A machine learning contest enhances automated freezing of gait detection and reveals time-of-day effects.

Freezing of gait (FOG) is a debilitating problem that markedly impairs the mobility and independence of 38-65% of people with Parkinson's disease. During a FOG episode, patients report that their feet are suddenly and inexplicably "glued" to the floor. The lack of a widely applicable, objective FOG detection method obstructs research and treatment. To address this problem, we organized a 3-month machine-learning contest, inviting experts from around the world to develop wearable sensor-based FOG detection algorithms. 1,379 teams from 83 countries submitted 24,862 solutions. The winning solutions demonstrated high accuracy, high specificity, and good precision in FOG detection, with strong correlations to gold-standard references. When applied to continuous 24/7 data, the solutions revealed previously unobserved patterns in daily living FOG occurrences. This successful endeavor underscores the potential of machine learning contests to rapidly engage AI experts in addressing critical medical challenges and provides a promising means for objective FOG quantification.

Gait asymmetry and symptom laterality in Parkinson's disease: two of a kind?

BACKGROUND: The laterality of motor symptoms is considered a key feature of Parkinson's disease (PD). Here, we investigated whether gait and turning asymmetry coincided with symptom laterality as determined by the MDS-UPRDS part III and whether it was increased compared to healthy controls (HC). METHODS: We analyzed the asymmetry of gait and turning with and without a cognitive dual task (DT) using motion capture systems and wearable sensors in 97 PD patients mostly from Hoehn & Yahr stage II and III and 36 age-matched HC. We also assessed motor symptom asymmetry using the bilateral sub-items of the MDS-UPDRS-III. Finally, we examined the strength of the association between gait asymmetry and symptom laterality. RESULTS: Participants with PD had increased gait but not more turning asymmetry compared to HC (p < 0.05). Only 53.7% of patients had a shorter step length on the more affected body side as determined by the MDS-UPDRS-III. Also, 54% took more time and 29% more steps during turns toward the more affected side. The degree of asymmetry in the different domains did not correlate with each other and was not influenced by DT-load. CONCLUSIONS: We found a striking mismatch between the side and the degree of asymmetry in different motor domains, i.e., in gait, turning, and distal symptom severity in individuals with PD. We speculate that motor execution in different body parts relies on different neural control mechanisms. Our findings warrant further investigation to understand the complexity of gait asymmetry in PD.

Towards a better understanding of anticipatory postural adjustments in people with Parkinson's disease.

INTRODUCTION: Previous studies have shown that anticipatory postural adjustments (APAs) are altered in people with Parkinson's disease but its meaning for locomotion is less understood. This study aims to investigate the association between APAs and gait initiation, gait and freezing of gait and how a dynamic postural control challenging training may induce changes in these features. METHODS: Gait initiation was quantified using wearable sensors and subsequent straight walking was assessed via marker-based motion capture. Additionally, turning and FOG-related outcomes were measured with wearable sensors. Assessments were conducted one week before (Pre), one week after (Post) and 4 weeks after (Follow-up) completion of a training intervention (split-belt treadmill training or regular treadmill training), under single task and dual task (DT) conditions. Statistical analysis included a linear mixed model for training effects and correlation analysis between APAs and the other outcomes for Pre and Post-Pre delta. RESULTS: 52 participants with Parkinson's disease (22 freezers) were assessed. We found that APA size in the medio-lateral direction during DT was positively associated with gait speed (p<0.001) and stride length (p<0.001) under DT conditions at Pre. The training effect was largest for first step range of motion and was similar for both training modes. For the associations between changes after the training (pooled sample) medio-lateral APA size showed a significant positive correlation with first step range of motion (p = 0.033) only in the DT condition and for the non-freezers only. CONCLUSIONS: The findings of this work revealed new insights into how APAs were not associated with first step characteristics and freezing and only baseline APAs during DT were related with DT gait characteristics. Training-induced changes in the size of APAs were related to training benefits in the first step ROM only in non-freezers. Based on the presented results increasing APA size through interventions might not be the ideal target for overall improvement of locomotion.

