Parkinson’s Disease Detection using Imputed Multimodal Datasets

Parkinson's disease is a progressive nervous system disorder that affects movement, often including tremors and work has been done on developing systems for early-stage detection of Parkinson’s disease. During our literature survey we observed that there are many available datasets, but most are small and often have multiple samples from the same subject. Either the current state-of-the-art models assume the source of the multi-modality to be the same, or they lack deployability in low-income areas, because of their dependence on means of data-collection which require expensive equipment. In our goal to build a more generalizable hybrid model, we decided to tackle this problem. We propose utilize information from 3 datasets available in the public domain.