Parkinson’s disease detection and monitoring system based on tremors and abnormal gait using Machine Learning

  • Unique Paper ID: 162981
  • Volume: 10
  • Issue: 11
  • PageNo: 541-545
  • Abstract:
  • This work presents a tremor and aberrant gait analysis-based Parkinson's disease (PD) identification and monitoring system. Using wearable device accelerometer and gyroscope data, the suggested approach uses feature extraction to capture minute movements that are linked to Parkinson's disease. In real-time applications, the K-Nearest Neighbors (KNN), Support-Vector Machine and Logistic regression classifier are implemented due to ease of use and effectiveness. The KNN algorithm demonstrates its effectiveness as a dependable diagnostic tool by achieving outstanding accuracy, having been trained on a dataset containing both PD and non-PD cases. Furthermore, the system serves as an uninterrupted tool that enables us to trace the advancement of a disease in real time. As a result, this makes it easier to modify treatment plans in a timely manner, which greatly enhances patient care thus providing overall health benefits. It serves as a non-invasive yet affordable option for early screening of Parkinson's disease. The lead time due to screening helps in early diagnosis and also provides individualized treatment plan which is a combination of wearable sensor technologies and machine learning.

Related Articles