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: | BO LING |
Institution: | University of Birmingham |
Department: | Computer Science School |
Country: | |
Proposed Analysis: | This is a summer project that is about to start, and the main research direction is medical image segmentation. In previous projects, I have tried to use U-net neural network to segment liver tumors, and have a preliminary grasp of related technologies. In this project, I will try an improved U-net neural network. To solve the problem that U-net is prone to gradient disappearance and low feature reuse rate, I will try to establish a dense connection between the current layer and all layers to maintain upsampling The long connection structure with downsampling can effectively alleviate the shortcomings of U-net. I have also learned about the data set provided by ADNI. For the small proportion of hippocampus, small size, and unclear edges, I will try to use data augmentation, size cropping and other methods to remove unnecessary edge samples. |
Additional Investigators |