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: | Jingyi Chi |
| Institution: | University of Pennsylvania |
| Department: | Computer Science |
| Country: | |
| Proposed Analysis: | A lot of the current problems in the medical domain are related to imaging. One of these problems is the generation of CT (Computed Tomography) images, since CT imaging requires long exposure to radiation which may harm the patients in terms of unexpected side effects. MRI is a comparatively safer method to produce images of inside the body. This gives us the motivation to work on development of methods for reliable estimation of CT images form corresponding MR images of the same subject. Some of the previous approaches used to solve the problem require heavy feature engineering, ensemble machine learning methods, and some deep learning methods like CNN, FCN to lead an end-to-end nonlinear mapping form MRI to CT. fHowever, with the advent of GANs and adversarial techniques, the same can be used for the task and to produce more realistic images. Further Auto-Context Model can be applied to implement a context-aware generative adversarial network for the task. Addition, methods to sharpen the output image to get better results can also be applied as an extension to the project. |
| Additional Investigators |
