Analysis of surface EMG signal morphology in Parkinson's disease

Diagnosis and Treatment of Tremor in Parkinson's Disease Using Mechanical Devices

BACKGROUND

Parkinsonian tremors are sometimes confused with essential tremors or other conditions. Recently, researchers conducted several studies on tremor evaluation using wearable sensors and devices, which may support accurate diagnosis. Mechanical devices are also commonly used to treat tremors and have been actively researched and developed. Here, we aimed to review recent progress and the efficacy of the devices related to Parkinsonian tremors.

METHODS

The PubMed and Scopus databases were searched for articles. We searched for "Parkinson disease" and "tremor" and "device".

RESULTS

Eighty-six articles were selected by our systematic approach. Many studies demonstrated that the diagnosis and evaluation of tremors in patients with PD can be done accurately by machine learning algorithms. Mechanical devices for tremor suppression include deep brain stimulation (DBS), electrical muscle stimulation, and orthosis. In recent years, adaptive DBS and optimization of stimulation parameters have been studied to further improve treatment efficacy.

CONCLUSIONS

Due to developments using state-of-the-art techniques, effectiveness in diagnosing and evaluating tremor and suppressing it using these devices is satisfactorily high in many studies. However, other than DBS, no devices are in practical use. To acquire high-level evidence, large-scale studies and randomized controlled trials are needed for these devices.

Using Deep Learning for Task and Tremor Type Classification in People with Parkinson's Disease

Hand tremor is one of the dominating symptoms of Parkinson's disease (PD), which significantly limits activities of daily living. Along with medications, wearable devices have been proposed to suppress tremor. However, suppressing tremor without interfering with voluntary motion remains challenging and improvements are needed. The main goal of this work was to design algorithms for the automatic identification of the tremor type and voluntary motions, using only surface electromyography (sEMG) data. Towards this goal, a bidirectional long short-term memory (BiLSTM) algorithm was implemented that uses sEMG data to identify the motion and tremor type of people living with PD when performing a task. Moreover, in order to automate the training process, hyperparamter selection was performed using a regularized evolutionary algorithm. The results show that the accuracy of task classification among 15 people living with PD was 84±8%, and the accuracy of tremor classification was 88±5%. Both models performed significantly above chance levels (20% and 33% for task and tremor classification, respectively). Thus, it was concluded that the trained models, based on using purely sEMG signals, could successfully identify the task and tremor types.

Changes in elbow flexion EMG morphology during adjustment of deep brain stimulator in advanced Parkinson’s disease

Objective

Deep brain stimulation (DBS) is an effective treatment for motor symptoms of advanced Parkinson’s disease (PD). Currently, DBS programming outcome is based on a clinical assessment. In an optimal situation, an objectively measurable feature would assist the operator to select the appropriate settings for DBS. Surface electromyographic (EMG) measurements have been used to characterise the motor symptoms of PD with good results; with proper methodology, these measurements could be used as an aid to program DBS.

Methods

Muscle activation measurements were performed for 13 patients who had advanced PD and were treated with DBS. The DBS pulse voltage, frequency, and width were changed during the measurements. The measured EMG signals were analysed with parameters that characterise the EMG signal morphology, and the results were compared to the clinical outcome of the adjustment.

Results

The EMG signal correlation dimension, recurrence rate, and kurtosis changed significantly when the DBS settings were changed. DBS adjustment affected the signal recurrence rate the most. Relative to the optimal settings, increased recurrence rates (median ± IQR) 1.1 ± 0.5 (−0.3 V), 1.3 ± 1.1 (+0.3 V), 1.7 ± 0.4 (−30 Hz), 1.7 ± 0.8 (+30 Hz), 2.0 ± 1.7 (+30 μs), and 1.5 ± 1.1 (DBS off) were observed. With optimal stimulation settings, the patients’ Unified Parkinson’s Disease Rating Scale motor part (UPDRS-III) score decreased by 35% on average compared to turning the device off. However, the changes in UPRDS-III arm tremor and rigidity scores did not differ significantly in any settings compared to the optimal stimulation settings.

