Investigation of Anticipatory Postural Adjustments during One-Leg Stance Using Inertial Sensors: Evidence from Subjects with Parkinsonism

Analysis of the characteristics of anticipatory postural adjustments in older adults using smartphones: Association between cognitive and balance functions

OBJECTIVES

Using smartphones, we aimed to clarify the characteristics of anticipatory postural adjustments (APA) in older adults and examine the relationship between cognitive and balance functions.

METHODS

The study participants were 10 young and 13 older adults. An accelerometer built into a smartphone was attached to the lower back (L5) of the participant, and acceleration in the mediolateral direction was measured using a one-leg stance (OLS). As APA features, we analyzed the time to the peak value in the stance direction (peak latency [PL]) and the amount of displacement to the peak value in the stance direction (peak magnitude [PM]). Additionally, the measured PL was divided by PM for each group to obtain the APA ratio (APAr). We investigated the relationship between the APAr and Mini-BESTest subitems.

RESULTS

Older adults showed delayed PL and decreased PM levels (p < 0.01). While in the Mini-BESTest sub-items, deductions were most common in the order of dual-task and single-leg standing, and most participants with low APAr scores were degraded in APA of sub-items. The correlation was observed between APAr and both TUG and dual-task cost (DTC) (r= -0.56, r= -0.67). According to the receiver operating characteristic curve, the APAr value was 1.71 in the older age group.

CONCLUSIONS

Older adults showed delayed PL and decreased PM, and APAr was associated with cognitive and locomotor functions. By evaluating the APAr at the initiation of movement, it may be possible to distinguish the APA of the older adluts from the possible to the impossible of OLS movement.

New quantitative evaluation of anticipatory postural adjustments using a smartphone in patients with Parkinson's disease

OBJECTIVE

To investigate a smartphone-enabled quantitative evaluation of anticipatory postural adjustments (APA) during one-leg stance (OLS) movements among individuals with Parkinson's disease (PD).

METHODS

This cross-sectional study included 10 young controls, 10 older individuals, and 13 individuals with PD. A smartphone and accelerometer were attached to the participants' lower back (L5), and the movements of the lower back toward the stance side during OLS were measured. For acceleration, the time to the peak value in the stance direction (peak latency [PL]) and the amount of displacement to the peak value in the stance direction (peak magnitude [PM]) were analyzed as APA characteristics. Additionally, the measured PL was divided by the PM for each group to obtain the APA ratio (APAr) as a new index.

RESULTS

Individuals with PD showed a delayed PL and decreased PM (vs. young controls: p = .002 for PL, p < .001 for PM) (vs. older individuals: p = .022 for PL, p = .001 for PM). The APAr clustered the young controls, older individuals, and individuals with PD. According to the receiver operating characteristic curve the APAr value was 0.95, and individuals in the PD group were identified (i.e. area under the curve: 0.98; sensitivity: 85.0%; specificity: 100%). Moreover the APAr was correlated with severity and balance ability in individuals with PD (p = .015 for NFOG-Q, p = .028 for UPDRS, p = .036 for TUG, p = .015 for Mini-BESTest, p = .018 for OLS time).

CONCLUSIONS

This smartphone-based evaluation using the APAr index was reflective of disease severity and decreased balance ability among individuals with PD. The facilitation of this measurement can help clinicians and physiotherapists quantitatively evaluate the APA of individuals with PD at laboratories and hospitals as well as in home environments.