On-Demand Cueing for Freezing of Gait in Parkinson's Disease: A Randomized Controlled Trial.

BACKGROUND: Cueing can alleviate freezing of gait (FOG) in people with Parkinson's disease (PD), but using the same cues continuously in daily life may compromise effectiveness. Therefore, we developed the DeFOG-system to deliver personalized auditory cues on detection of a FOG episode. OBJECTIVES: We aimed to evaluate the effects of DeFOG during a FOG-provoking protocol: (1) after 4 weeks of DeFOG-use in daily life against an active control group; (2) after immediate DeFOG-use (within-group) in different medication states. METHOD: In this randomized controlled trial, 63 people with PD and daily FOG were allocated to the DeFOG or active control group. Both groups received feedback on their daily living step counts using the device, but the DeFOG group also received on-demand cueing. Video-rated FOG severity was compared pre- and post-intervention through a FOG-provoking protocol administered at home off and on-medication, but without using DeFOG. Within-group effects were tested by comparing FOG during the protocol with and without DeFOG. RESULTS: DeFOG-use during the 4 weeks was similar between groups, but we found no between-group differences in FOG-severity. However, the within-group analysis showed that FOG was alleviated by DeFOG (effect size d = 0.57), regardless of medication state. Combining DeFOG and medication yielded an effect size of d = 0.67. CONCLUSIONS: DeFOG reduced FOG considerably in a population of severe freezers both off and on medication. Nonetheless, 4 weeks of DeFOG-use in daily life did not ameliorate FOG during the protocol unless DeFOG was worn. These findings suggest that on-demand cueing is only effective when used, similar to other walking aids. © 2024 International Parkinson and Movement Disorder Society.

Freezing of gait assessment with inertial measurement units and deep learning: effect of tasks, medication states, and stops.

BACKGROUND: Freezing of gait (FOG) is an episodic and highly disabling symptom of Parkinson's Disease (PD). Traditionally, FOG assessment relies on time-consuming visual inspection of camera footage. Therefore, previous studies have proposed portable and automated solutions to annotate FOG. However, automated FOG assessment is challenging due to gait variability caused by medication effects and varying FOG-provoking tasks. Moreover, whether automated approaches can differentiate FOG from typical everyday movements, such as volitional stops, remains to be determined. To address these questions, we evaluated an automated FOG assessment model with deep learning (DL) based on inertial measurement units (IMUs). We assessed its performance trained on all standardized FOG-provoking tasks and medication states, as well as on specific tasks and medication states. Furthermore, we examined the effect of adding stopping periods on FOG detection performance. METHODS: Twelve PD patients with self-reported FOG (mean age 69.33 ± 6.02 years) completed a FOG-provoking protocol, including timed-up-and-go and 360-degree turning-in-place tasks in On/Off dopaminergic medication states with/without volitional stopping. IMUs were attached to the pelvis and both sides of the tibia and talus. A temporal convolutional network (TCN) was used to detect FOG episodes. FOG severity was quantified by the percentage of time frozen (%TF) and the number of freezing episodes (#FOG). The agreement between the model-generated outcomes and the gold standard experts' video annotation was assessed by the intra-class correlation coefficient (ICC). RESULTS: For FOG assessment in trials without stopping, the agreement of our model was strong (ICC (%TF) = 0.92 [0.68, 0.98]; ICC(#FOG) = 0.95 [0.72, 0.99]). Models trained on a specific FOG-provoking task could not generalize to unseen tasks, while models trained on a specific medication state could generalize to unseen states. For assessment in trials with stopping, the agreement of our model was moderately strong (ICC (%TF) = 0.95 [0.73, 0.99]; ICC (#FOG) = 0.79 [0.46, 0.94]), but only when stopping was included in the training data. CONCLUSION: A TCN trained on IMU signals allows valid FOG assessment in trials with/without stops containing different medication states and FOG-provoking tasks. These results are encouraging and enable future work investigating automated FOG assessment during everyday life.

Accuracy of consumer-based activity trackers as measuring tool and coaching device in breast and colorectal cancer survivors.

