Identifying the regional substrates predictive of Alzheimer's disease progression through a convolutional neural network model and occlusion

• Published in: Human brain mapping Vol. 43; no. 18; pp. 5509 - 5519
• Main Authors: Kwak, Kichang, Stanford, William, Dayan, Eran
• Format: Journal Article
• Language: English
• Published: Hoboken, USA John Wiley & Sons, Inc 15.12.2022
• Subjects: Alzheimer Disease - pathology; Alzheimer's disease; Artificial neural networks; Atrophy; brain atrophy; Cerebral blood flow; Clinical trials; Cognitive ability; Cognitive Dysfunction - diagnostic imaging; Cognitive Dysfunction - etiology; Deep learning; Dementia; Dementia disorders; Demographics; Diagnosis; Disease Progression; Humans; Ischemia; Magnetic resonance imaging; Magnetic Resonance Imaging - methods; mild cognitive impairment; Neural networks; Neural Networks, Computer; neurodegeneration; Neurodegenerative diseases; Occlusion; occlusion analysis; Prediction models; Substrates; neurodegeneration; occlusion analysis; deep learning; mild cognitive impairment; Alzheimer's disease; brain atrophy
• ISSN: 1065-9471; 1097-0193; 1097-0193
• DOI: 10.1002/hbm.26026

Progressive brain atrophy is a key neuropathological hallmark of Alzheimer's disease (AD) dementia. However, atrophy patterns along the progression of AD dementia are diffuse and variable and are often missed by univariate methods. Consequently, identifying the major regional atrophy patterns underlying AD dementia progression is challenging. In the current study, we propose a method that evaluates the degree to which specific regional atrophy patterns are predictive of AD dementia progression, while holding all other atrophy changes constant using a total sample of 334 subjects. We first trained a dense convolutional neural network model to differentiate individuals with mild cognitive impairment (MCI) who progress to AD dementia versus those with a stable MCI diagnosis. Then, we retested the model multiple times, each time occluding different regions of interest (ROIs) from the model's testing set's input. We also validated this approach by occluding ROIs based on Braak's staging scheme. We found that the hippocampus, fusiform, and inferior temporal gyri were the strongest predictors of AD dementia progression, in agreement with established staging models. We also found that occlusion of limbic ROIs defined according to Braak stage III had the largest impact on the performance of the model. Our predictive model reveals the major regional patterns of atrophy predictive of AD dementia progression. These results highlight the potential for early diagnosis and stratification of individuals with prodromal AD dementia based on patterns of cortical atrophy, prior to interventional clinical trials. Progressive brain atrophy is a key neuropathological hallmark of Alzheimer's disease (AD). However, atrophy patterns along the progression of AD are diffuse and variable and are often missed by univariate methods. Here, we propose a method based upon deep learning that evaluates the degree to which specific regional atrophy patterns are predictive of AD progression, while holding all other atrophy changes constant. Our results show that atrophy in the hippocampus, fusiform, and inferior temporal gyri was the strongest predictor of AD progression.