Objective estimation of m-CTSIB balance test scores using wearable sensors and machine learning

Accurate balance assessment is important in healthcare for identifying and managing conditions affecting stability and coordination. It plays a key role in preventing falls, understanding movement disorders, and designing appropriate therapeutic interventions across various age groups and medical conditions. However, traditional balance assessment methods often suffer from subjectivity, lack of comprehensive balance assessments and remote assessment capabilities, and reliance on specialized equipment and expert analysis. In response to these challenges, our study introduces an innovative approach for estimating scores on the Modified Clinical Test of Sensory Interaction on Balance (m-CTSIB). Utilizing wearable sensors and advanced machine learning algorithms, we offer an objective, accessible, and efficient method for balance assessment. We collected comprehensive movement data from 34 participants under four different sensory conditions using an array of inertial measurement unit (IMU) sensors coupled with a specialized system to evaluate ground truth m-CTSIB balance scores for our analysis. This data was then preprocessed, and an extensive array of features was extracted for analysis. To estimate the m-CTSIB scores, we applied Multiple Linear Regression (MLR), Support Vector Regression (SVR), and XGBOOST algorithms. Our subject-wise Leave-One-Out and 5-Fold cross-validation analysis demonstrated high accuracy and a strong correlation with ground truth balance scores, validating the effectiveness and reliability of our approach. Key insights were gained regarding the significance of specific movements, feature selection, and sensor placement in balance estimation. Notably, the XGBOOST model, utilizing the lumbar sensor data, achieved outstanding results in both methods, with Leave-One-Out cross-validation showing a correlation of 0.96 and a Mean Absolute Error (MAE) of 0.23 and 5-fold cross-validation showing comparable results with a correlation of 0.92 and an MAE of 0.23, confirming the model's consistent performance. This finding underlines the potential of our method to revolutionize balance assessment practices, particularly in settings where traditional methods are impractical or inaccessible.

Clinical Static Balance Assessment: A Narrative Review of Traditional and IMU-Based Posturography in Older Adults and Individuals with Incomplete Spinal Cord Injury

Maintaining a stable upright posture is essential for performing activities of daily living, and impaired standing balance may impact an individual's quality of life. Therefore, accurate and sensitive methods for assessing static balance are crucial for identifying balance impairments, understanding the underlying mechanisms of the balance deficiencies, and developing targeted interventions to improve standing balance and prevent falls. This review paper first explores the methods to quantify standing balance. Then, it reviews traditional posturography and recent advancements in using wearable inertial measurement units (IMUs) to assess static balance in two populations: older adults and those with incomplete spinal cord injury (iSCI). The inclusion of these two groups is supported by their large representation among individuals with balance impairments. Also, each group exhibits distinct aspects in balance assessment due to diverse underlying causes associated with aging and neurological impairment. Given the high vulnerability of both demographics to balance impairments and falls, the significance of targeted interventions to improve standing balance and mitigate fall risk becomes apparent. Overall, this review highlights the importance of static balance assessment and the potential of emerging methods and technologies to improve our understanding of postural control in different populations.

Assessing balance in people with bilateral vestibulopathy using the Mini-Balance Evaluation Systems Test (Mini-BESTest): feasibility and comparison with healthy control data

OBJECTIVES

Bilateral vestibulopathy (BVP) leads to unsteadiness when walking, which worsens in darkness or on uneven ground, as well as falls. Since simple balance tests struggle to distinguish between BVP and healthy participants, we aimed (1) to test if the Mini-BESTest is feasible in BVP, (2) how people with BVP perform on the Mini-BESTest and (3) to compare these scores with healthy reference data.

METHODS

Fifty participants with BVP completed the Mini-BESTest. 12-month falls incidence was obtained by questionnaire. To compare the overall and sub-scores between our participants with BVP and those of healthy participants from the literature (n = 327; obtained via PubMed searches), Mann-Whitney U tests were used. Sub scores within the BVP group were also compared. Spearman correlations were used to investigate the relationships between Mini-BESTest score and age.

RESULTS

No floor or ceiling effects were observed. Participants with BVP had significantly lower Mini-BESTest total scores than the healthy group. Anticipatory, reactive postural control and sensory orientation sub scores of the Mini-BESTest were significantly lower in BVP, while dynamic gait sub scores were not significantly different. A stronger negative correlation between age and Mini-BESTest total score was found in BVP than in the healthy group. Scores did not differ between patients with different falls history.

