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: | Dheenadhayalan Ramadoss |
| Institution: | Indian Institute of Information Technology, Kottayam |
| Department: | Computer Science |
| Country: | |
| Proposed Analysis: | We propose a brain MRI classification method to classify Mild Cognitive Impairment or early Alzheimer's Disease from Normal Cognitive by generating severity based labels to train Convolution Neural Network. Proposed method utilizes the MRI scans of a patient taken at different time of diagnosis and attempts to create a new labels for the dataset. New labels are generated by integrating the severity (this is found by comparing two MRI scans of a patient which are obtained at significantly spaced time interval) and actual labels of the dataset. A Convolution Neural Network is then trained using early MRI scan data and the generated labels. An increase in classification accuracy is expected since the CNN is trained on a fuzzy label, which contains the information of both presence and severity of the disease rather than the binary label which indicates only the presence of disease. |
| Additional Investigators | |
| Investigator's Name: | Carol Hargreaves |
| Proposed Analysis: | We propose a brain MRI classification method to classify Mild Cognitive Impairment or early Alzheimer's Disease from Normal Cognitive by generating severity based labels to train Convolution Neural Network. Proposed method utilizes the MRI scans of a patient taken at different time of diagnosis and attempts to create a new labels for the dataset. New labels are generated by integrating the severity (this is found by comparing two MRI scans of a patient which are obtained at significantly spaced time interval) and actual labels of the dataset. A Convolution Neural Network is then trained using early MRI scan data and the generated labels. An increase in classification accuracy is expected since the CNN is trained on a fuzzy label, which contains the information of both presence and severity of the disease rather than the binary label which indicates only the presence of disease. |
