Monitoring Of Tremor Research Articles

Spatial Distribution of Tremor Episodes From Long‐Term Monitoring in the Northern Cascadia Subduction Zone

AbstractLarge bursts of non‐volcanic tremor (“major” tremor episodes) correlated with geodetic deformation recur regularly in the Cascadia subduction zone and are often called episodic tremor and slip (ETS). Minor episodes of tremor between ETS are ubiquitous but have been understudied. This paper assesses time‐invariant characteristics of tremor episodes of all sizes within northern Cascadia. We derive a catalog of tremor episodes ranging in size from 10 to >13,000 tremor events using the results of 17 years of tremor monitoring. Minor episodes represent ∼96% of all 896 tremor episodes and their occurrence varies on 10‐km scales. Using estimates for the depth of the forearc Moho and subducting slab, we observe an association between the location of the forearc mantle corner (FMC) and tremor occurrence that leads to along‐dip modality. Bimodality, present in southern Washington and Vancouver Island, represents the segmentation of major and minor episodes up‐dip and down‐dip of the FMC, respectively. Unimodality, present in Puget Sound, results when the FMC is located near the down‐dip edge of the ETS zone and no segmentation occurs. We also use our extensive tremor episode catalog alongside three‐dimensional regional tomographic velocity models to reassess the relationship between tremor activity and low Vp/Vs signatures in the forearc. We do not find a correlation between tremor episode recurrence intervals and Vp/Vs, contrary to some previous work, suggesting that controls on silica precipitation in the forearc crust are not dominant controls of tremor episode recurrence, or that the association is not widely observable.

Continuous and Unconstrained Tremor Monitoring in Parkinson's Disease Using Supervised Machine Learning and Wearable Sensors.

Accurately assessing the severity and frequency of fluctuating motor symptoms is important at all stages of Parkinson's disease management. Contrarily to time-consuming clinical testing or patient self-reporting with uncertain reliability, recordings with wearable sensors show promise as a tool for continuously and objectively assessing PD symptoms. While wearables-based clinical assessments during standardised and scripted tasks have been successfully implemented, assessments during unconstrained activity remain a challenge. We developed and implemented a supervised machine learning algorithm, trained and tested on tremor scores. We evaluated the algorithm on a 67-hour database comprising sensor data and clinical tremor scores for 24 Parkinson patients at four extremities for periods of about 3 hours. A random 25% subset of the labelled samples was used as test data, the remainder as training data. Based on features extracted from the sensor data, a Support Vector Machine was trained to predict tremor severity. Due to the inherent imbalance in tremor scores, we applied dataset rebalancing techniques. Our classifier demonstrated robust performance in detecting tremor events with a sensitivity of 0.90 on the test-portion of the resampled dataset. The overall classification accuracy was high at 0.88. We implemented an accurate classifier for tremor monitoring in free-living environments that can be trained even with modestly sized and imbalanced datasets. This advancement offers significant clinical value in continuously monitoring Parkinson's disease symptoms beyond the hospital setting, paving the way for personalized management of PD, timely therapeutic adjustments, and improved patient quality of life.

ELectronic Archimedes spiral Neural Network (ELANN)

The Archimedes spiral is a clinical tool that aids in the diagnosis and monitoring of essential tremor. However, spiral ratings may vary based on experience and training of the rating physician. This study sought to generate an objective standard model for tremor evaluation using convolutional neural networks.One senior movement disorders neurologist (Neurologist 1) with over 30 years of clinical experience used the Bain and Findley Spirography Rating Scale to rate 1653 Archimedes spiral images from 46 essential tremor patients (mild to severe tremor) and 75 control subjects (no to mild tremor). Neurologist 1's labels were used as the reference standard to train the model. After training the model, a randomly selected subset of spiral testing data was re-evaluated by Neurologist 1, by a second senior movement disorders neurologist (Neurologist 2) with over 27 years of clinical experience, and by our model. Cohen's Weighted Kappa 95% confidence intervals were calculated from all rater comparisons to determine if our model performs with the same proficiency as two senior movement disorders neurologists.The Cohen's Weighted Kappa 95% confidence intervals for the agreement between the reference standard scores and Neurologist 1's rerated scores, for the agreement between the reference standard scores and Neurologist 2's scores, and for the agreement between the reference standard scores and our model's scores were 0.93–0.98, 0.86–0.94, and 0.89–0.96, respectively.With overlapping Cohen's Weighted Kappa 95% confidence intervals for all agreement comparisons, we demonstrate that our model evaluates spirals with the same proficiency as two senior movement disorders neurologists.

