Single- and dual-task gait performance and their diagnostic value in early-stage Parkinson's disease

Machine learning for early detection and severity classification in people with Parkinson's disease

Early detection of Parkinson's disease (PD) and accurate assessment of disease progression are critical for optimizing treatment and rehabilitation. However, there is no consensus on how to effectively detect early-stage PD and classify motor symptom severity using gait analysis. This study evaluated the accuracy of machine learning models in classifying early and moderate-stages of PD based on spatiotemporal gait features at different walking speeds. A total of 178 participants were recruited, including 103 individuals with PD (61 early-stage, 42 moderate-stage) and 75 healthy controls. Participants performed a walking test on a 24-m walkway at three speeds: preferred walking speed (PWS), 20% faster (HWS), and 20% slower (LWS). Key features-walking speed at PWS, stride length at HWS, and the coefficient of variation (CV) of the stride length at LWS-achieved a classification accuracy of 78.1% using the random forest algorithm. For early PD detection, the stride length at HWS and CV of the stride length at LWS provided an accuracy of 67.3% with Naïve Bayes. Walking at PWS was the most critical feature for distinguishing early from moderate PD, with an accuracy of 69.8%. These findings suggest that assessing gait over consecutive steps under different speed conditions may improve the early detection and severity assessment of individuals with PD.

Impaired dual-task gait in Parkinson's disease is associated with brain morphology changes

In Parkinson's disease (PD), impaired gait and cognition affect daily activities, particularly in the more advanced stages of the disease. This study investigated the relationship between gait parameters, cognitive performance, and brain morphology in patients with early untreated PD. 64 drug-naive PD patients and 47 healthy controls (HC) participated in the study. Single- and dual-task gait (counting task) were examined using an expanded Timed Up & Go Test measured on a GaitRite walkway. Measurements included gait speed, stride length, and cadence. A brain morphometry analysis was performed on T1-weighted magnetic resonance (MR) images. In PD patients compared to HC, gait analysis revealed reduced speed (p < 0.001) and stride length (p < 0.001) in single-task gait, as well as greater dual-task cost (DTC) for speed (p = 0.007), stride length (p = 0.014) and cadence (p = 0.029). Based on the DTC measures in HC, PD patients were further divided into two subgroups with normal DTC (PD-nDTC) and abnormally increased DTC (PD-iDTC). For PD-nDTC, voxel-based morphometric correlation analysis revealed a positive correlation between a cluster in the left primary motor cortex and stride-length DTC (r = 0.57, p = 0.027). For PD-iDTC, a negative correlation was found between a cluster in the right lingual gyrus and the DTC for gait cadence (r=-0.35, pFWE = 0.018). No significant correlations were found in HC. The associations found between brain morphometry and gait performance with a concurrent cognitive task may represent the substrate for gait and cognitive impairment occurring since the early stages of PD.

Non-invasive brain stimulation enhances motor and cognitive performances during dual tasks in patients with Parkinson's disease: a systematic review and meta-analysis

BACKGROUND

Parkinson's disease (PD) induces progressive deficits in motor and cognitive functions as well as impaired dual-task performance requiring both motor and cognitive functions. This systematic review and meta-analysis evaluated the effects of non-invasive brain stimulation (NIBS) on dual-task performance in patients with PD.

METHODS

11 studies met the following inclusion criteria: (a) patients with PD, (b) NIBS intervention, (c) comparison with the sham stimulation group, (d) motor and cognitive performance outcomes during dual tasks, and (e) randomized controlled trials with parallel or crossover designs. Individual effect size (i.e., comparison) was quantified by comparing motor and cognitive performances changes during dual tasks between active NIBS and sham stimulation conditions. Thus, higher values of the overall effect size indicate more improvements in either motor or cognitive performances after NIBS. Moreover, moderator variable analyses determined whether NIBS effects on dual-task performances differed depending on targeted brain regions. Finally, meta-regression analyses determined whether NIBS effects on dual-task performances were associated with demographic characteristics.

