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: | keee luuu |
| Institution: | ynu |
| Department: | stat |
| Country: | |
| Proposed Analysis: | There is currently no clear treatment for Alzheimer's disease, but if diagnosed and treated with intervention at an early stage, the progression of the disease can be slowed. Intelligent diagnostic methods based on medical imaging data are therefore important for the prevention and treatment of Alzheimer's disease. Traditional machine learning algorithms for medical imaging data usually consist of two modules: feature extraction and pattern classification. As medical image data is usually high-dimensional data and the data structure is complex. When dealing with such high-dimensional data, traditional algorithms usually need to perform vectorisation operations on the data, and this will lose the spatial structure information between brain region data. A tensor is a higher-order generalisation of a matrix that can directly represent this type of high-dimensional data. Thus, tensor is an important tool for representing medical imaging data. I wish to investigate a discriminant analysis method based on the tensor t-SVD, which will be accelerated by the Discrete Cosine Transform (DCT) due to the influence of the tensor product in the computation process. The tensor t-SVD discriminant analysis combined with the K-Nearest Neighbor (KNN) algorithm is finally used in the analysis of human brain fMRI datasets, with the expectation of achieving better classification prediction results. |
| Additional Investigators |