(Invited) 3D Morphable Electronic Systems Via Deterministic Microfolding

DNA and proteins fold in three dimensions (3D) to enable functions that sustain life. Emulation of such reversible folding schemes for functional materials can create promising opportunities for advancing a wide range of technologies. In particular, morphing 3D mesostructures via multiple dimensions for high-performance materials including monocrystalline silicon can enable unconventional designs in sensory robotics, biomedical devices, microelectronics, and microelectromechanical systems. Existing approaches to morphing functional materials mostly rely on planar processes, which often complicates needed mechanisms for structural reconfiguration and limits the choice of materials composition and dimension. This presentation will introduce a bioinspired microfolding strategy to realize 3D reconfigurable microelectronic systems in freestanding forms with various advanced materials and complex architectures. The microfolding mechanism allows access to any transitional states of the folded 3D structures in a reversible fashion, to modulate functionalities on demand. Demonstrations on microantennas for telecommunication, wearable accelerometers for tremor monitoring, and epicardial bioelectronic probes for cardiac mapping, all of which are achieved via the microfolding strategy, further highlight the high potential for a broad range of applications in healthcare and communications industry.

A Novel Accelerometry Method to Perioperatively Quantify Essential Tremor Based on Fahn–Tolosa–Marin Criteria

The disease status, progression, and treatment effect of essential tremor (ET) patients are currently assessed with clinical scores, such as the Fahn–Tolosa–Marin Clinical Rating Scale for Tremor (FTM). The use of objective and rater-independent monitoring of tremors may improve clinical care for patients with ET. Therefore, the focus of this study is to develop an objective accelerometry-based method to quantify ET, based on FTM criteria. Thirteen patients with ET and thirteen matched healthy participants underwent FTM tests to rate tremor severity, paired with tri-axial accelerometric measurements at the index fingers. Analogue FTM assessments were performed by four independent raters based on video recordings. Quantitative measures were derived from the accelerometric data, e.g., the area under the curve of power in the 4–8 Hz frequency band (AUCP) and maximal tremor amplitude. As such, accelerometric tremor scores were computed, using thresholds based on healthy measurements and FTM criteria. Agreement between accelerometric and clinical FTM scores was analyzed with Cohen’s kappa coefficient. It was assessed whether there was a relationship between mean FTM scores and the natural logarithm (ln) of the accelerometric outcome measures using linear regression. The agreement between accelerometric and FTM scores was substantial for resting and intention tremor tests (≥72.7%). However, the agreement between accelerometric postural tremor data and clinical FTM ratings (κ = 0.459) was low, although their logarithmic (ln) relationship was substantial (R2 ≥ 0.724). Accelerometric test–retest reliability was good to excellent (ICC ≥ 0.753). This pilot study shows that tremors can be quantified with accelerometry, using healthy thresholds and FTM criteria. The test–retest reliability of the accelerometric tremor scoring algorithm indicates that our low-cost accelerometry-based approach is a promising one. The proposed easy-to-use technology could diminish the rater dependency of FTM scores and enable physicians to monitor ET patients more objectively in clinical, intraoperative, and home settings.

Continuous Monitoring of Essential Tremor: Standards and Challenges.

The value of continuous symptom monitoring in people with essential tremor is uncertain. To determine the relationship between tremor amplitude measured with wearable inertial sensors and clinician- and patient-rated measures. For 14 days, patients (1) wore inertial sensors on both wrists, (2) self-rated their tremor using a diary, (3) drew spirals, and (4) completed an activities of daily living scale once daily. Patients were also scored using The Essential Tremor Rating Scale (TETRAS) performance in the clinic by a clinician. We found strong correlations in patient-reported metrics of tremor, but weak correlations between these data and both the inertial sensor data and the in-clinic TETRAS scores. The patient experience of tremor during normal daily activities may differ from the transducer-based measures of tremor amplitude and rating scales of tremor severity. Future studies should consider how to record features of tremor that are important to patients.

