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: | Hanrui Zhang |
Institution: | University of Michigan |
Department: | DCMB |
Country: | |
Proposed Analysis: | We propose to construct a machine learning model to tele-diagnose Alzheimer's disease from accelerometer records. Once we obtained data from ADNI, we will preprocess the data by filtering noise, handling missing data, and segmenting it into time windows. Focus on feature engineering to extract gait characteristics, postural control, activity levels, sleep patterns, and temporal patterns. Use feature selection techniques, such as LASSO or mutual information, to identify relevant features. Experiment with machine learning models, including logistic regression, support vector machines, and deep learning architectures. Use k-fold cross-validation for model selection and hyperparameter optimization with grid search or random search. Evaluate the final model on an independent test dataset, considering metrics like accuracy, sensitivity, specificity, F1 score, and AUC-ROC. Enhance interpretability with LIME or SHAP to understand the contribution of individual features to predictions. Deploy the model in a clinical setting, ensuring ethical considerations and data privacy regulations are met. Continuously monitor and update the model for accuracy and relevance as new data become available. |
Additional Investigators |