Intelligent IoT Anklets for Monitoring the Assessment of Parkinson’s Diseases

Parkinson’s disease (PD) is one of the fastest-growing neurological diseases in the world, characterized by impaired speech and walking abilities in patients. Currently, doctors usually assess the severity of PD based on these indicators. However, these methods have problems such as vague diagnostic criteria, low quantification, and poor accuracy. Therefore, how quantifying and assessing PD and its severity level by accurately and timely measuring the gait characteristics of patients is a great challenge. This article aims to develop a portable wearable smart anklet that can identify and grade PD patients by quantitatively measuring their gait characteristics. The method of this article mainly includes three aspects: first, an Internet of Things anklet (
$19\times 26\times6$
mm) is designed and manufactured to measure the three-axis acceleration and three-axis angular velocity at the patient’s ankle; second, the gait data of 40 PD patients (ten-grade one patient, 16-grade two patients, and 14-grade three patients) are collected and 150 gait features are extracted; finally, feature engineering and optimized K-nearest neighbor (KNN) multifeature classification algorithm are used to detect and accurately classify the abnormal gait patterns of PD patients. The results of this article show that the method can effectively quantify the degree of illness of the patient’s gait in nine parameters and achieve a classification accuracy of 96.5%, significantly better than other existing methods based on gait measurement. This novel, convenient, and efficient PD diagnosis auxiliary tool will provide strong support for the formulation of personalized treatment plans for PD patients.