Spatiotemporal Distribution of Shallow Tremors Along the Nankai Trough, Southwest Japan, as Determined From Waveform Amplitudes and Cross‐Correlations

AbstractShallow slow earthquakes in subduction zones are related to spatiotemporal variations of stress on the plate boundary adjacent to a megathrust earthquake. A dense network (DONET) of cabled ocean‐bottom seismometers (OBSs) deployed in the Nankai Trough seismogenic zone, southwest Japan, has enabled real‐time monitoring of tremors. Because the amplitudes of tremor signals are very low, even when recorded by OBSs, we developed a hybrid method that simultaneously uses traveltimes based on cross‐correlations and amplitudes of waveforms to determine tremor hypocenters. We applied our method to waveforms observed during five years of continuous DONET recording from April 2016 to September 2021 to detect more than 6,500 tremors, including three major episodes in 2016, 2018, and 2020–2021. The tremors of each episode were distributed over distances of >100 km along the trench axis in two regions, southeast of the Kii Peninsula and off Cape Muroto. We also estimated seismic energy rates and found that for the three major tremor episodes, energy rates increased gradually with time. The 2018 and 2020–2021 episodes began near a subducted seamount and a subducted ridge, respectively, and were likely related to fluid movements controlled by the topography of the downgoing plate. We showed that the tremors were spatiotemporally synchronized with slow slip events, and their spatial variations of energy rate corresponded to stress heterogeneity on the shallow plate boundary. Future real‐time monitoring of shallow tremors will contribute to our understanding of changes of stress state due to slow earthquakes at the shallow plate boundary.

Cerebellar Dysfunction as a Source of Dystonic Phenotypes in Mice

There is now a substantial amount of compelling evidence demonstrating that the cerebellum may be a central locus in dystonia pathogenesis. Studies using spontaneous genetic mutations in rats and mice, engineered genetic alleles in mice, shRNA knockdown in mice, and conditional genetic silencing of fast neurotransmission in mice have all uncovered a common set of behavioral and electrophysiological defects that point to cerebellar cortical and cerebellar nuclei dysfunction as a source of dystonic phenotypes. Here, we revisit the Ptf1aCre/+;Vglut2flox/flox mutant mouse to define fundamental phenotypes and measures that are valuable for testing the cellular, circuit, and behavioral mechanisms that drive dystonia. In this model, excitatory neurotransmission from climbing fibers is genetically eliminated and, as a consequence, Purkinje cell and cerebellar nuclei firing are altered in vivo, with a prominent and lasting irregular burst pattern of spike activity in cerebellar nuclei neurons. The resulting impact on behavior is that the mice have developmental abnormalities, including twisting of the limbs and torso. These behaviors continue into adulthood along with a tremor, which can be measured with a tremor monitor or EMG. Importantly, expression of dystonic behavior is reduced upon cerebellar-targeted deep brain stimulation. The presence of specific combinations of disease-like features and therapeutic responses could reveal the causative mechanisms of different types of dystonia and related conditions. Ultimately, an emerging theme places cerebellar dysfunction at the center of a broader dystonia brain network.

Intraoperative Quantification of MDS-UPDRS Tremor Measurements Using 3D Accelerometry: A Pilot Study.

The most frequently used method for evaluating tremor in Parkinson’s disease (PD) is currently the internationally standardized Movement Disorder Society—Unified PD Rating Scale (MDS-UPDRS). However, the MDS-UPDRS is associated with limitations, such as its inherent subjectivity and reliance on experienced raters. Objective motor measurements using accelerometry may overcome the shortcomings of visually scored scales. Therefore, the current study focuses on translating the MDS-UPDRS tremor tests into an objective scoring method using 3D accelerometry. An algorithm to measure and classify tremor according to MDS-UPDRS criteria is proposed. For this study, 28 PD patients undergoing neurosurgical treatment and 26 healthy control subjects were included. Both groups underwent MDS-UPDRS tests to rate tremor severity, while accelerometric measurements were performed at the index fingers. All measurements were performed in an off-medication state. Quantitative measures were calculated from the 3D acceleration data, such as tremor amplitude and area-under-the-curve of power in the 4–6 Hz range. Agreement between MDS-UPDRS tremor scores and objective accelerometric scores was investigated. The trends were consistent with the logarithmic relationship between tremor amplitude and MDS-UPDRS score reported in previous studies. The accelerometric scores showed a substantial concordance (>69.6%) with the MDS-UPDRS ratings. However, accelerometric kinetic tremor measures poorly associated with the given MDS-UPDRS scores (R2 < 0.3), mainly due to the noise between 4 and 6 Hz found in the healthy controls. This study shows that MDS-UDPRS tremor tests can be translated to objective accelerometric measurements. However, discrepancies were found between accelerometric kinetic tremor measures and MDS-UDPRS ratings. This technology has the potential to reduce rater dependency of MDS-UPDRS measurements and allow more objective intraoperative monitoring of tremor.