RESULTS

The random-effects model meta-analysis revealed that NIBS significantly improved motor (73 comparisons from 11 studies) and cognitive (12 comparisons from four studies) performances during dual tasks in patients with PD. Specifically, anodal transcranial direct current stimulation protocols on the dorsolateral prefrontal cortex were effective. Moreover, greater improvements in motor performance during dual tasks significantly correlated with decreased age and increased proportion of females, respectively.

CONCLUSION

This meta-analysis suggests that excitatory stimulation on the dorsolateral prefrontal cortex may be effective for improving dual-task performance in patients with PD.

A Review of Recent Advances in Cognitive-Motor Dual-Tasking for Parkinson's Disease Rehabilitation

BACKGROUND

Parkinson's disease is primarily characterized by the degeneration of motor neurons, leading to significant impairments in movement. Initially, physical therapy was predominantly employed to address these motor issues through targeted rehabilitation exercises. However, recent research has indicated that cognitive training can enhance the quality of life for patients with Parkinson's. Consequently, some researchers have posited that the simultaneous engagement in computer-assisted motor and cognitive dual-task (CADT) may yield superior therapeutic outcomes.

METHODS

A comprehensive literature search was performed across various databases, and studies were selected following PRISMA guidelines, focusing on CADT rehabilitation interventions.

RESULTS

Dual-task training enhances Parkinson's disease (PD) rehabilitation by automating movements and minimizing secondary task interference. The inclusion of a sensor system provides real-time feedback to help patients make immediate adjustments during training. Furthermore, CADT promotes more vigorous participation and commitment to training exercises, especially those that are repetitive and can lead to patient boredom and demotivation. Virtual reality-tailored tasks, closely mirroring everyday challenges, facilitate more efficient patient adaptation post-rehabilitation.

CONCLUSIONS

Although the current studies are limited by small sample sizes and low levels, CADT rehabilitation presents as a significant, effective, and potential strategy for PD.

Machine learning-based gait adaptation dysfunction identification using CMill-based gait data

Background

Combining machine learning (ML) with gait analysis is widely applicable for diagnosing abnormal gait patterns.

Objective

To analyze gait adaptability characteristics in stroke patients, develop ML models to identify individuals with GAD, and select optimal diagnostic models and key classification features.

Methods

This study was investigated with 30 stroke patients (mean age 42.69 years, 60% male) and 50 healthy adults (mean age 41.34 years, 58% male). Gait adaptability was assessed using a CMill treadmill on gait adaptation tasks: target stepping, slalom walking, obstacle avoidance, and speed adaptation. The preliminary analysis of variables in both groups was conducted using t-tests and Pearson correlation. Features were extracted from demographics, gait kinematics, and gait adaptability datasets. ML models based on Support Vector Machine, Decision Tree, Multi-layer Perceptron, K-Nearest Neighbors, and AdaCost algorithm were trained to classify individuals with and without GAD. Model performance was evaluated using accuracy (ACC), sensitivity (SEN), F1-score and the area under the receiver operating characteristic (ROC) curve (AUC).

Results

The stroke group showed a significantly decreased gait speed (p = 0.000) and step length (SL) (p = 0.000), while the asymmetry of SL (p = 0.000) and ST (p = 0.000) was higher compared to the healthy group. The gait adaptation tasks significantly decreased in slalom walking (p = 0.000), obstacle avoidance (p = 0.000), and speed adaptation (p = 0.000). Gait speed (p = 0.000) and obstacle avoidance (p = 0.000) were significantly correlated with global F-A score in stroke patients. The AdaCost demonstrated better classification performance with an ACC of 0.85, SEN of 0.80, F1-score of 0.77, and ROC-AUC of 0.75. Obstacle avoidance and gait speed were identified as critical features in this model.

