Parkinson’s disease (PD) is a major global health concern due to its debilitating motor symptoms, particularly tremors. This pilot study presents an interpretable Internet of Medical Things (IoMT)-based system for monitoring hand tremors in PD patients, integrating a 6-axis inertial sensor for motion tracking, a low-power wireless microcontroller for data acquisition, and a LightGBM gradient-boosting model for tremor classification.
During acquisition, triaxial accelerometer and gyroscope signals are sampled at 66.67 Hz, covering the characteristic 4–6 Hz Parkinsonian tremor band. Data is transmitted via a RESTful API over Wi-Fi and stored in a relational database. Signal processing includes noise reduction, temporal segmentation, and frequency-domain feature extraction using FFT, after which LightGBM classifies motor activity as Parkinsonian tremor, voluntary movement, or no tremor. A web-based interface further enables real-time visualization and clinical data entry.
As a feasibility study, the system achieved strong performance—95.64% accuracy and an F1-score of 0.95—on 10,314 samples from nine participants (four with PD, five healthy controls), supporting its potential as a low-cost, interpretable tool for tremor monitoring.
Autor(es):Toque Erick
Gutiérrez-Flores, César
Abarca, Victoria E.
Vila, Sebastian
Silva-Salas, Rosa M.
Elias, Dante A.
Año: 2026
Título de la revista: Results in Engineering
Volumen: 29
Url: https://doi.org/10.1016/j.rineng.2026.109608