CONCLUSION

The Mini-BESTest is feasible in BVP. Our results confirm the commonly reported balance deficits in BVP. The stronger negative association between age and balance in BVP might reflect the age-related decline in the remaining sensory systems with which people with BVP compensate.

The instrumented single leg stance test detects early balance impairment in people with multiple sclerosis

Balance impairment is frequent in people with multiple sclerosis (pwMS) and affects risk of falls and quality of life. By using inertial measurement units (IMUs) on the Single Leg Stance Test (SLS) we aimed to discriminate healthy controls (HC) from pwMS and detect differences in balance endurance and quality. Thirdly, we wanted to test the correlation between instrumented SLS parameters and self-reported measures of gait and balance. Fifty-five pwMS with mild (EDSS<4) and moderate disability (EDSS≥4) and 20 HC performed the SLS with 3 IMUs placed on the feet and sacrum and filled the Twelve Item Multiple Sclerosis Walking Scale (MSWS-12) questionnaire. A linear mixed model was used to compare differences in the automated balance measures. Balance duration was significantly longer in HC compared to pwMS (p < 0.001) and between the two disability groups (p < 0.001). Instrumented measures identified that trunk stability (normalized mediolateral and antero-posterior center of mass stability) had the strongest association with disability (R2 marginal 0.30, p < 0.001) and correlated well with MSWS-12 (R = 0.650, p < 0.001). PwMS tended to overestimate own balance compared to measured balance duration. The use of both self-reported and objective assessments from IMUs can secure the follow-up of balance in pwMS.

Reliability and validity of quantitative evaluation of anticipatory postural adjustments using smartphones

[Purpose] This study aimed to investigate the reliability and validity of the quantitative evaluation of anticipatory postural adjustments using smartphones. [Participants and Methods] The study included 10 young control participants who underwent a one-legged stance with an accelerometer and a smartphone that were simultaneously attached to their lower back (L5). Acceleration was measured as the mediolateral component of the lumbar movement toward the stance side. The peak value of the time (peak latency) and the amount of displacement (peak magnitude) in the stance side direction of the lumbar acceleration were analyzed as anticipatory postural adjustment features. Intra-rater reliability was calculated for both accelerometer and smartphone measurements, while inter-rater reliability was calculated for smartphone measurements by two examiners. Validity was determined for both accelerometer and smartphone measurements. [Results] In this study, the intra-rater reliability of the peak latency and peak magnitude in accelerometer and smartphone measurements was confirmed, as was the inter-rater reliability in smartphone measurements. The intra-rater reliability was confirmed through re-testing, while the validity of the accelerometer and smartphone measurements was also confirmed. [Conclusion] The findings of this study suggest that the use of smartphones to measure anticipatory postural adjustments is highly reliable and valid, making it a useful clinical balance index. The method is simple and can be used for continuous patient monitoring.

Association of 7-Day Profiles of Motor Activity in Marital Dyads with One Component Affected by Parkinson's Disease

(1) Background: A noticeable association between the motor activity (MA) profiles of persons living together has been found in previous studies. Social actigraphy methods have shown that this association, in marital dyads composed of healthy individuals, is greater than that of a single person compared to itself. This study aims at verifying the association of MA profiles in dyads where one component is affected by Parkinson's disease (PD). (2) Methods: Using a wearable sensor-based social actigraphy approach, we continuously monitored, for 7 days, the activities of 27 marital dyads including one component with PD. (3) Results: The association of motor activity profiles within a marital dyad (cross-correlation coefficient 0.344) is comparable to the association of any participant with themselves (0.325). However, when considering the disease severity quantified by the UPDRS III score, it turns out that the less severe the symptoms, the more associated are the MA profiles. (4) Conclusions: Our findings suggest that PD treatment could be improved by leveraging the MA of the healthy spouse, thus promoting lifestyles also beneficial for the component affected by PD. The actigraphy approach provided valuable information on habitual functions and motor fluctuations, and could be useful in investigating the response to treatment.