Insights into the dynamics of the 2010 Eyjafjallajökull eruption using seismic interferometry and network covariance matrix analyses

Applying seismic interferometry and network covariance matrix-based analyses to detect and locate the source of volcanic tremor during the 2010 Eyjafjallajökull effusive flank and explosive-effusive summit eruptions has provided new insights into this iconic event. The tremor source locations derived from the network covariance matrix approach were spatially distinct during the two eruptions. The tremor was radiated between the surface and 5–6 km depth during the effusive flank eruption, including an apparently progressive upward migration in early April 2010, but was strictly confined to the surface during the summit eruption. Each phase of the summit eruption left a distinct fingerprint in the seismic records. Effusive phases radiated continuous tremor between 0.6 and 5 Hz, whereas explosive phases produced tremor in a more pulsating fashion over a wider frequency band (0.2–10 Hz). A period of intermittent tremor bursts (called banded tremor) on 15 April, associated with formation of a new vent at the summit, was most likely generated by magma-gas-meltwater interaction within a subglacial enclosure. The banded tremor ceased following an abrupt draining of the newly formed subglacial cauldron, resulting in a large slurry glacial meltwater flood (jökulhlaup). This study highlights the importance of new data processing methodologies for future monitoring of volcanic tremor in real-time.

Seismic energy radiation and along-strike heterogeneities of shallow tectonic tremors at the Nankai Trough and Japan Trench

Shallow slow earthquakes have been documented along shallow plate interfaces near trenches. Recent geophysical observation networks located offshore of Japan enable us to analyze shallow tremors in the Nankai Trough and the Japan Trench. Onshore seismic stations are also important for detecting shallow very low frequency earthquakes (VLFEs) and for evaluating their seismicity prior to the deployment of offshore observation networks. This study analyzes data from ocean bottom seismometers to estimate the seismic energy radiation of such tremors, from which we observe along-strike heterogeneity. Tremors with higher seismic energy rates tend to have longer recurrence intervals, which has also been observed for deep tremors. This study also estimates seismic moment releases of shallow VLFEs at the Japan Trench observed by onshore seismic stations using Green's function in a local three-dimensional seismic velocity model. The scaled energy, which is the ratio of the seismic energy (rate) to the seismic moment (rate), is ~10−9. Shallow slow earthquakes located off the Kii Peninsula and Cape Muroto in the Nankai Trough have higher values (10−9–10−8), while shallow slow earthquakes located off the Kii channel in the Nankai Trough and in the Japan Trench have lower values (10−10–10−9), which are similar to the values estimated for deep slow earthquakes. By comparing the cumulative seismic energy of the shallow tremors with the cumulative seismic moment of shallow VLFEs, we evaluate the monitoring of shallow tremors and shallow VLFEs in the Japan Trench, suggesting that monitoring shallow VLFEs based on data from onshore seismic stations could miss many (~90% at most) of the seismic moment releases by small magnitude but frequent events.

Clinical Characteristics and Electrophysiological Biomarkers of Parkinson's Disease Developed From Essential Tremor.

Background and Objective: Parkinson's disease developed from essential tremor (ET-PD) is a distinct clinical syndrome that is different from essential tremor (ET) and Parkinson's disease (PD). There is currently a lack of research on ET-PD. Tremor characteristics (amplitude and frequency) are primary quantitative indexes for diagnosing and monitoring of tremors. In this study, we aimed to explore specific clinical and electrophysiological biomarkers for the identification of ET-PD.Methods: The study included patients with ET-PD (n = 22), ET (n = 42), and tremor-dominant PD (t-PD, n = 47). We collected demographic data, clinical characteristics (including motor and non-motor symptoms), and tremor analysis. The frequency, amplitude, contracting patterns of resting tremor and postural tremor were collected. The analysis of ET-PD and ET/t-PD was compared. The receiver operating characteristic (ROC) curve was used to analyze the electrophysiological features in distinguishing ET-PD from ET or t-PD.Results: Compared with ET, hyposmia, bradykinesia, rigidity, postural abnormality, and resting tremor were more common in the ET-PD group (P = 0.01, 0.003, 0.001, 0.001, 0.019, respectively). The postural tremor frequencies of the head, upper limbs, and lower limbs were significantly lower in the ET-PD than in the ET (P = 0.007, 0.003, 0.035, respectively), which were the most appropriate variables for distinguishing ET-PD from ET (AUC: 0.775, 0.727, and 0.701, respectively). Compared with t-PD, bradykinesia, rigidity, postural abnormality (both P < 0.001), and resting tremor (P = 0.024) were less common in the ET-PD. The postural tremor amplitudes of the head and upper limbs were significantly higher in the ET-PD than in the t-PD (P = 0.022, 0.001, respectively), which were the most appropriate variables for distinguishing ET-PD from t-PD (AUC: 0.793 and 0.716).Conclusions: Hyposmia and electrophysiological biomarkers (postural tremor frequencies and amplitudes) help early recognition of ET-PD.