PURPOSE: Consumer-based activity trackers are used to measure and promote PA. We studied the accuracy of a wrist- and waist-worn activity tracker in cancer survivors and compared these results to a healthy age-matched control group. METHODS: Twenty-two cancer survivors and 35 healthy subjects wore an activity tracker at the waist and at the wrist combined with a reference activity monitor at the waist (Dynaport Movemonitor). The devices were worn for 14 consecutive days. The mean daily step count from both activity trackers was compared with the reference activity monitor to investigate accuracy and agreement (paired t-test, intraclass correlation, Bland-Altman plots). To evaluate the accuracy as a coaching tool, day-by-day differences within patients were calculated. The Kendall correlation coefficient was used to test the consistency of ranking daily steps between the activity trackers and the reference activity monitor. RESULTS: The wrist-worn wearable significantly overestimated the daily step count in the cancer group (mean ± SDΔ: + 1305 (2685) steps per day; p = 0.033) and in the healthy control group (mean ± SDΔ: + 1598 (2927) steps per day; p = 0.003). The waist-worn wearable underestimated the step count in both groups, although this was not statistically significant. As a coaching device, moderate (r = 0.642-0.670) and strong (r = 0.733-0.738) accuracy was found for the wrist- and waist-worn tracker, respectively, for detecting day-by-day variability in both populations. CONCLUSION: Our results show that wrist-worn activity trackers significantly overestimate daily step count in both cancer survivors and healthy control subjects. Based on the accuracy, in particular, the waist-worn activity tracker could possibly be used as a coaching tool.

Validation of Commercial Activity Trackers in Everyday Life of People with Parkinson's Disease.

Maintaining physical activity is an important clinical goal for people with Parkinson's disease (PwPD). We investigated the validity of two commercial activity trackers (ATs) to measure daily step counts. We compared a wrist- and a hip-worn commercial AT against the research-grade Dynaport Movemonitor (DAM) during 14 days of daily use. Criterion validity was assessed in 28 PwPD and 30 healthy controls (HCs) by a 2 × 3 ANOVA and intraclass correlation coefficients (ICC2,1). The ability to measure daily step fluctuations compared to the DAM was studied by a 2 × 3 ANOVA and Kendall correlations. We also explored compliance and user-friendliness. Both the ATs and the DAM measured significantly fewer steps/day in PwPD compared to HCs (p < 0.01). Step counts derived from the ATs showed good to excellent agreement with the DAM in both groups (ICC2,1 > 0.83). Daily fluctuations were detected adequately by the ATs, showing moderate associations with DAM-rankings. While compliance was high overall, 22% of PwPD were disinclined to use the ATs after the study. Overall, we conclude that the ATs had sufficient agreement with the DAM for the purpose of promoting physical activity in mildly affected PwPD. However, further validation is needed before clinical use can be widely recommended.

Split-Belt Treadmill Training to Improve Gait Adaptation in Parkinson's Disease.

BACKGROUND: Gait deficits in people with Parkinson's disease (PD) are triggered by circumstances requiring gait adaptation. The effects of gait adaptation training on a split-belt treadmill (SBT) are unknown in PD. OBJECTIVE: We investigated the effects of repeated SBT versus tied-belt treadmill (TBT) training on retention and automaticity of gait adaptation and its transfer to over-ground walking and turning. METHODS: We recruited 52 individuals with PD, of whom 22 were freezers, in a multi-center randomized single-blind controlled study. Training consisted of 4 weeks of supervised treadmill training delivered three times per week. Tests were conducted pre- and post-training and at 4-weeks follow-up. Turning (primary outcome) and gait were assessed over-ground and during a gait adaptation protocol on the treadmill. All tasks were performed with and without a cognitive task. RESULTS: We found that SBT-training improved gait adaptation with moderate to large effects sizes (P < 0.02) compared to TBT, effects that were sustained at follow-up and during dual tasking. However, better gait adaptation did not transfer to over-ground turning speed. In both SBT- and TBT-arms, over-ground walking and Movement Disorder Society-Unified Parkinson's Disease Rating Scale III (MDS-UPDRS-III scores were improved, the latter of which reached clinically meaningful effects in the SBT-group only. No impact was found on freezing of gait. CONCLUSION: People with PD are able to learn and retain the ability to overcome asymmetric gait-speed perturbations on a treadmill remarkably well, but seem unable to generalize these skills to asymmetric gait off-treadmill. Future study is warranted into gait adaptation training to boost the transfer of complex walking skills. © 2022 The Authors. Movement Disorders published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society.

