Abstract #4529

Evaluating advanced multi-shell diffusion MRI microstructural biomarkers of Alzheimer’s disease

Julio Ernesto Villalon Reina¹, Talia Miriam Nir², Sophia Thomopoulos², Lauren E Salminen³, Neda M Jahanshad², Rutger Fick⁴, Matteo Frigo⁵, Rachid Deriche⁵, and Paul M Thompson²

¹USC Imaging Genetics Center, University of Southern California, Los Angeles, CA, United States, ²USC Stevens Neuroimaging and Informatics Institute, University of Southern California, Marina del Rey, CA, United States, ³USC Imaging Genetics Center, University of Southern California, Marina del Rey, CA, United States, ⁴Therapanacea, Paris, France, ⁵Athena Project Team, Inria Sophia-Antipolis Méditerranée, Université Côte d'Azur, Nice, France

To identify microstructure-based biomarkers sensitive to cognitive impairment, we used ADNI-3 multi-shell dMRI data to estimate 18 measures from seven dMRI models and assessed their ability to predict mild cognitive impairment (MCI). For each measure, we used TV-L1 regularized logistic regression to find cohesive clusters of brain tissue that contribute to correct classification. We found that tensor-based (DTI) diffusivity and multi-compartment spherical mean technique (MC-SMT) measures showed the highest prediction accuracy, but differential anatomical distributions of classifying voxels. MC-SMT may offer greater sensitivity and specificity to MCI than DTI as MC-SMT resulted in the highest recall and fewest classifying voxels.

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

Join Here