Alzheimer Disease progression based on LSTM and CNN deep learning techniques Lead Investigator: Shaker El-Sappagh Institution : Inha University E-Mail : shaker_elsapagh@yahoo.com Proposal ID : 1184 Proposal Description: The goal is to use the dataset to build an accurate deep learning model to predict the progression of the disease in the future. We will use the most recent technique to build the most accurate model. A. We will use the dataset to predict the progression of Alzheimer disease based on the most recent deep learning models including CNN and LSTM. B. We will examine the effect of different biomarker groups and patient's demographics and medical history on the prediction accuracy of the model. C. We will try to utilize the time series nature of the data to predict multiple steps in the future based on the capabilities of the RNN and the large number of samples in the dataset. D. We expect to receive data with all features or biomarkers that may affect the diagnoses, evaluation, and prediction of patient conditions. These data include diagnosis data (baseline diagnosis, diagnoses at different stages of the disease development), Neuropsychological data, biospecimen, imaging features including MRI, symptoms, complications, questionnaires, demographics, etc.) Please note that as much data I get, the results will be much more accurate and trusted. E. The main outcome from our research will be the prediction of the patient next state of the disease based on his/her clinical dementia rating or the prediction of the most sensitive biomarkers. F. The data will be preprocessed to handle outliers, missing values, and other to improve the quality and prepare it to the deep learning model.