Asymmetric Influence of Dual-Task Interference on Anticipatory Postural Adjustments in One-Leg Stance

This study investigated the differences of anticipatory postural adjustments (APAs) in a one-leg stance (OLS) that appear according to lower-extremity dominance and dual-task interference. Thirteen young, healthy, male volunteers performed the OLS task under the following six conditions: (1) dominant leg (DL), single-task; (2) DL, dual-task, with a low level of cognitive load (DT1/2); (3) DL, dual-task, with a high level of cognitive load (DT + 1); (4) non-dominant leg (NDL), single-task; (5) NDL, DT1/2; and (6) NDL, DT + 1. In order to measure the subjects' APA, we used the medial-lateral displacement of their centers of pressure and gravity from the force plate and the time-series data of joint angular motions, recorded using a 3D motion analysis system. In the NDL under the dual-task condition, the onset of APA was delayed and the amplitude declined, which resulted in an increase in the duration of the APA period. The number of components identified by principal component analysis differed according to the dominant foot, and the change caused by cognitive load was found only in the NDL. As the cognitive load increased, the variance of the principal component decreased. These findings show that dual-task interference asymmetrically influences APA according to limb dominance, which reorganizes the coordination strategy of joints' angular motion.

Review of Active Extracorporeal Medical Devices to Counteract Freezing of Gait in Patients with Parkinson Disease

Parkinson Disease (PD) primarily affects older adults. It is the second-most common neurodegenerative disease after Alzheimer’s disease. Currently, more than 10 million people suffer from PD, and this number is expected to grow, considering the increasing global longevity. Freezing of Gait (FoG) is a symptom present in approximately 80% of advanced-stage PD’s patients. FoG episodes alter the continuity of gait, and may be the cause of falls that can lead to injuries and even death. The recent advances in the development of hardware and software systems for the monitoring, stimulus, or rehabilitation of patients with FoG has been of great interest to researchers because detection and minimization of the duration of FoG events is an important factor in improving the quality of life. This article presents a review of the research on non-invasive medical devices for FoG, focusing on the acquisition, processing, and stimulation approaches used.

Ability of Wearable Accelerometers-Based Measures to Assess the Stability of Working Postures

With the rapid development and widespread application of wearable inertial sensors in the field of human motion capture, the low-cost and non-invasive accelerometer (ACC) based measures have been widely used for working postural stability assessment. This study systematically investigated the abilities of ACC-based measures to assess the stability of working postures in terms of the ability to detect the effects of work-related factors and the ability to classify stable and unstable working postures. Thirty young males participated in this study and performed twenty-four load-holding tasks (six working postures × two standing surfaces × two holding loads), and forty-three ACC-based measures were derived from the ACC data obtained by using a 17 inertial sensors-based motion capture system. ANOVAs, t-tests and machine learning (ML) methods were adopted to study the factors’ effects detection ability and the postural stability classification ability. The results show that almost all forty-three ACC-based measures could (p < 0.05) detect the main effects of Working Posture and Load Carriage, and their interaction effects. However, most of them failed in (p ≥ 0.05) detecting Standing Surface’s main or interaction effects. Five measures could detect both main and interaction effects of all the three factors, which are recommended for working postural stability assessment. The performance in postural stability classification based on ML was also good, and the feature set exerted a greater influence on the classification accuracy than sensor configuration (i.e., sensor placement locations). The results show that the pelvis and lower legs are recommended locations overall, in which the pelvis is the first choice. The findings of this study have proved that wearable ACC-based measures could assess the stability of working postures, including the work-related factors’ effects detection ability and stable-unstable working postures classification ability. However, researchers should pay more attention to the measure selection, sensors placement, feature selection and extraction in practical applications.