Development of digital biomarkers for resting tremor and bradykinesia using a wrist-worn wearable device

Objective assessment of Parkinson’s disease symptoms during daily life can help improve disease management and accelerate the development of new therapies. However, many current approaches require the use of multiple devices, or performance of prescribed motor activities, which makes them ill-suited for free-living conditions. Furthermore, there is a lack of open methods that have demonstrated both criterion and discriminative validity for continuous objective assessment of motor symptoms in this population. Hence, there is a need for systems that can reduce patient burden by using a minimal sensor setup while continuously capturing clinically meaningful measures of motor symptom severity under free-living conditions. We propose a method that sequentially processes epochs of raw sensor data from a single wrist-worn accelerometer by using heuristic and machine learning models in a hierarchical framework to provide continuous monitoring of tremor and bradykinesia. Results show that sensor derived continuous measures of resting tremor and bradykinesia achieve good to strong agreement with clinical assessment of symptom severity and are able to discriminate between treatment-related changes in motor states.

A Smart Inertial System for 24h Monitoring and Classification of Tremor and Freezing of Gait in Parkinson’s Disease

The main objective of this study is the realization of advanced methods of automatic evaluation and continuous monitoring tools for the support to the medical evaluation of Parkinson’s disease (PD) and to the follow up of the patient also through the presentation and classification of data through a universally recognized clinical scale. We have developed an embedded wearable device integrating a tri-axial accelerometer, a tri-axial gyroscope, and a tri-axial magnetometer, realizing a sensor with embedded data fusion and aggregation algorithms. Two different classification and detection algorithms are implemented to provide a comprehensive description and objective quantification of the two most investigated PD’s symptoms as tremor and freezing of gait (FoG). The system is usable in clinical and diagnostic settings showing its effectiveness above all in patients home monitoring through the upload in Cloud and the real-time processing of the acquired data. In the latter case, it allows the detection and objectification of PD symptoms manifested by the patient in everyday life, thereby providing the doctor with essential information for the dosage and effective personalization of drug therapy. For the system validation, a series of tests were carried out according to a defined experimental protocol on a sample of PD patients and healthy subjects with the presence of a video recording system and under the supervision of a neurologists team. The results obtained show a system’s accuracy of 97.7% for tremor classification and 99.7% for FoG events detection.

Quantitative Assessment of Head Tremor in Patients with Essential Tremor and Cervical Dystonia by Using Inertial Sensors.

Tremor is most common among the movement disabilities that affect older people, having a prevalence rate of 4.6% in the population older than 65 years. Despite this, distinguishing different types of tremors is clinically challenging, often leading to misdiagnosis. However, due to advances in microelectronics and wireless communication, it is now possible to easily monitor tremor in hospitals and even in home environments. In this paper, we propose an architecture of a system for remote health-care and one possible implementation of such system focused on head tremor monitoring. In particular, the aim of the study presented here was to test new tools for differentiating essential tremor from dystonic tremor. To that aim, we propose a number of temporal and spectral features that are calculated from measured gyroscope signals, and identify those that provide optimal differentiation between two groups. The mean signal amplitude feature results in sensitivity = 0.8537 and specificity = 0.8039 in distinguishing patients having cervical dystonia with or without tremor. In addition, mean signal amplitude was shown to be significantly higher in patients with essential tremor than in patients with cervical dystonia, whereas the mean peak frequency is not different between two groups.

A Two-Stage Tremor Detection Algorithm for Wearable Inertial Sensors During Normal Daily Activities.

Continuous monitoring of tremor with wearable wrist sensors during normal daily activities is more difficult than in a clinical setting when subjects perform prescribed activities because some normal daily activities resemble tremor, many normal movements contain frequency content that overlaps with the tremor frequency, and the tremor amplitude has a large dynamic range during normal daily activities. We describe a novel two-stage algorithm that offers improvement at discriminating tremor from other activities. Some of this improvement is attained by using prior domain knowledge that tremor occurs over a narrow range of frequencies for an individual, but the mean tremor frequency may vary significantly between individuals in a study population. We validated the algorithm in continuous recordings from people with Parkinson's disease and matched control subjects. The algorithm has good face validity, a low rate of false positives on recordings from control subjects (<; 1.1%), and good correspondence with the constancy of rest tremor as measured by this question on the MDS-UPDRS (ρ = 0.54).