Connecting real-world digital mobility assessment to clinical outcomes for regulatory and clinical endorsement-the Mobilise-D study protocol.

BACKGROUND: The development of optimal strategies to treat impaired mobility related to ageing and chronic disease requires better ways to detect and measure it. Digital health technology, including body worn sensors, has the potential to directly and accurately capture real-world mobility. Mobilise-D consists of 34 partners from 13 countries who are working together to jointly develop and implement a digital mobility assessment solution to demonstrate that real-world digital mobility outcomes have the potential to provide a better, safer, and quicker way to assess, monitor, and predict the efficacy of new interventions on impaired mobility. The overarching objective of the study is to establish the clinical validity of digital outcomes in patient populations impacted by mobility challenges, and to support engagement with regulatory and health technology agencies towards acceptance of digital mobility assessment in regulatory and health technology assessment decisions. METHODS/DESIGN: The Mobilise-D clinical validation study is a longitudinal observational cohort study that will recruit 2400 participants from four clinical cohorts. The populations of the Innovative Medicine Initiative-Joint Undertaking represent neurodegenerative conditions (Parkinson's Disease), respiratory disease (Chronic Obstructive Pulmonary Disease), neuro-inflammatory disorder (Multiple Sclerosis), fall-related injuries, osteoporosis, sarcopenia, and frailty (Proximal Femoral Fracture). In total, 17 clinical sites in ten countries will recruit participants who will be evaluated every six months over a period of two years. A wide range of core and cohort specific outcome measures will be collected, spanning patient-reported, observer-reported, and clinician-reported outcomes as well as performance-based outcomes (physical measures and cognitive/mental measures). Daily-living mobility and physical capacity will be assessed directly using a wearable device. These four clinical cohorts were chosen to obtain generalizable clinical findings, including diverse clinical, cultural, geographical, and age representation. The disease cohorts include a broad and heterogeneous range of subject characteristics with varying chronic care needs, and represent different trajectories of mobility disability. DISCUSSION: The results of Mobilise-D will provide longitudinal data on the use of digital mobility outcomes to identify, stratify, and monitor disability. This will support the development of widespread, cost-effective access to optimal clinical mobility management through personalised healthcare. Further, Mobilise-D will provide evidence-based, direct measures which can be endorsed by regulatory agencies and health technology assessment bodies to quantify the impact of disease-modifying interventions on mobility. TRIAL REGISTRATION: ISRCTN12051706.

Daily-Living Freezing of Gait as Quantified Using Wearables in People With Parkinson Disease: Comparison With Self-Report and Provocation Tests.

OBJECTIVE: Freezing of gait (FOG) is an episodic, debilitating phenomenon that is common among people with Parkinson disease. Multiple approaches have been used to quantify FOG, but the relationships among them have not been well studied. In this cross-sectional study, we evaluated the associations among FOG measured during unsupervised daily-living monitoring, structured in-home FOG-provoking tests, and self-report. METHODS: Twenty-eight people with Parkinson disease and FOG were assessed using self-report questionnaires, percentage of time spent frozen (%TF) during supervised FOG-provoking tasks in the home while off and on dopaminergic medication, and %TF evaluated using wearable sensors during 1 week of unsupervised daily-living monitoring. Correlations between those 3 assessment approaches were analyzed to quantify associations. Further, based on the %TF difference between in-home off-medication testing and in-home on-medication testing, the participants were divided into those responding to Parkinson disease medication (responders) and those not responding to Parkinson disease medication (nonresponders) in order to evaluate the differences in the other FOG measures. RESULTS: The %TF during unsupervised daily living was mild to moderately correlated with the %TF during a subset of the tasks of the in-home off-medication testing but not the on-medication testing or self-report. Responders and nonresponders differed in the %TF during the personal "hot spot" task of the provoking protocol while off medication (but not while on medication) but not in the total scores of the self-report questionnaires or the measures of FOG evaluated during unsupervised daily living. CONCLUSION: The %TF during daily living was moderately related to FOG during certain in-home FOG-provoking tests in the off-medication state. However, this measure of FOG was not associated with self-report or FOG provoked in the on-medication state. These findings suggest that to fully capture FOG severity, it is best to assess FOG using a combination of all 3 approaches. IMPACT: These findings suggest that several complementary approaches are needed to provide a complete assessment of FOG severity.