Conclusion

Stroke patients walk slower with shorter SL and more asymmetry of SL and ST. Their gait adaptability was decreased, particularly in obstacle avoidance and speed adaptation. The faster gait speed and better obstacle avoidance were correlated with better functional mobility. The AdaCost identifies individuals with GAD and facilitates clinical decision-making. This advances the future development of user-friendly interfaces and computer-aided diagnosis systems.

Cholinergic nucleus degeneration and its association with gait impairment in Parkinson's disease

BACKGROUND

The contribution of cholinergic degeneration to gait disturbance in Parkinson's disease (PD) is increasingly recognized, yet its relationship with dopaminergic-resistant gait parameters has been poorly investigated. We investigated the association between comprehensive gait parameters and cholinergic nucleus degeneration in PD.

METHODS

This cross-sectional study enrolled 84 PD patients and 69 controls. All subjects underwent brain structural magnetic resonance imaging to assess the gray matter density (GMD) and volume (GMV) of the cholinergic nuclei (Ch123/Ch4). Gait parameters under single-task (ST) and dual-task (DT) walking tests were acquired using sensor wearables in PD group. We compared cholinergic nucleus morphology and gait performance between groups and examined their association.

RESULTS

PD patients exhibited significantly decreased GMD and GMV of the left Ch4 compared to controls after reaching HY stage > 2. Significant correlations were observed between multiple gait parameters and bilateral Ch123/Ch4. After multiple testing correction, the Ch123/Ch4 degeneration was significantly associated with shorter stride length, lower gait velocity, longer stance phase, smaller ankle toe-off and heel-strike angles under both ST and DT condition. For PD patients with HY stage 1-2, there were no significant degeneration of Ch123/4, and only right side Ch123/Ch4 were corrected with the gait parameters. However, as the disease progressed to HY stage > 2, bilateral Ch123/Ch4 nuclei showed correlations with gait performance, with more extensive significant correlations were observed in the right side.

CONCLUSIONS

Our study demonstrated the progressive association between cholinergic nuclei degeneration and gait impairment across different stages of PD, and highlighting the potential lateralization of the cholinergic nuclei's impact on gait impairment. These findings offer insights for the design and implementation of future clinical trials investigating cholinergic treatments as a promising approach to address gait impairments in PD.

The impact of anxiety on gait impairments in Parkinson's disease: insights from sensor-based gait analysis

BACKGROUND

Sensor-based gait analysis provides a robust quantitative tool for assessing gait impairments and their associated factors in Parkinson's disease (PD). Anxiety is observed to interfere with gait clinically, but this has been poorly investigated. Our purpose is to utilize gait analysis to uncover the effect of anxiety on gait in patients with PD.

METHODS

We enrolled 38 and 106 PD patients with and without anxiety, respectively. Gait parameters were quantitively examined and compared between two groups both in single-task (ST) and dual-task (DT) walking tests. Multiple linear regression was applied to evaluate whether anxiety independently contributed to gait impairments.

RESULTS

During ST, PD patients with anxiety presented significantly shorter stride length, lower gait velocity, longer stride time and stance time, longer stance phase, smaller toe-off (TO) and heel-strike (HS) angles than those without anxiety. While under DT status, the differences were diminished. Multiple linear regression analysis demonstrated that anxiety was an independent factor to a serials of gait parameters, particularly ST-TO (B = -2.599, (-4.82, -0.38)), ST-HS (B = -2.532, (-4.71, -0.35)), ST-TO-CV (B = 4.627, (1.71, 7.64)), ST-HS-CV(B = 4.597, (1.66, 7.53)), ST stance phase (B = 1.4, (0.22, 2.58)), and DT stance phase (B = 1.749, (0.56, 2.94)).

CONCLUSION

Our study discovered that anxiety has a significant impact on gait impairments in PD patients, especially exacerbating shuffling steps and prolonging stance phase. These findings highlight the importance of addressing anxiety in PD precision therapy to achieve better treatment outcomes.