There are many active research projects accessing and applying shared ADNI data. Use the search above to find specific research focuses on the active ADNI investigations. This information is requested annually as a requirement for data access.
Principal Investigator | |
Principal Investigator's Name: | Yanfei Wang |
Institution: | The University of Texas Health Science Center at Houston |
Department: | School of biomedical informatics |
Country: | |
Proposed Analysis: | We will 1) use a multi-view-based classification algorithm to access the volume changes on the segmented ROIs of AD patients’ brains; 2) develop a Long Short-Term Memory (LSTM) model to trace the developments of AD by the high-level presentations of the ROIs in the previous step. This will help us to identify significant image biomarkers from AD brain imaging and classify AD stages using bioinformatics approaches. Then we would like to predict AD progression based on the clinical and imaging features of the AD patients. We will develop computational approaches to estimate sub-region tissue biochemical properties from AD imaging data; to design an elastic material model to estimate the regional change of hippocampal tissue during AD progression, extract biomechanical properties using the nonlinear finite element models (FEM); and also to develop a novel biomechanical property-based machine learning approach to model the relationships between AD progression, age, biomechanical property, cerebral tissue volume, and cognitive ability. |
Additional Investigators | |
Investigator's Name: | Guangming Zhang |
Proposed Analysis: | We will design an elastic material model for cerebral tissue based on the regional change in AD progression, and extract biomechanical properties using the nonlinear FEM models; We will develop a novel biomechanical property-based machine learning approach to model the relationships between AD progression, age, biomechanical property, cerebral tissue volume, and cognitive ability |
Investigator's Name: | Lei You |
Proposed Analysis: | We will use 3D rendering to increase the amount of training images with limited sagittal CSO cases; adopt transfer-learning to fine-tune the pre-trained deep neural networks; classify the sagittal CSO cases by utilizing the predictions of multiple rendered images of each case. |