Abstract #0899

Machine learning and rapid multi-parametric relaxometry can differentiate demyelinating disorders with high accuracy

Gabriel Mangeat^1,2, Russell Ouellette^2,3,4, Maxime Wabartha¹, Virginija Danylaité Karrenbauer^3,5, Nikola Stikov¹, Marcel Warntjes^6,7, Nikola Stikov^1,8, Caterina Mainero^2,9, Julien Cohen-Adad^1,10, and Tobias Granberg^2,3,4,9

¹NeuroPoly Lab, Institute of Biomedical Engineering, Polytechnique Montreal, Montreal, QC, Canada, ²Athinoula A. Martinos Center for Biomedical Imaging, MGH, Charlestown, MA, United States, ³Department of Clinical Neuroscience, Karolinska Institutet, Stockholm, Sweden, ⁴Department of Radiology, Division of Neuroradiology, Karolinska University Hospital, Stockholm, Sweden, ⁵Department of Neurology, Karolinska University Hospital, Stockholm, Sweden, ⁶Center for Medical Imaging Science and Visualization, CMIV, Linköping, Sweden, ⁷SyntheticMR, Linköping, Sweden, ⁸Montreal Health Institute, Montreal, QC, Canada, ⁹Harvard Medical School, Boston, MA, United States, ¹⁰Functional Neuroimaging Unit, CRIUGM, Université de Montréal, Montreal, QC, Canada

Hereditary diffuse leukoencephalopathy with spheroids (HDLS) and multiple sclerosis (MS) are both demyelinating and neurodegenerative disorders that can be hard to distinguish clinically and radiologically. Here, we present a machine learning method that relies on rapid multi-parametric relaxometry and volumetry to achieve a robust classification of HDLS vs. MS. Linear discriminant analysis was shown to be a favorable approach compared to non-linear options. A leave-one-out cross-validation show a detection rate of 100% and 0% false positives for both conditions, which suggests that computer-assistance maybe helpful in accurately diagnosing these disorders.

This abstract and the presentation materials are available to members only; a login is required.

Join Here