Background: Parkinson’s disease (PD) is the second most common neurodegenerative disease. Despite the awareness of PD for exactly two centuries, at present, there is no precise predictor for either onset or progression of disease thanks to the heterogeneity of PD. Scientists have spent tremendous efforts on the development or identification of new biomarkers, but none of them really find "The One", which fulfill its commitment as the prophet of PD. In order to breakthrough this deadlock, launching a new predicting model which integrates multiple parameters, including brand new biomarkers, novel and objective clinical assessment is essential and urgent. In the past, the contents of cerebrospinal fluid and peripheral blood are the main sources of biomarkers. However, obtaining cerebrospinal fluid is invasive whereas blood brain barrier blocks the communication between peripheral blood to brain parenchyma. Neural-derived exosomes isolated from peripheral blood are ideal for tackling both issues. Exosomes are small vesicles secreted from the cells and the content of exosomes are almost identical to cytoplasm, which enable researchers to investigate the neurons in remote. On the other hand, motor symptoms are still cardinal presentations of PD and the significance cannot be underestimated. However, the lack of objective and universal assessment limits the application as disease predictor. Recent study demonstrated that artificial intelligence (AI) assisted speech analysis accurately reflects the motor symptoms of PD, which introduces a new route for the assessment of PD. Hypothesis: This project hypothesizes that a novel prediction model of PD should base on assessments from multiple domains, which covers novel biomarkers and AI-assisted motor symptom analysis. Aims: This project aims to set up an integrated prediction model of PD for both the high-risk groups and patients. Materials and methods: This project will initiate three cohorts and all of the participants will be followed for 2 years. The first cohort is the discovery cohort, which recruits 40 PD patients to set up the prediction model for the progression of disease. The prediction will be based on the 1. the level of aggregated proteins in the neural exosomes, including α-synuclein, β-amyloid, tau and TDP-43 2. the speech analysis. The second cohort is the validation cohort, which recruits another 40 de novo PD patients and will be utilized to test the accuracy of the disease progression according to the prediction model. The third cohort will enroll people with rapid eye movement sleep behavioral disorders, who are recognized as high risk of PD according to previous literatures. The prediction model will be applied to the third cohort to test the capability of predicting onset of PD. Expecting Results: This integrated prediction model is able to predict 1.the progression of PD for diagnosed patients 2.the onset of PD for high risk people.
|Effective start/end date||8/1/18 → 7/31/19|
- Parkinson’s disease
- prediction model
- artificial intelligence
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.