Automated freezing of gait assessment with marker-based motion capture and multi-stage spatial-temporal graph convolutional neural networks.

BACKGROUND: Freezing of gait (FOG) is a common and debilitating gait impairment in Parkinson's disease. Further insight into this phenomenon is hampered by the difficulty to objectively assess FOG. To meet this clinical need, this paper proposes an automated motion-capture-based FOG assessment method driven by a novel deep neural network. METHODS: Automated FOG assessment can be formulated as an action segmentation problem, where temporal models are tasked to recognize and temporally localize the FOG segments in untrimmed motion capture trials. This paper takes a closer look at the performance of state-of-the-art action segmentation models when tasked to automatically assess FOG. Furthermore, a novel deep neural network architecture is proposed that aims to better capture the spatial and temporal dependencies than the state-of-the-art baselines. The proposed network, termed multi-stage spatial-temporal graph convolutional network (MS-GCN), combines the spatial-temporal graph convolutional network (ST-GCN) and the multi-stage temporal convolutional network (MS-TCN). The ST-GCN captures the hierarchical spatial-temporal motion among the joints inherent to motion capture, while the multi-stage component reduces over-segmentation errors by refining the predictions over multiple stages. The proposed model was validated on a dataset of fourteen freezers, fourteen non-freezers, and fourteen healthy control subjects. RESULTS: The experiments indicate that the proposed model outperforms four state-of-the-art baselines. Moreover, FOG outcomes derived from MS-GCN predictions had an excellent (r = 0.93 [0.87, 0.97]) and moderately strong (r = 0.75 [0.55, 0.87]) linear relationship with FOG outcomes derived from manual annotations. CONCLUSIONS: The proposed MS-GCN may provide an automated and objective alternative to labor-intensive clinician-based FOG assessment. Future work is now possible that aims to assess the generalization of MS-GCN to a larger and more varied verification cohort.

Dual Task Turning in Place: A Reliable, Valid, and Responsive Outcome Measure of Freezing of Gait.

BACKGROUND: Freezing of gait (FOG) is a complex symptom in Parkinson's disease (PD) that is both elusive to elicit and varied in its presentation. These complexities present a challenge to measuring FOG in a sensitive and reliable way, precluding therapeutic advancement. OBJECTIVE: We investigated the reliability, validity, and responsiveness of manual video annotations of the turning-in-place task and compared it to the sensor-based FOG ratio. METHODS: Forty-five optimally medicated people with PD and FOG performed rapid alternating 360° turns without and with an auditory stroop dual task, thrice over two consecutive days. The tasks were video recorded, and inertial sensors were placed on the lower back and shins. Interrater reliability between three raters, criterion validity with self-reported FOG, and responsiveness to single-session split-belt treadmill (SBT) training were investigated and contrasted with the sensor-based FOG ratio. RESULTS: Visual ratings showed excellent agreement between raters for the percentage time frozen (%TF) (ICC [intra-class correlation coefficient] = 0.99), the median duration of a FOG episode (ICC = 0.90), and the number of FOG episodes (ICC = 0.86). Dual tasking improved the sensitivity and validity of visual FOG ratings resulting in increased FOG detection, criterion validity with self-reported FOG ratings, and responsiveness to a short SBT intervention. The sensor-based FOG ratio, on the contrary, showed complex FOG presentation-contingent relationships with visual and self-reported FOG ratings and limited responsiveness to SBT training. CONCLUSIONS: Manual video annotations of FOG during dual task turning in place generate reliable, valid, and sensitive outcomes for investigating therapeutic effects on FOG. © 2021 International Parkinson and Movement Disorder Society.

