Activity Number:
|
215
- Contributed Poster Presentations: Section on Statistical Learning and Data Science
|
Type:
|
Contributed
|
Date/Time:
|
Tuesday, August 4, 2020 : 10:00 AM to 2:00 PM
|
Sponsor:
|
Section on Statistical Learning and Data Science
|
Abstract #312963
|
|
Title:
|
Early Prediction of Alzheimer’s Disease with Deep Learning Using Data Integration of MRI Data and Clinical Data
|
Author(s):
|
Lisa Neums* and Jinxiang Hu and Jeffrey Thompson
|
Companies:
|
University of Kansas Medical Center and University of Kansas Medical Center and University of Kansas Medical Center
|
Keywords:
|
Alzheimer's disease;
MRI data;
longitudinal;
prediction of progression;
data integration;
deep neural networks
|
Abstract:
|
Alzheimer’s disease (AD) progresses slowly but the timing and severity of symptoms for each patient varies strongly as each patient is affected by the disease differently. Studies using longitudinal MRI data or clinical data, which observe the progression of AD in each patient over time, have been conducted to increase the accuracy of diagnosis of patients in early stages. Another important aspect is the prediction of the progression of AD to address the varying development of the disease in each individual. So far, no study has been presented to include data integration of MRI data and clinical data for the prediction of the progression in AD. Here we show a novel method for longitudinal MRI data in combination with traditional cognitive tests and other measures to predict the progression of Alzheimer’s disease as early as possible. Our method uses exploratory factor analysis to obtain the best clinical predictor variables and recurrent convolutional neural networks to incorporate longitudinal image data.
|
Authors who are presenting talks have a * after their name.