inferior medial aspect of the temporal lobes. Conclusion: These results corroborate our earlier publications which indicated that DKI shows more promise in identifying TLE‐associated pathological features than DTI. Also, the locations of the contributory MK voxels were in areas with high fiber crossing and complex fiber anatomy. These traits result in non‐Gaussian water diffusion, and hence render DTI less likely to detect abnormalities. If the location of consistent microstructural abnormalities can be better understood, then it may be possible in the future to identify the various phenotypes of TLE. This is important since treatment outcome varies dependent on type of TLE. Abstract : This study applies machine learning to diffusion‐derived measures (mean diffusivity, MD; fractional anisotropy, FA; and mean kurtosis, MK) to predict medial temporal lobe epilepsy (TLE) with linear classifiers that operate on a voxel‐wise level. The best classification was achieved with MK, corroborating our previous work that kurtosis‐based imaging is more sensitive to microstructural abnormalities associated with TLE than traditional diffusion metrics. The strong prediction performance of the models trained with the MK‐images indicates that there may be a common pathological pattern across individuals with TLE.