Conclusion

Adjustment of DBS treatment alters the muscle activation patterns in PD patients. The changes in the muscle activation patterns can be observed with EMG, and the parameters calculated from the signals differ between optimal and non-optimal settings of DBS. This provides a possibility for using the EMG-based measurement to aid the clinicians to adjust the DBS.

Wearable sensors during drawing tasks to measure the severity of essential tremor

Commonly used methods to assess the severity of essential tremor (ET) are based on clinical observation and lack objectivity. This study proposes the use of wearable accelerometer sensors for the quantitative assessment of ET. Acceleration data was recorded by inertial measurement unit (IMU) sensors during sketching of Archimedes spirals in 17 ET participants and 18 healthy controls. IMUs were placed at three points (dorsum of hand, posterior forearm, posterior upper arm) of each participant's dominant arm. Movement disorder neurologists who were blinded to clinical information scored ET patients on the Fahn-Tolosa-Marin rating scale (FTM) and conducted phenotyping according to the recent Consensus Statement on the Classification of Tremors. The ratio of power spectral density of acceleration data in 4-12 Hz to 0.5-4 Hz bands and the total duration of the action were inputs to a support vector machine that was trained to classify the ET subtype. Regression analysis was performed to determine the relationship of acceleration and temporal data with the FTM scores. The results show that the sensor located on the forearm had the best classification and regression results, with accuracy of 85.71% for binary classification of ET versus control. There was a moderate to good correlation (r² = 0.561) between FTM and a combination of power spectral density ratio and task time. However, the system could not accurately differentiate ET phenotypes according to the Consensus classification scheme. Potential applications of machine-based assessment of ET using wearable sensors include clinical trials and remote monitoring of patients.

Technology-based therapy-response and prognostic biomarkers in a prospective study of a de novo Parkinson's disease cohort

Early noninvasive reliable biomarkers are among the major unmet needs in Parkinson's disease (PD) to monitor therapy response and disease progression. Objective measures of motor performances could allow phenotyping of subtle, undetectable, early stage motor impairments of PD patients. This work aims at identifying prognostic biomarkers in newly diagnosed PD patients and quantifying therapy-response. Forty de novo PD patients underwent clinical and technology-based kinematic assessments performing motor tasks (MDS-UPDRS part III) to assess tremor, bradykinesia, gait, and postural stability (T0). A visit after 6 months (T1) and a clinical and kinematic assessment after 12 months (T2) where scheduled. A clinical follow-up was provided between 30 and 36 months after the diagnosis (T3). We performed an ANOVA for repeated measures to compare patients' kinematic features at baseline and at T2 to assess therapy response. Pearson correlation test was run between baseline kinematic features and UPDRS III score variation between T0 and T3, to select candidate kinematic prognostic biomarkers. A multiple linear regression model was created to predict the long-term motor outcome using T0 kinematic measures. All motor tasks significantly improved after the dopamine replacement therapy. A significant correlation was found between UPDRS scores variation and some baseline bradykinesia (toe tapping amplitude decrement, p = 0.009) and gait features (velocity of arms and legs, sit-to-stand time, p = 0.007; p = 0.009; p = 0.01, respectively). A linear regression model including four baseline kinematic features could significantly predict the motor outcome (p = 0.000214). Technology-based objective measures represent possible early and reproducible therapy-response and prognostic biomarkers.