Boxing with and without Kicking Techniques for People with Parkinson's Disease: An Explorative Pilot Randomized Controlled Trial

BACKGROUND

People with Parkinson's disease (PD) benefit from boxing exercise. Adding kicking variations to the boxing may provide additional benefit to improve balance. However, the benefits and adherence to such trainings is unknown.

OBJECTIVE

To explore the feasibility, safety, and benefits on balance of boxing training combined with kicking techniques in comparison to boxing without kicking in PD.

METHODS

Participants were randomized to group-based boxing training with kicking techniques (BK) or to group-based boxing alone training (BO). Both groups trained for one hour, once a week, for a period of 10 weeks. Participants were assessed at baseline and ten weeks post-intervention for difference in balance, fear of falling, balance confidence, walking ability, and quality of life.

RESULTS

Twenty-nine people with PD (median age 64 years; median disease duration 5 years) participated. Both interventions were feasible and acceptable for all participants. No adverse events occurred. Most participants (BK 80%; BO 75%) were satisfied with the training. We found no significant between group difference on either the primary (Mini-BEST) or secondary outcomes. The within group comparison showed that balance improved in both groups after the intervention (BK 22.60 (2.7) to 25.33 (2.64) p = 0.02; BO 23.09 (3.44) to 25.80 (2.39); p = 0.01 on the Mini BEST test).

CONCLUSION

Both types of boxing seem to be feasible and safe. Adding kicking techniques to boxing does not improve balance significantly more than boxing alone. Incorporation of kicking may be a valuable addition to the exercise therapy repertoire.

Discussion of Research Priorities for Gait Disorders in Parkinson's Disease

Gait and balance abnormalities develop commonly in Parkinson's disease and are among the motor symptoms most disabling and refractory to dopaminergic or other treatments, including deep brain stimulation. Efforts to develop effective therapies are challenged by limited understanding of these complex disorders. There is a major need for novel and appropriately targeted research to expedite progress in this area. The Scientific Issues Committee of the International Parkinson and Movement Disorder Society has charged a panel of experts in the field to consider the current knowledge gaps and determine the research routes with highest potential to generate groundbreaking data. © 2021 International Parkinson and Movement Disorder Society.

Postural Adjustments and Biomechanics During Gait Initiation and Obstacle Negotiation: A Comparison Between Akinetic-Rigid and Hyperkinetic Parkinson's Disease

Background: Individuals with Parkinson's disease (PD) exhibit different combinations of motor symptoms. The most frequent subtypes are akinetic-rigid (AK-R) and hyperkinetic (HYP). Motor symptoms, such as rigidity and bradykinesia, can directly affect postural adjustments and performance in daily tasks, like gait initiation and obstacles negotiation, increasing the risk of falls and functional dependence. Objective: To compare postural adjustments and biomechanical parameters during the gait initiation and obstacle negotiation of people with AK-R and HYP PD and correlate with functional mobility and risk of falls. Methods: Cross-sectional study. Thirty-three volunteers with PD were divided into two groups according to clinical motor manifestations: AK-R (n = 16) and HYP (n = 17). We assessed the anticipatory (APA), compensatory (CPA) postural adjustments analyzing kinematic, kinetic and, electromyographic parameters during the gait initiation and obstacle negotiation tests. We applied independent T-tests and Pearson correlation tests for comparisons and correlations, respectively (α = 0.05). Results: In the APA phase of the gait initiation test, compared to the functional HYP group, the AK-R group showed shorter time for single support (p = 0.01), longer time for double support (p = 0.01) accompanied by a smaller first step (size, p = 0.05; height, p = 0.04), and reduced muscle activation of obliquus internus (p = 0.02). Similarly, during the first step in the obstacle negotiation test, the AK-R group showed less step height (p = 0.01) and hip excursion (p = 0.02), accompanied by a reduced mediolateral displacement of the center of pressure (p = 0.02) during APA, and activation of the gluteus medius (p = 0.02) and the anterior tibialis (p = 0.04) during CPA in comparison with HYP group. Conclusion: The findings suggest that people with AK-R present impaired postural adjustments during gait initiation and obstacles negotiation compared to hyperkinetic PD. Based on defined motor symptoms, the proposition presented here revealed consistent postural adjustments during complex tasks and, therefore, may offer new insights onto PD motor evaluation and neurorehabilitation.

