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: | Diogo Shiraishi |
| Institution: | UNICAMP |
| Department: | Neurology |
| Country: | |
| Proposed Analysis: | The data will be used in test-retest evaluation of deep learning segmentation models. The temporary abstract of the study is pasted below. Spinocerebellar ataxias (SCAs) represent a heterogeneous group of neurodegenerative diseases that share cerebellar ataxia as a clinical symptom common. Such diseases do not have an approved treatment and, thus, several groups have been working on disease-modifying agents. However, there is a lack of variables responses sensitive enough to assess the real potential of these drugs. Considering that hereditary ataxias are rare and progress slowly, this imposes a serious limitation on these studies. On the other hand, studies based on images of magnetic resonance imaging have shown to be useful in the search for biomarkers. As the cerebellum is the main structure affected in SCAs, neuroimaging tools specialized in its evaluation have emerged over time, but due to the organ compactness and irregular nature, this task has proven to be challenging. In this sense, such methods are still not very accurate in terms of segmentation and need a lot of manual correction, which takes a lot of processing time. Faced with important limitations methodological aspects of the main methods of cerebellar segmentation, we propose to create a cerebellar segmentation method based on deep learning algorithms with two different networks, one for healthy subjects and one for patients with cerebellar damage, to increase the accuracy of the segmentation and decrease the time of image processing. |
| Additional Investigators |
