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: | Leonel Mera |
Institution: | University of Antioquia |
Department: | Bioengineering |
Country: | |
Proposed Analysis: | Although there are currently treatments to mitigate Parkinson's disease, they depend on early differential diagnosis. Most of them are subject to clinical criteria established by expert medical evaluations and the error rate can be up to 24% in specialized medical centers. In fact, many factors can lead to a misdiagnosis such as essential tremor, drug-induced parkinsonism, and vascular parkinsonism. However, the integration of tools such as magnetic resonance imaging has made it possible to rule out other pathologies that cause parkinsonism. Furthermore, the development of high-field resonators is giving more relevance to this technique. This is even being considered in the search for a biomarker for Parkinson's progression. On the other hand, artificial intelligence has made it possible to automate and tackle problems as complex as pattern recognition, predictions, and classifications. In fact, one of the areas that have grown the most is artificial neural networks, which have even surpassed human performance. With the multilayer perceptron, point data or characteristics can be integrated. Convolutional neural networks allow images to be included and even time series can be used with recurrent networks. Furthermore, their generalization and their flexibility allow not only to work with data and images individually but also with multimodal structures where the different types of data can be integrated. Based on the previous antecedents, the research question arises: is it possible to develop a machine-learning algorithm to create an automatic tool in the differential diagnosis of Parkinson's disease? Also, could the features and algorithms be used to generate a predictor of Parkinson's progress status? |
Additional Investigators |