AUTHOR=Qureshi Muhammad Naveed Iqbal , Ryu Seungjun , Song Joonyoung , Lee Kun Ho , Lee Boreom 

TITLE=Evaluation of Functional Decline in Alzheimer’s Dementia Using 3D Deep Learning and Group ICA for rs-fMRI Measurements

JOURNAL=Frontiers in Aging Neuroscience

VOLUME=Volume 11 - 2019

YEAR=2019

URL=https://www.frontiersin.org/journals/aging-neuroscience/articles/10.3389/fnagi.2019.00008

DOI=10.3389/fnagi.2019.00008

ISSN=1663-4365

ABSTRACT=Purpose: To perform automatic assessment of dementia severity using deep learning framework applied to resting state functional MRI (rs-fMRI) data. 
Method: We divided 133 Alzheimer’s disease patients with Clinical Dementia Rating (CDR) scores from 0.5 to 3 into two groups based on dementia severity; the groups with very mild/mild (CDR: 0.5–1) and moderate to severe (CDR: 2–3) dementia consisted of 77 and 56 subjects, respectively. We used rs-fMRI to extract functional connectivity features, calculated using independent component analysis (ICA), and performed automated severity classification with three-dimensional convolutional neural networks based on deep learning. 
Results: The mean balanced classification accuracy was 0.923±0.042 (p<0.001) with a specificity of 0.946±0.019 and sensitivity of 0.896±0.077. The rs-fMRI data indicated that the medial frontal, sensorimotor, executive control, dorsal attention, and visual related networks mainly correlated with dementia severity. 
Conclusions: Our CDR-based novel classification using rs-fMRI is an acceptable objective severity indicator. In the absence of trained neuropsychologists, dementia severity can be objectively and accurately classified using 3D-deep learning framework with rs-fMRI independent components.