Solving The Maze of Diagnosing Parkinson’s Disease based on Portable EEG Sensing to be Adaptable to Go In-The-Wild

Solving The Maze of Diagnosing Parkinson’s Disease based on Portable EEG Sensing to be Adaptable to Go In-The-Wild

---

연구 분야: Strategies

---

---

학회: NSysS '20: Proceedings of the 7th International Conference on Networking, Systems and Security

---

초록

Parkinson’s disease is a common and highly threatening neurodegenerative disease, which has no confirmed well-adopted diagnosis method to date. All research efforts in this regard focus on diagnosing in controlled environment such as laboratories. Going beyond this common limitation, in this paper, we exploit the Electroencephalogram (EEG) signal to effectively come up with a portable Parkinson’s disease diagnosis system that can go in-the-wild. To do so, we collect EEG signals from both Parkinson’s patients and healthy people in different emotional states. Our statistical analysis finds a substantial differences in the signals in several emotional states for Parkinson’s patients and healthy people. Accordingly, we develop a machine learning based model to diagnose Parkinson’s disease that results in 97% accuracy.

---