A Statistical Method for Exploratory Data Analysis Based on 2D and 3D Area under Curve Diagrams: Parkinson's Disease Investigation

A statistical method for exploratory data analysis based on 2D and 3D area under curve (AUC) diagrams was developed. The method was designed to analyze electroencephalogram (EEG), electromyogram (EMG), and tremorogram data collected from patients with Parkinson's disease. The idea of the method of wave train electrical activity analysis is that we consider the biomedical signal as a combination of the wave trains. The wave train is the increase in the power spectral density of the signal localized in time, frequency, and space. We detect the wave trains as the local maxima in the wavelet spectrograms. We do not consider wave trains as a special kind of signal. The wave train analysis method is different from standard signal analysis methods such as Fourier analysis and wavelet analysis in the following way. Existing methods for analyzing EEG, EMG, and tremor signals, such as wavelet analysis, focus on local time-frequency changes in the signal and therefore do not reveal the generalized properties of the signal. Other methods such as standard Fourier analysis ignore the local time-frequency changes in the characteristics of the signal and, consequently, lose a large amount of information that existed in the signal. The method of wave train electrical activity analysis resolves the contradiction between these two approaches because it addresses the generalized characteristics of the biomedical signal based on local time-frequency changes in the signal. We investigate the following wave train parameters: wave train central frequency, wave train maximal power spectral density, wave train duration in periods, and wave train bandwidth. We have developed special graphical diagrams, named AUC diagrams, to determine what wave trains are characteristic of neurodegenerative diseases. In this paper, we consider the following types of AUC diagrams: 2D and 3D diagrams. The technique of working with AUC diagrams is illustrated by examples of analysis of EMG in patients with Parkinson's disease and healthy volunteers. It is demonstrated that new regularities useful for the high-accuracy diagnosis of Parkinson's disease can be revealed using the method of analyzing the wave train electrical activity and AUC diagrams.

A-WEAR Bracelet for Detection of Hand Tremor and Bradykinesia in Parkinson's Patients

Parkinson's disease patients face numerous motor symptoms that eventually make their life different from those of normal healthy controls. Out of these motor symptoms, tremor and bradykinesia, are relatively prevalent in all stages of this disease. The assessment of these symptoms is usually performed by traditional methods where the accuracy of results is still an open question. This research proposed a solution for an objective assessment of tremor and bradykinesia in subjects with PD (10 older adults aged greater than 60 years with tremor and 10 older adults aged greater than 60 years with bradykinesia) and 20 healthy older adults aged greater than 60 years. Physical movements were recorded by means of an AWEAR bracelet developed using inertial sensors, i.e., 3D accelerometer and gyroscope. Participants performed upper extremities motor activities as adopted by neurologists during the clinical assessment based on Unified Parkinson's Disease Rating Scale (UPDRS). For discriminating the patients from healthy controls, temporal and spectral features were extracted, out of which non-linear temporal and spectral features show greater difference. Both supervised and unsupervised machine learning classifiers provide good results. Out of 40 individuals, neural net clustering discriminated 34 individuals in correct classes, while the KNN approach discriminated 91.7% accurately. In a clinical environment, the doctor can use the device to comprehend the tremor and bradykinesia of patients quickly and with higher accuracy.

Assessment of Motor Impairments in Early Untreated Parkinson's Disease Patients: The Wearable Electronics Impact

OBJECTIVE

The complex nature of Parkinson's disease (PD) makes difficult to rate its severity, mainly based on the visual inspection of motor impairments. Wearable sensors have been demonstrated to help overcoming such a difficulty, by providing objective measures of motor abnormalities. However, up to now, those sensors have been used on advanced PD patients with evident motor impairment. As a novelty, here we report the impact of wearable sensors in the evaluation of motor abnormalities in newly diagnosed, untreated, namely de novo, patients.

METHODS

A network of wearable sensors was used to measure motor capabilities, in 30 de novo PD patients and 30 healthy subjects, while performing five motor tasks. Measurement data were used to determine motor features useful to highlight impairments and were compared with the corresponding clinical scores. Three classifiers were used to differentiate PD from healthy subjects.

RESULTS

Motor features gathered from wearable sensors showed a high degree of significance in discriminating the early untreated de novo PD patients from the healthy subjects, with 95% accuracy. The rates of severity obtained from the measured features are partially in agreement with the clinical scores, with some highlighted, though justified, exceptions.