Modelling and identification of characteristic kinematic features preceding freezing of gait with convolutional neural networks and layer-wise relevance propagation.

BACKGROUND: Although deep neural networks (DNNs) are showing state of the art performance in clinical gait analysis, they are considered to be black-box algorithms. In other words, there is a lack of direct understanding of a DNN's ability to identify relevant features, hindering clinical acceptance. Interpretability methods have been developed to ameliorate this concern by providing a way to explain DNN predictions. METHODS: This paper proposes the use of an interpretability method to explain DNN decisions for classifying the movement that precedes freezing of gait (FOG), one of the most debilitating symptoms of Parkinson's disease (PD). The proposed two-stage pipeline consists of (1) a convolutional neural network (CNN) to model the reduction of movement present before a FOG episode, and (2) layer-wise relevance propagation (LRP) to visualize the underlying features that the CNN perceives as important to model the pathology. The CNN was trained with the sagittal plane kinematics from a motion capture dataset of fourteen PD patients with FOG. The robustness of the model predictions and learned features was further assessed on fourteen PD patients without FOG and fourteen age-matched healthy controls. RESULTS: The CNN proved highly accurate in modelling the movement that precedes FOG, with 86.8% of the strides being correctly identified. However, the CNN model was unable to model the movement for one of the seven patients that froze during the protocol. The LRP interpretability case study shows that (1) the kinematic features perceived as most relevant by the CNN are the reduced peak knee flexion and the fixed ankle dorsiflexion during the swing phase, (2) very little relevance for FOG is observed in the PD patients without FOG and the healthy control subjects, and (3) the poor predictive performance of one subject is attributed to the patient's unique and severely flexed gait signature. CONCLUSIONS: The proposed pipeline can aid clinicians in explaining DNN decisions in clinical gait analysis and aid machine learning practitioners in assessing the generalization of their models by ensuring that the predictions are based on meaningful kinematic features.

Repeated Gait Perturbation Training in Parkinson's Disease and Healthy Older Adults: A Systematic Review and Meta-Analysis.

Background: Gait impairments are common in healthy older adults (HOA) and people with Parkinson's disease (PwPD), especially when adaptations to the environment are required. Traditional rehabilitation programs do not typically address these adaptive gait demands in contrast to repeated gait perturbation training (RGPT). RGPT is a novel reactive form of gait training with potential for both short and long-term consolidation in HOA and PwPD. The aim of this systematic review with meta-analysis is to determine whether RGPT is more effective than non-RGPT gait training in improving gait and balance in HOA and PwPD in the short and longer term. Methods: This review was conducted according to the PRISMA-guidelines and pre-registered in the PROSPERO database (CRD42020183273). Included studies tested the effects of any form of repeated perturbations during gait in HOA and PwPD on gait speed, step or stride length. Studies using balance scales or sway measures as outcomes were included in a secondary analysis. Effects of randomized controlled trials (RCT) on RGPT were pooled using a meta-analysis of final measures. Results: Of the 4421 studies, eight studies were deemed eligible for review, of which six could be included in the meta-analysis, totaling 209 participants (159 PwPD and 50 HOA). The studies were all of moderate quality. The meta-analysis revealed no significant effects of RGPT over non-RGPT training on gait performance (SMD = 0.16; 95% CI = -0.18, 0.49; Z = 0.92; P = 0.36). Yet, in some individual studies, favorable effects on gait speed, step length and stride length were observed immediately after the intervention as well as after a retention period. Gait variability and asymmetry, signifying more direct outcomes of gait adaptation, also indicated favorable RGPT effects in some individual studies. Conclusion: Despite some promising results, the pooled effects of RGPT on gait and balance were not significantly greater as compared to non-RGPT gait training in PwPD and HOA. However, these findings could have been driven by low statistical power. Therefore, the present review points to the imperative to conduct sufficiently powered RCT's to verify the true effects of RGPT on gait and balance in HOA and PwPD. Systematic Review Registration: https://www.crd.york.ac.uk/prospero/display_record.php? Identifier: CRD42020183273.