Activity-aware essential tremor evaluation using deep learning method based on acceleration data

BackgroundEssential tremor (ET), one of the most common neurological disorders is typically evaluated with validated rating scales which only provide a subjective assessment during a clinical visit, underestimating the fluctuations tremor during different daily activities. Motion sensors have shown favorable performances in both quantifying tremor and voluntary human activity recognition (HAR). ObjectiveTo create an automated system of a reference scale using motion sensors supported by deep learning algorithms to accurately rate ET severity during voluntary activities, and to propose an IOTA based blockchain application to share anonymously tremor data. MethodA smartwatch-based tremor monitoring system was used to collect motion data from 20 subjects while they were doing standard tasks. Two neurologists rated ET by Fahn-Tolosa Marin Tremor Rating Scale (FTMTRS). Supported by deep learning techniques, activity classification models (ACMs) and tremor evaluation models (TEMs) were created and algorithms were implemented, to distinguish voluntary human activities and evaluate tremor severity respectively. ResultA practical application example showed that the proposed ACMs can classify six typical activities with high accuracy (89.73%–98.84%) and the results produced by the TEMs are significantly correlated with the FTMTRS ratings of two neurologists (r1 = 0.92, p1 = 0.008; r2 = 0.93, p2 = 0.007). ConclusionThis study demonstrated that motion sensor data, supported by deep learning algorithms, can be used to classify human activities and evaluate essential tremor severity during different activities.

A novel device for continuous monitoring of tremor and other motor symptoms.

The clinical assessment of Parkinson's disease (PD) symptoms is typically performed with neurological examinations and simple motor tests. However, this only takes into account the severity of motor symptoms during the length of the recording and fails to capture variations in a patient's motor state, which change continuously during the day. Most of the current methods for long-term monitoring of extrapyramidal symptoms are based on the use of a wearable magneto-inertial device that evaluates the frequential content of signals in the range of movement disorders. However, the typical daily motor activities performed by patients may have a power spectrum into the same range of motor symptoms, and habitual activity may be indistinguishable from that due to movement disorders. In this work, we report a new device and method for the continuous and long-term monitoring of tremor due to PD and other movement disorders to reduce the probability of mistaking the discrimination between extrapyramidal symptoms and normal daily activity. The method is based on the evaluation of frequential data content from multi-axial sensors and on the identification of specific movement patterns that Parkinsonian and extrapyramidal symptoms are typically associated with. In this study, 16 patients with movement disorders were recruited. While results need to be extended with further studies and clinical trials, the proposed device appears promising and suitable for the use as part of clinical trials and routine clinical practice for supporting the evaluation of motor symptoms, disease progression, and the quantification of therapeutic effects.

A classification system for assessment and home monitoring of tremor in patients with Parkinson's disease

Tremor is one of the most common symptoms of Parkinson's disease (PD), which is widely being used in the diagnosis procedure. Accurate estimation of PD tremor based on Unified PD Rating Scale (UPDRS) provides aid for physicians in prescription and home monitoring. This article presents a robust design of a classification system to estimate PD patient's hand tremors and the results of the proposed system as compared to the UPDRS. A smartphone accelerometer sensor is used for accurate and noninvasive data acquisition. We applied short-time Fourier transform to time series data of 52 PD patients. Features were extracted based on the severity of PD patients' hand tremor. The wrapper method was employed to determine the most discriminative subset of the extracted features. Four different classifiers were implemented for achieving best possible accuracy in the estimation of PD hand tremor based on UPDRS. Of the four tested classifiers, the Naive Bayesian approach proved to be the most accurate one. The classification result for the assessment of PD tremor achieved close to 100% accuracy by selecting an optimum combination of extracted features of the acceleration signal acquired. For home health-care monitoring, the proposed algorithm was also implemented on a cost-effective embedded system equipped with a microcontroller, and the implemented classification algorithm achieved 93.8% average accuracy. The accuracy result of both implemented systems on MATLAB and microcontroller is acceptable in comparison with the previous works.

A classification system for assessment and home monitoring of tremor in patients with Parkinson's disease

A classification system for assessment and home monitoring of tremor in patients with Parkinson's disease