CONCLUSION

Our findings support the feasibility of adopting wearable sensors in the detection of motor anomalies in early, untreated, PD patients.

SIGNIFICANCE

This work demonstrates that subtle motor impairments, occurring in de novo patients, can be evidenced by means of wearable sensors, providing clinicians with instrumental tools as suitable supports for early diagnosis, and subsequent management.

Automatic Classification of Tremor Severity in Parkinson's Disease Using a Wearable Device

Although there is clinical demand for new technology that can accurately measure Parkinsonian tremors, automatic scoring of Parkinsonian tremors using machine-learning approaches has not yet been employed. This study aims to fill this gap by proposing machine-learning algorithms as a way to predict the Unified Parkinson’s Disease Rating Scale (UPDRS), which are similar to how neurologists rate scores in actual clinical practice. In this study, the tremor signals of 85 patients with Parkinson’s disease (PD) were measured using a wrist-watch-type wearable device consisting of an accelerometer and a gyroscope. The displacement and angle signals were calculated from the measured acceleration and angular velocity, and the acceleration, angular velocity, displacement, and angle signals were used for analysis. Nineteen features were extracted from each signal, and the pairwise correlation strategy was used to reduce the number of feature dimensions. With the selected features, a decision tree (DT), support vector machine (SVM), discriminant analysis (DA), random forest (RF), and k-nearest-neighbor (kNN) algorithm were explored for automatic scoring of the Parkinsonian tremor severity. The performance of the employed classifiers was analyzed using accuracy, recall, and precision, and compared to other findings in similar studies. Finally, the limitations and plans for further study are discussed.

Reliability of quadriceps surface electromyography measurements is improved by two vs. single site recordings

Purpose

The reliability of surface electromyography (sEMG) is typically modest even with rigorous methods, and therefore further improvements in sEMG reliability are desirable. This study compared the between-session reliability (both within participant absolute reliability and between-participant relative reliability) of sEMG amplitude from single vs. average of two distinct recording sites, for individual muscle (IM) and whole quadriceps (WQ) measures during voluntary and evoked contractions.

Methods

Healthy males (n = 20) performed unilateral isometric knee extension contractions: voluntary maximum and submaximum (60%), as well as evoked twitch contractions on two separate days. sEMG was recorded from two distinct sites on each superficial quadriceps muscle.

Results

Averaging two recording sites vs. using single site measures improved reliability for IM and WQ measurements during voluntary (16–26% reduction in within-participant coefficient of variation, CV_W) and evoked contractions (40–56% reduction in CV_W).

Conclusions

For sEMG measurements from large muscles, averaging the recording of two distinct sites is recommended as it improves within-participant reliability. This improved sensitivity has application to clinical and research measurement of sEMG amplitude.

Assessing Motor Fluctuations in Parkinson's Disease Patients Based on a Single Inertial Sensor

Altered movement control is typically the first noticeable symptom manifested by Parkinson's disease (PD) patients. Once under treatment, the effect of the medication is very patent and patients often recover correct movement control over several hours. Nonetheless, as the disease advances, patients present motor complications. Obtaining precise information on the long-term evolution of these motor complications and their short-term fluctuations is crucial to provide optimal therapy to PD patients and to properly measure the outcome of clinical trials. This paper presents an algorithm based on the accelerometer signals provided by a waist sensor that has been validated in the automatic assessment of patient's motor fluctuations (ON and OFF motor states) during their activities of daily living. A total of 15 patients have participated in the experiments in ambulatory conditions during 1 to 3 days. The state recognised by the algorithm and the motor state annotated by patients in standard diaries are contrasted. Results show that the average specificity and sensitivity are higher than 90%, while their values are higher than 80% of all patients, thereby showing that PD motor status is able to be monitored through a single sensor during daily life of patients in a precise and objective way.