Protocol for the DeFOG trial: A randomized controlled trial on the effects of smartphone-based, on-demand cueing for freezing of gait in Parkinson's disease.

BACKGROUND: Freezing of gait (FOG) is a highly incapacitating symptom that affects many people with Parkinson's disease (PD). Cueing triggered upon real-time FOG detection (on-demand cueing) shows promise for FOG treatment. Yet, the feasibility of implementation and efficacy in daily life is still unknown. Therefore, this study aims to investigate the effectiveness of DeFOG: a smartphone and sensor-based on-demand cueing solution for FOG. METHODS: Sixty-two PD patients with FOG will be recruited for this single-blind, multi-center, randomized controlled phase II trial. Patients will be randomized into either the intervention group or the active control group. For four weeks, both groups will receive feedback about their physical activity using the wearable DeFOG system in daily life. In addition, the intervention group will also receive on-demand auditory cueing and instructions. Before and after the intervention, home-based assessments will be performed to evaluate the primary outcome, i.e., "percentage time frozen" during a FOG-provoking protocol. Secondary outcomes include the training effects on physical activity monitored over 7 days and the user-friendliness of the technology. DISCUSSION: The DeFOG trial will investigate the effectiveness of personalized on-demand cueing in a controlled design, delivered for 4 weeks in the patient's home environment. We anticipate that DeFOG will reduce FOG to a greater degree than in the control group and we will explore the impact of the intervention on physical activity levels. We expect to gain in-depth insight into whether and how patients control FOG using cueing methods in their daily lives. TRIAL REGISTRATION: Clinicaltrials.gov NCT03978507.

A systematic review on exercise and training-based interventions for freezing of gait in Parkinson's disease.

Freezing of gait (FOG) in Parkinson's disease (PD) causes severe patient burden despite pharmacological management. Exercise and training are therefore advocated as important adjunct therapies. In this meta-analysis, we assess the existing evidence for such interventions to reduce FOG, and further examine which type of training helps the restoration of gait function in particular. The primary meta-analysis across 41 studies and 1838 patients revealed a favorable moderate effect size (ES = -0.37) of various training modalities for reducing subjective FOG-severity (p < 0.00001), though several interventions were not directly aimed at FOG and some included non-freezers. However, exercise and training also proved beneficial in a secondary analysis on freezers only (ES = -0.32, p = 0.007). We further revealed that dedicated training aimed at reducing FOG episodes (ES = -0.24) or ameliorating the underlying correlates of FOG (ES = -0.40) was moderately effective (p < 0.01), while generic exercises were not (ES = -0.14, p = 0.12). Relevantly, no retention effects were seen after cessation of training (ES = -0.08, p = 0.36). This review thereby supports the implementation of targeted training as a treatment for FOG with the need for long-term engagement.

How many gait initiation trials are necessary to reliably detect anticipatory postural adjustments and first step characteristics in healthy elderly and people with Parkinson's disease?

BACKGROUND: The gait initiation (GI) process can be characterized by anticipatory postural adjustments (APAs) and first step characteristics. However, even within a constrained environment, it is unclear how many trials are necessary to obtain a reliable measurement of the GI process within one assessment. RESEARCH QUESTION: How many gait initiation trials are necessary to reliably detect APAs and first step characteristics in healthy elderly (HC) and people with Parkinson's disease with Freezing of Gait (PD + FOG) under single (ST) and dual task (DT) conditions and are there any potential systematic errors? METHODS: Thirty-eight PD + FOG (ON-medication) and 30 HC performed 5 trials of GI under ST and DT (auditory stroop test). APAs and first-step-outcomes were captured with IMUs placed on the lower back and on each foot. Intraclass correlation coefficients (ICCs) and the standard error of measurement (SEM) were computed to investigate reliability and mixed model analysis to find potential systematic errors. Additionally, we computed an estimation for the number of necessary trials to reach acceptable reliability (ICC = 0.75) for each outcome. RESULTS: ICCs varied from low reliability to excellent reliability across outcomes in PD + FOG and HC. ICCs were comparable under ST and DT for most outcomes. SEM results confirmed the ICC results. A systematic error was found for the first trial in first step ROM. Number of necessary trials varied largely across outcomes. SIGNIFICANCE: Within-session reliability varied across outcomes but was similar for PD + FOG and HC, and ST and DT. ML size of APA and first step ROM were most reliable, whereas APA duration and latency were least reliable. Depending on the outcome of interest, future studies should conduct multiple trials of GI to increase reliability.