Balance Deficits due to Cerebellar Ataxia: A Machine Learning and Cloud-Based Approach

Cerebellar ataxia (CA) refers to the disordered movement that occurs when the cerebellum is injured or affected by disease. It manifests as uncoordinated movement of the limbs, speech, and balance. This study is aimed at the formation of a simple, objective framework for the quantitative assessment of CA based on motion data. We adopted the Recurrence Quantification Analysis concept in identifying features of significance for the diagnosis. Eighty-six subjects were observed undertaking three standard neurological tests (Romberg's, Heel-shin and Truncal ataxia) to capture 213 time series inertial measurements each. The feature selection was based on engaging six different common techniques to distinguish feature subset for diagnosis and severity assessment separately. The Gaussian Naive Bayes classifier performed best in diagnosing CA with an average double cross-validation accuracy, sensitivity, and specificity of 88.24%, 85.89%, and 92.31%, respectively. Regarding severity assessment, the voting regression model exhibited a significant correlation (0.72 Pearson) with the clinical scores in the case of the Romberg's test. The Heel-shin and Truncal tests were considered for diagnosis and assessment of severity concerning subjects who were unable to stand. The underlying approach proposes a reliable, comprehensive framework for the assessment of postural stability due to cerebellar dysfunction using a single inertial measurement unit.

Fifteen Years of Wireless Sensors for Balance Assessment in Neurological Disorders

Balance impairment is a major mechanism behind falling along with environmental hazards. Under physiological conditions, ageing leads to a progressive decline in balance control per se. Moreover, various neurological disorders further increase the risk of falls by deteriorating specific nervous system functions contributing to balance. Over the last 15 years, significant advancements in technology have provided wearable solutions for balance evaluation and the management of postural instability in patients with neurological disorders. This narrative review aims to address the topic of balance and wireless sensors in several neurological disorders, including Alzheimer’s disease, Parkinson’s disease, multiple sclerosis, stroke, and other neurodegenerative and acute clinical syndromes. The review discusses the physiological and pathophysiological bases of balance in neurological disorders as well as the traditional and innovative instruments currently available for balance assessment. The technical and clinical perspectives of wearable technologies, as well as current challenges in the field of teleneurology, are also examined.

Wearable Inertial Sensors to Assess Standing Balance: A Systematic Review

Wearable sensors are de facto revolutionizing the assessment of standing balance. The aim of this work is to review the state-of-the-art literature that adopts this new posturographic paradigm, i.e., to analyse human postural sway through inertial sensors directly worn on the subject body. After a systematic search on PubMed and Scopus databases, two raters evaluated the quality of 73 full-text articles, selecting 47 high-quality contributions. A good inter-rater reliability was obtained (Cohen's kappa = 0.79). This selection of papers was used to summarize the available knowledge on the types of sensors used and their positioning, the data acquisition protocols and the main applications in this field (e.g., "active aging", biofeedback-based rehabilitation for fall prevention, and the management of Parkinson's disease and other balance-related pathologies), as well as the most adopted outcome measures. A critical discussion on the validation of wearable systems against gold standards is also presented.

Balancing between the two: Are freezing of gait and postural instability in Parkinson's disease connected?