Neuroelectronics and Biooptics: Closed-Loop Technologies in Neurological Disorders

Brain-implanted devices are no longer a futuristic idea. Traditionally, therapies for most neurological disorders are adjusted based on changes in clinical symptoms and diagnostic measures observed over time. These therapies are commonly pharmacological or surgical, requiring continuous or irreversible treatment regimens that cannot respond rapidly to fluctuations of symptoms or isolated episodes of dysfunction. In contrast, closed-loop systems provide intervention only when needed by detecting abnormal neurological signals and modulating them with instantaneous feedback. Closed-loop systems have been applied to several neurological conditions (most notably epilepsy and movement disorders), but widespread use is limited by conceptual and technical challenges. Herein, we discuss how advances in experimental closed-loop systems hold promise for improved clinical benefit in patients with neurological disorders.

A portable system for quantitative assessment of parkinsonian rigidity

Rigidity is one of the primary symptoms of Parkinson's disease. Passive flexion and extension of the elbow is used to assess rigidity in this study. An examiner flexes and extends the subject's elbow joint through a rigidity assessment cuff attached around the wrist. Each assessment lasts for 10 seconds. Two force sensor boxes and an inertial measurement unit are used to measure the applied force and the state of the elbow movement. Elastic and viscous values will be obtained through a least squares estimation with all the data. 9 healthy subjects were tested with this system in two experimental conditions: 1) normal state (relaxed); 2) imitated rigidity state. Also the subjects were performed the assessment task with different frequencies and elbow movement ranges. The imitated rigidity action increases viscosity and elasticity. The effect sizes (Cohen's d) of the viscosity and elasticity between normal state and imitated state are 1.61 and 1.36 respectively, which means the difference is significant. Thus, this system can detect the on-off fluctuations of parkinsonian rigidity. Both wrist movement angle and frequency have small effect on the viscosity, but have elevated effect on the elasticity.

Technologies for Assessment of Motor Disorders in Parkinson's Disease: A Review

Parkinson’s Disease (PD) is characterized as the commonest neurodegenerative illness that gradually degenerates the central nervous system. The goal of this review is to come out with a summary of the recent progress of numerous forms of sensors and systems that are related to diagnosis of PD in the past decades. The paper reviews the substantial researches on the application of technological tools (objective techniques) in the PD field applying different types of sensors proposed by previous researchers. In addition, this also includes the use of clinical tools (subjective techniques) for PD assessments, for instance, patient self-reports, patient diaries and the international gold standard reference scale, Unified Parkinson Disease Rating Scale (UPDRS). Comparative studies and critical descriptions of these approaches have been highlighted in this paper, giving an insight on the current state of the art. It is followed by explaining the merits of the multiple sensor fusion platform compared to single sensor platform for better monitoring progression of PD, and ends with thoughts about the future direction towards the need of multimodal sensor integration platform for the assessment of PD.

Non-linear dynamics in parkinsonism

Over the last 30 years, the functions (and dysfunctions) of the sensory-motor circuitry have been mostly conceptualized using linear modelizations which have resulted in two main models: the “rate hypothesis” and the “oscillatory hypothesis.” In these two models, the basal ganglia data stream is envisaged as a random temporal combination of independent simple patterns issued from its probability distribution of interval interspikes or its spectrum of frequencies respectively. More recently, non-linear analyses have been introduced in the modelization of motor circuitry activities, and they have provided evidences that complex temporal organizations exist in basal ganglia neuronal activities. Regarding movement disorders, these complex temporal organizations in the basal ganglia data stream differ between conditions (i.e., parkinsonism, dyskinesia, healthy control) and are responsive to treatments (i.e., l-DOPA, deep brain stimulation). A body of evidence has reported that basal ganglia neuronal entropy (a marker for complexity/irregularity in time series) is higher in hypokinetic state. In line with these findings, an entropy-based model has been recently formulated to introduce basal ganglia entropy as a marker for the alteration of motor processing and a factor of motor inhibition. Importantly, non-linear features have also been identified as a marker of condition and/or treatment effects in brain global signals (EEG), muscular activities (EMG), or kinetic of motor symptoms (tremor, gait) of patients with movement disorders. It is therefore warranted that the non-linear dynamics of motor circuitry will contribute to a better understanding of the neuronal dysfunctions underlying the spectrum of parkinsonian motor symptoms including tremor, rigidity, and hypokinesia.

