O uso de sensores inerciais para caracterização e classificação do tremor de punho em indivíduos com a doença de Parkinson e correlação com a escala de avaliação subjetiva: UPDRS

Abstract

Introduction: Human tremor is a clinical disorder characterized by involuntary movement resulting from contractions of antagonistic muscles. The physiological tremor is associated with natural processes and the pathological tremor associated with several factors, such as neurological dysfunctions. One possible cause of the disease is Parkinson's disease (DP). Among the clinical signs associated with DP, the present study focused on tremor. For an understanding of the tremor, there are subjective and objective methods. Subjective use clinical scales of severity, such as the Unified Parkinson's Disease Rating Scale (UPDRS). However, this scale depends on the experience and knowledge of the evaluator. Thus, studies suggest objective methods. These use inertial sensors, such as accelerometer, magnetometer and gyroscope, as they measure the activities of the evaluated member in real time. Objective: Use of inertial sensors for the characterization and classification of hand tremor in individuals with DP and correlation with the UPDRS, motor examination - part III. Methodology: This study was supported by the Ethics Committee on Research in Human Beings of the Federal University of Uberlândia (CEP: 270,782 and CEP: 2,001,535). Twenty-two subjects with DP participated in the study, allocated to only one group. The limb most affected by the tremor was submitted to activity. The protocol was: (i) reading and signing the informed consent form; completing an identification form, and a questionnaire assessing the severity of the disease, UPDRS; (ii) collection of activities: wrist at rest and in mild extension maintained (0o), without and with load of 92 g and 184 g. After the collection, the results were generated and the analyzed items were: (i) characteristics of the tremor signal by amplitude and frequency; (ii) different variables of the experimental protocol; (iii) different inertial sensors; (iv) at different times of the protocol. Statistical analysis was used to estimate the mean and standard deviation, and the Spearman's test, for the correlation. The analyses were performed using software R. p <0.05. Results: the equipment collected data objectively, characterized, and classified the tremor signs. There was a strong correlation between the characteristics of the signals between experimental variables and different protocols. The load used (92 and 184 g) did not have a significant effect. The characteristics, median of RMS and wE4 of the signal were those that presented a higher mean of correlation for the analysis. The accelerometer and gyroscope sensors presented a strong correlation for all the experimental variables. The magnetometer had a weak correlation. The two sensors (1 and 2) presented a strong correlation, thus, as their coordinates (x, y, z). Conclusion: The biomedical device, characterized and classified the cuff tremor in PD. Strong correlation between the characteristics of the signal, with the sensors, accelerometer and gyroscope, and with the UPDRS scale (motor examination, part III) was found. Strong correlations between the coordinates and the two sensors have shown that they can be used together, or separately, due to similarity and any are sensitive to capture information from the tremor. The charge added in this study (92 and 184 g) to obtain a weak correlation did not sensitize the signal. The method in question can replace only the motor examination - part III of the UPDRS and reproduced for other tremors, and used by similar research, by allows, easy manipulation of its aspects and parameters. Keywords: Inertial sensors. Biomedical signal processing. Tremor. Parkinson’s Disease. UPDRS.