Postural instability and freezing of gait (FoG) are key features of Parkinson's disease (PD) closely related to falls. Growing evidence suggests that co-existing postural deficits could influence the occurrence and severity of FoG. To date, the exact nature of this interrelationship remains largely unknown. We analyzed the complex interaction between postural instability and gait disturbance by comparing the findings available in the posturographic literature between patients with and without FoG. Results showed that FoG and postural instability are intertwined, can influence each other behaviorally and may coincide neurologically. The most common FoG-related postural deficits included weight-shifting impairments, and inadequate scaling and timing of postural responses most apparent at forthcoming postural changes under time constraints. Most likely, a negative cycle of combined and more severe postural deficits in people with FoG will enhance postural stability breakdown. As such, the wide brain network deficiencies involved in FoG may also concurrently influence postural stability. Future work needs to examine whether training interventions targeting both symptoms will have extra clinical benefits on fall frequency.

Teleneurology and mobile technologies: the future of neurological care

Neurological disorders are the leading cause of global disability. However, for most people around the world, current neurological care is poor. In low-income countries, most individuals lack access to proper neurological care, and in high-income countries, distance and disability limit access. With the global proliferation of smartphones, teleneurology - the use of technology to provide neurological care and education remotely - has the potential to improve and increase access to care for billions of people. Telestroke has already fulfilled this promise, but teleneurology applications for chronic conditions are still in their infancy. Similarly, few studies have explored the capabilities of mobile technologies such as smartphones and wearable sensors, which can guide care by providing objective, frequent, real-world assessments of patients. In low-income settings, teleneurology can increase the capacity of local care systems through professional development, diagnostic support and consultative services. In high-income settings, teleneurology is likely to promote the expansion and migration of neurological care away from institutions, incorporate systems of asynchronous communication (such as e-mail), integrate clinicians with diverse skill sets and reach new populations. Inertia, outdated policies and social barriers - especially the digital divide - will slow this progress at considerable cost. However, a future increasingly will be possible in which neurological care can be accessed by anyone, anywhere. Here, we examine the emerging evidence regarding the benefits of teleneurology for chronic conditions, its role and risks in low-income countries and the promise of mobile technologies to measure disease status and deliver care. We conclude by discussing the future trends, barriers and timing for the adoption of teleneurology.

Correlations between Motor Symptoms across Different Motor Tasks, Quantified via Random Forest Feature Classification in Parkinson's Disease

Background

Objective assessments of Parkinson’s disease (PD) patients’ motor state using motion capture techniques are still rarely used in clinical practice, even though they may improve clinical management. One major obstacle relates to the large dimensionality of motor abnormalities in PD. We aimed to extract global motor performance measures covering different everyday motor tasks, as a function of a clinical intervention, i.e., deep brain stimulation (DBS) of the subthalamic nucleus.

Methods

We followed a data-driven, machine-learning approach and propose performance measures that employ Random Forests with probability distributions. We applied this method to 14 PD patients with DBS switched-off or -on, and 26 healthy control subjects performing the Timed Up and Go Test (TUG), the Functional Reach Test (FRT), a hand coordination task, walking 10-m straight, and a 90° curve.

Results

For each motor task, a Random Forest identified a specific set of metrics that optimally separated PD off DBS from healthy subjects. We noted the highest accuracy (94.6%) for standing up. This corresponded to a sensitivity of 91.5% to detect a PD patient off DBS, and a specificity of 97.2% representing the rate of correctly identified healthy subjects. We then calculated performance measures based on these sets of metrics and applied those results to characterize symptom severity in different motor tasks. Task-specific symptom severity measures correlated significantly with each other and with the Unified Parkinson’s Disease Rating Scale (UPDRS, part III, correlation of r² = 0.79). Agreement rates between different measures ranged from 79.8 to 89.3%.

Conclusion

The close correlation of PD patients’ various motor abnormalities quantified by different, task-specific severity measures suggests that these abnormalities are only facets of the underlying one-dimensional severity of motor deficits. The identification and characterization of this underlying motor deficit may help to optimize therapeutic interventions, e.g., to “automatically” adapt DBS settings in PD patients.