Tremor analysis with wearable sensors correlates with outcome after thalamic deep brain stimulation

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

Background

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.

Methods

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.

Results

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.

Conclusion

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.

WHIGET and TETRAS Ratings of Action Tremor in Patients with Essential Tremor: Substantial Association and Agreement

Background

Evaluating tremor severity is a critical component of diagnosing and clinically managing patients with essential tremor (ET). We examined the comparability of tremor severity ratings derived from two frequently used tremor rating scales: the Washington Heights-Inwood Genetic Study of Essential Tremor (WHIGET) rating scale and the Tremor Research Group Essential Tremor Rating Scale (TETRAS).

Methods

A trained assistant administered and videotaped a neurological examination, including eight items assessing upper limb action tremor (arms outstretched, arms in the wingbeat position, finger-nose-finger maneuver, and drawing of Archimedes spirals). An experienced movement disorders neurologist reviewed the videos and assigned WHIGET and TETRAS ratings. We calculated associations between TETRAS and WHIGET ratings using Spearman rank order correlations. Subsequently, we collapsed these ratings into four tremor severity categories (absent, mild, moderate, severe) and then two broader tremor severity categories (absent/mild, moderate/severe). We calculated weighted Kappa coefficients to assess agreement between category assignments based on the TETRAS and the WHIGET.

Results

Spearman's r' s were significant for all items (p's ≤ 0.001, mean r = 0.89). Weighted Kappa's revealed substantial to near perfect agreement for all eight items (mean k = 0.86, range = 0.64 to 1.00).

Conclusion

Analyses revealed substantial strength of association and substantial to near perfect agreement between items rated with the WHIGET and TETRAS scales. These data indicated that ratings provided by each scale are highly comparable.

Objective clinical registration of tremor, bradykinesia, and rigidity during awake stereotactic neurosurgery: a scoping review

Tremor, bradykinesia, and rigidity are incapacitating motor symptoms that can be suppressed with stereotactic neurosurgical treatment like deep brain stimulation (DBS) and ablative surgery (e.g., thalamotomy, pallidotomy). Traditionally, clinicians rely on clinical rating scales for intraoperative evaluation of these motor symptoms during awake stereotactic neurosurgery. However, these clinical scales have a relatively high inter-rater variability and rely on experienced raters. Therefore, objective registration (e.g., using movement sensors) is a reasonable extension for intraoperative assessment of tremor, bradykinesia, and rigidity. The main goal of this scoping review is to provide an overview of electronic motor measurements during awake stereotactic neurosurgery. The protocol was based on the PRISMA extension for scoping reviews. After a systematic database search (PubMed, Embase, and Web of Science), articles were screened for relevance. Hundred-and-three articles were subject to detailed screening. Key clinical and technical information was extracted. The inclusion criteria encompassed use of electronic motor measurements during stereotactic neurosurgery performed under local anesthesia. Twenty-three articles were included. These studies had various objectives, including correlating sensor-based outcome measures to clinical scores, identifying optimal DBS electrode positions, and translating clinical assessments to objective assessments. The studies were highly heterogeneous in device choice, sensor location, measurement protocol, design, outcome measures, and data analysis. This review shows that intraoperative quantification of motor symptoms is still limited by variable signal analysis techniques and lacking standardized measurement protocols. However, electronic motor measurements can complement visual evaluations and provide objective confirmation of correct placement of the DBS electrode and/or lesioning. On the long term, this might benefit patient outcomes and provide reliable outcome measures in scientific research.

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.