Parkinson disease and exercise

Parkinson disease (PD) is a progressive, neurodegenerative movement disorder. PD was originally attributed to neuronal loss within the substantia nigra pars compacta, and a concomitant loss of dopamine. PD is now thought to be a multisystem disorder that involves not only the dopaminergic system, but other neurotransmitter systems whose role may become more prominent as the disease progresses (189). PD is characterized by four cardinal symptoms, resting tremor, rigidity, bradykinesia, and postural instability, all of which are motor. However, PD also may include any combination of a myriad of nonmotor symptoms (195). Both motor and nonmotor symptoms may impact the ability of those with PD to participate in exercise and/or impact the effects of that exercise on those with PD. This article provides a comprehensive overview of PD, its symptoms and progression, and current treatments for PD. Among these treatments, exercise is currently at the forefront. People with PD retain the ability to participate in many forms of exercise and generally respond to exercise interventions similarly to age-matched subjects without PD. As such, exercise is currently an area receiving substantial research attention as investigators seek interventions that may modify the progression of the disease, perhaps through neuroprotective mechanisms.

EMG signal morphology in essential tremor and Parkinson's disease

The aim of this work was to differentiate patients with essential tremor from patients with Parkinson's disease. The electromyographic signal from the biceps brachii muscle was measured during isometric tension from 17 patients with essential tremor, 35 patients with Parkinson's disease, and 40 healthy controls. The EMG signals were high pass filtered and divided to smaller segments from which histograms were calculated using 200 histogram bins. EMG signal histogram shape was analysed with a feature dimension reduction method, the principal component analysis, and the shape parameters were used to differentiate between different patient groups. The height of the histogram and the side difference between left and right hand were the best discriminators between essential tremor and Parkinson's disease groups. With this method, it was possible to discriminate 13/17 patients with essential tremor from 26/35 patients with Parkinson's disease and 14/17 patients with essential tremor from 29/40 healthy controls.

Analysis of EMG and acceleration signals for quantifying the effects of deep brain stimulation in Parkinson's disease

Deep brain stimulation (DBS) is effective in reducing motor symptoms in Parkinson's disease (PD). However, objective methods for quantifying its efficacy are lacking. We present a principal component (PC)-based tracking method for quantifying the effects of DBS in PD by using electromyography (EMG) and acceleration measurements. Ten parameters capturing PD characteristic signal features were initially extracted from isometric EMG and acceleration recordings. Using a PC approach, the original parameters were transformed into a smaller number of PCs. Finally, the effects of DBS were quantified by examining the PCs in a low-dimensional feature space. The EMG and acceleration data from 13 PD patients with DBS ON and OFF, and 13 healthy age-matched controls were used for analysis. Clinical evaluation of patients showed that their motor symptoms were effectively reduced with DBS. The analysis results showed that the signal characteristics of 12 patients were more similar to those of the healthy controls with DBS ON than with DBS OFF. These observations indicate that the PC-based tracking method can be used to objectively quantify the effects of DBS on the neuromuscular function of PD patients. Further studies are suggested to estimate the clinical sensitivity of the method to different types of PD.

