CMStatistics 2021: Start Registration
View Submission - CMStatistics
Title: Transfer learning for cognitive reserve quantification Authors:  Xi Zhu - New York State Psychiatric Institute (United States)
Yi Liu - Columbia University (United States)
Christian Habeck - Columbia University (United States)
Yaakov Stern - Columbia University (United States)
Seonjoo Lee - Columbia University/New York State Psychiatric Institute (United States) [presenting]
Abstract: Cognitive reserve has been introduced to explain individual differences in susceptibility to cognitive or functional impairment in the presence of age or pathology. We developed a deep learning model to quantify the CR as residual variance in memory performance using the structural MRI data from a lifespan healthy cohort. The generalizability of the sMRI-based deep learning model was tested in two independent healthy and Alzheimer cohorts using a transfer learning framework. Structural MRIs were collected from three cohorts: 495 healthy adults from RANN, 620 healthy participants (age 36-100) from lifespan Human Connectome Project Aging (HCPA), and 941 subjects from Alzheimer's Disease Neuroimaging Initiative (ADNI). Cognitive reserve was quantified by residuals which subtract the predicted memory from the true memory. Cascade neural network (CNN) models were used to train the RANN dataset for memory prediction. The CNN model trained on the RANN dataset exhibited a strong linear correlation between true and predicted memory based on the chosen T1 cortical thickness and volume predictors. In addition, the model generated from healthy lifespan data (RANN) was able to generalize to independent healthy lifespan data (HCPA) and older demented participants (ADNI) across different scanner types. The estimated CR was correlated with CR proxies such as education and IQ across all three datasets.