Motor-Cognitive Treadmill Training With Virtual Reality in Parkinson's Disease: The Effect of Training Duration.

Treadmill training with virtual reality (TT + VR) has been shown to improve gait performance and to reduce fall risk in Parkinson's disease (PD). However, there is no consensus on the optimal training duration. This study is a sub-study of the V-TIME randomized clinical trial (NCT01732653). In this study, we explored the effect of the duration of training based on the motor-cognitive interaction on motor and cognitive performance and on fall risk in subjects with PD. Patients in Hoehn and Yahr stages II-III, aged between 40 and 70 years, were included. In total, 96 patients with PD were assigned to 6 or 12 weeks of TT + VR intervention, and 77 patients completed the full protocol. Outcome measures for gait and cognitive performance were assessed at baseline, immediately after training, and at 1- and 6-month follow-up. The incident rate of falls in the 6-month pre-intervention was compared with that in the 6-month post-intervention. Dual-task gait performance (gait speed, gait speed variability and stride length under cognitive dual task and obstacle negotiation, and the leading foot clearance in obstacle negotiation) improved similarly in both groups with gains sustained at 6-month follow-up. A higher decrease in fall rate and fear of falling were observed in participants assigned to the 12-week intervention than the 6-week intervention. Improvements in cognitive functions (i.e., executive functions, visuospatial ability, and attention) were seen only in participants enrolled in 12-week training up to 1-month follow-up but vanished at the 6-month evaluation. Our results suggest that a longer TT + VR training leads to greater improvements in cognitive functions especially those directly addressed by the virtual environment.

Short-Term Effects of Single-Session Split-Belt Treadmill Training on Dual-Task Performance in Parkinson's Disease and Healthy Elderly.

Background: Dual-tasking is challenging for people with Parkinson's disease and freezing of gait (PD+FOG) and can exacerbate freezing episodes and falls. Split-belt treadmill training (SBT) is a novel tool to train complex gait and may improve dual-task (DT) walking and turning. Objective: To investigate the single-session effects of SBT on DT walking and DT turning performance in PD+FOG and older adults (OA), compared to regular treadmill training. Methods: Forty-five PD+FOG and 36 OA participated in a single training session (30 min). They were randomized into one of four training groups: (A) SB75-steady belt speed ratio 0.75:1; (B) SB50-steady belt speed ratio 0.5:1; (C) SBCR-changing belt speed ratios between 0.75:1 and 0.5:1; and (D) Tied-Belt (TBT). Over-ground straight-line gait and an alternating turning in place task combined with a cognitive dual-task (DT) (auditory Stroop) were assessed pre- and post-training, and the following day (retention). Constrained longitudinal data analysis was used to investigate the training effects for all participants and for PD+FOG alone. Results: DT gait speed improved at post-training for all groups (p < 0.001). However, SBT (SB50 and SBCR) led to larger post-training improvements compared to TBT, which were still visible at retention (SB50). For mean DT turning speed and Stroop response time while walking, only SBT groups showed significant improvements at post-training or retention. DT stride length, peak DT turning speed, and Stroop performance index while walking also showed larger gains in SBT compared to TBT. Results for PD+FOG alone showed similar effects although with smaller effect sizes. Conclusions: A single session of SBT in PD+FOG and OA showed larger short-term effects on DT walking and turning compared to TBT. Cognitive DT performance was also improved in SBT, likely due to reduced cortical control of gait. These results illustrate the potential for SBT to improve DT during complex gait and possibly reduce fall risk in clinical and healthy populations.