An EMG-based system for continuous monitoring of clinical efficacy of Parkinson's disease treatments

Current methods for assessing the efficacy of treatments for Parkinson's disease (PD) rely on physician rated scores. These methods pose three major shortcomings: 1) the subjectivity of the assessments, 2) the lack of precision on the rating scale (6 discrete levels), and 3) the inability to assess symptoms except under very specific conditions and/or for very specific tasks. To address these shortcomings, a portable system was developed to continuously monitor Parkinsonian symptoms with quantitative measures based on electrical signals from muscle activity (EMG). Here, we present the system design and the implementation of methods for system validation. This system was designed to provide continuous measures of tremor, rigidity, and bradykinesia which are related to the neurophysiological source without the need for multiple bulky experimental apparatuses, thus allowing more precise, quantitative indicators of the symptoms which can be measured during practical daily living tasks. This measurement system has the potential to improve the diagnosis of PD as well as the evaluation of PD treatments, which is an important step in the path to improving PD treatments.

Multi-joint movements with reversal in Parkinson's disease: Kinematics and electromyography

Subjects with Parkinson's disease (PD) presented difficulties in the performance of multi-joint movements. The purpose of the study was to determine whether the slowness of such movements was caused by the generation of non-linear trajectories and/or by a reduction or a deficit in the modulation of EMG activity. Nine healthy subjects and 10 subjects with PD performed multi-joint movements involving elbow and shoulder with reversal towards three targets in the sagittal plane without any constraint. The movement kinematics were calculated using X and Y coordinates of the markers positioned on the joints. EMG signals were recorded for the muscles related to these movements. The results revealed that subjects with PD presented a lower linear speed and the differences between them and healthy subjects increased with target distance. The trajectory was found to be linear and both groups of subjects had few errors in the targets despite the slower muscle activity in subjects with PD. Another interesting finding was the EMG pattern of subjects with PD. They showed a difficulty in modulating the activity of agonists and antagonists during the different movement phases. The low speed movements of PD subjects were attributable to the low EMG activity and difficulty in modulating the bursts of muscle activity.

Variability of EMG patterns: a potential neurophysiological marker of Parkinson's disease?

OBJECTIVE

This study evaluated whether changes in the electromygraphic (EMG) pattern during rapid point-to-point movements in individuals diagnosed with PD can: (1) distinguish PD subjects from healthy subjects and (2) determine if differences in the EMG pattern reflect disease severity in PD.

METHODS

Three groups of 10 PD subjects and 10 age/sex-matched healthy subjects performed rapid 72 degree point-to-point elbow flexion movements. PD subjects were divided, a priori, into three groups based upon off medication motor UPDRS score.

RESULTS

Measures related to the EMG pattern distinguished all PD subjects and 9 out of 10 healthy subjects, resulting in 100% sensitivity. Further, significant correlations were shown between EMG measures and the motor UPDRS score. After 30 months, the one healthy subject whose EMG pattern was abnormal was reexamined. The EMG measures remained abnormal and the motor UPDRS score went from 0 to 10. Parkinson's disease was diagnosed.

CONCLUSION

Measures related to the variability of the EMG pattern during rapid point-to-point movements provide neurophysiological measures that objectively distinguish PD subjects from healthy subjects. These measures also correlate with disease severity.

SIGNIFICANCE

EMG measures may provide a non-invasive measure that is sensitive and specific for identifying individuals with PD.

Surface EMG and acceleration signals in Parkinson's disease: feature extraction and cluster analysis

We present an advanced method for feature extraction and cluster analysis of surface electromyograms (EMG) and acceleration signals in Parkinson's disease (PD). In the method, 12 different EMG and acceleration signal features are extracted and used to form high-dimensional feature vectors. The dimensionality of these vectors is then reduced by using the principal component approach. Finally, the cluster analysis of feature vectors is performed in a low-dimensional eigenspace. The method was tested with EMG and acceleration data of 42 patients with PD and 59 healthy controls. The obtained discrimination between patients and controls was promising. According to clustering results, one cluster contained 90% of the healthy controls and two other clusters 76% of the patients. Seven patients with severe motor dysfunctions were distinguished in one of the patient clusters. In the future, the clinical value of the method should be further evaluated.