Predicting progression of Alzheimer?s Disease cognitive decline with Survival
Net 
   Lead Investigator:    Juan Vizcarra
   Institution      :    Georgia Institute of Technology Emory University
   E-Mail           :    jvizcar@emory.edu
   Proposal ID      :    1383



   Proposal Description:
   1) Implement a modified version of a machine learning model to predict progression of cognitive decline in the UDS dataset. Of interest is predicting accurately subjects who will progress from normal cognition to MCI and MCI to AD with time to progression predictions. This approach will benefit in that it   has the potential of identifying key factors that strongly predict whether a subject will progress in cognitive decline or not and how fast.                   2) Modify published Survival net model (predicts survival in cancer dataset) to predict progression from cognitive states. Identify key factors from the UDS dataset that are important in the prediction. Incorporate features from the NP dataset to include in survival net predictions and identify key features from this added dataset. Develop a pipeline that will extract key NP features and combine with UDS features to predict cognitive declines and predictive factors.