Machine learning to detect microstructural brain changes in patients with amnestic mild cognitive impairment based on NODDI
Xiuwei Fu1, Yu Zhang2, Tongtong Li2, Yuanyuan Chen3, Xianchang Zhang4, and Hongyan Ni5
1Department of Radiology, Tianjin Medical University General Hospital, Tianjin, China, 2Department of Radiology, First Central Clinical institution, Tianjin Medical University, Tianjin, China, 3Institute of Medical Engineering and Translational Medicine, Tianjin University, Tianjin, China, 4MR Collaboration, Siemens Healthineers Ltd., Beijing, China, 5Department of Radiology, Tianjin First Central Hospital, Tianjin, China
This study investigated the changes in brain microstructure in patients with amnestic mild cognitive impairment (aMCI) using neurite orientation dispersion and density imaging (NODDI) combined with machine learning. Neurite density index (NDI) was significantly decreased in white matter, orientation dispersion index (ODI) was significantly decreased in gray matter, and volume fraction of isotropic water molecules (Viso) was significantly increased in the aMCI group. Further correlation and receiver operating characteristic (ROC) curve analyses showed NODDI may reflect the clinical cognitive status of aMCI. NODDI combined with a machine learning algorithm could be a promising alternative for early diagnosis of MCI.
This abstract and the presentation materials are available to members only;
a login is required.