Reliability of the sub-components of the instrumented timed up and go test in ambulatory children with traumatic brain injury and typically developed controls

Employing machine learning to enhance fracture recovery insights through gait analysis

This study aimed to explore the potential of gait analysis coupled with supervised machine learning models as a predictive tool for assessing post-injury complications such as infection, malunion, or hardware irritation among individuals with lower extremity fractures. We prospectively identified participants with lower extremity fractures at a tertiary academic center. These participants underwent gait analysis with a chest-mounted inertial measurement unit device. Using customized software, the raw gait data were preprocessed, emphasizing 12 essential gait variables. The data were standardized, and several machine learning models, including XGBoost, logistic regression, support vector machine, LightGBM, and Random Forest, were trained, tested, and evaluated. Special attention was given to class imbalance, addressed using the synthetic minority oversampling technique (SMOTE). Additionally, we introduced a novel methodology to compute the post-injury recovery rate for gait variables, which operates independently of the time difference between the gait analyses of different participants. XGBoost was identified as the optimal model both before and after the application of SMOTE. Before using SMOTE, the model achieved an average test area under the ROC curve (AUC) of 0.90, with a 95% confidence interval (CI) of [0.79, 1.00], and an average test accuracy of 86%, with a 95% CI of [75%, 97%]. Through feature importance analysis, a pivotal role was attributed to the duration between the occurrence of the injury and the initial gait analysis. Data patterns over time revealed early aggressive physiological compensations, followed by stabilization phases, underscoring the importance of prompt gait analysis. χ2 analysis indicated a statistically significant higher readmission rate among participants with underlying medical conditions (p = 0.04). Although the complication rate was also higher in this group, the association did not reach statistical significance (p = 0.06), suggesting a more pronounced impact of medical conditions on readmission rates rather than on complications. This study highlights the transformative potential of integrating advanced machine learning techniques like XGBoost with gait analysis for orthopedic care. The findings underscore a shift toward a data-informed, proactive approach in orthopedics, enhancing patient outcomes through early detection and intervention. The χ2 analysis added crucial insights into the broader clinical implications, advocating for a comprehensive treatment strategy that accounts for the patient's overall health profile. The research paves the way for personalized, predictive medical care in orthopedics, emphasizing the importance of timely and tailored patient assessments.

Effectiveness of Physical Therapy on Ataxia-Telangiectasia: A Case Report

BACKGROUND AND PURPOSE

This case report investigated the benefits of a 12-week physical therapy program for a child with ataxia-telangiectasia (AT).

CASE DESCRIPTION

A 9-year-old girl with a diagnosis of AT participated. The physical therapy program consisted of balance and strength exercise and Wii Fit Balance-based video games training with a pediatric physical therapist for 12 weeks.

MEASUREMENTS

The motor performance, Gross Motor Function Measurement (GMFM), Pediatric Berg Balance Scale (PBBS), Trunk Control Measurement Scale (TCMS), participation as measured by the Life Habits Questionnaire (LIFE-H), and the Pediatric Quality of Life Inventory (PedsQL).

OUTCOMES

Positive changes were observed in the TCMS, PBBS, GMFM, and motor performance, participation, and quality of life.

CONCLUSIONS

Notable improvements were observed in both body structure and function, and activities and participation level.

WHAT THIS ADDS TO EVIDENCE

This case report is the first to support the effectiveness of physical therapy in a child with AT.

Reliability and concurrent validity of a modified timed up and go test for healthy preschoolers

The purpose of this study was to assess the interrater and test-retest reliability and the concurrent validity of the modified timed up and go test for preschoolers. As such, we aim to determine the most suited outcome of the modified timed up and go test: the best or the average performance. Thirty-two children (age 3-5) performed three timed up and go test trials as fast as possible on two separate occasions. During the first session, two researchers recorded the time to perform the task simultaneously. For reliability analyses, intra-class correlation coefficients (ICCs) and the minimal detectable change were determined. A Pearson correlation coefficient was calculated to determine concurrent validity between the timed up and go test and the balance subscale of the Movement Assessment Battery for Children, 2nd edition. Interrater (ICC > 0.97) and test-retest (ICC > 0.75) reliability were good both for the average and the best timed up and go test performance. A minimal detectable change of 1.86 s was found for the best performance, and 2.30 s for the average performance. Only the best timed up and go test performance correlated significantly with the balance subscale of the Movement Assessment Battery for Children, 2nd edition, though fair (r = -0.347, p = 0.007).Conclusion: The modified timed up and go test for preschoolers using the best performance is reliable and recommended to reduce standard and measurement error. What is Known: • A large variety of timed up and go test protocols is available for children • The protocols differ in the instructions on walking speed (self-selected/fastest), the use of an extra motivation (e.g., touch a star on the wall) and the applied outcome (average/best performance) What is New: • The best timed up and go test performance induces more consistent test results between raters and sessions and also less standard and measurement error in 3- to 5-year-old children • The best timed up and go test performance should be preferred over the average performance to achieve both reliable and valid test results in 3- to 5-year-old children.

Normative database of spatiotemporal gait parameters using inertial sensors in typically developing children and young adults

BACKGROUND

Inertial sensors are increasingly useful to clinicians and researchers to detect gait deficits. Reference values are necessary for comparison to children with gait abnormalities.

OBJECTIVE

To present a normative database of spatiotemporal gait and turning parameters in 164 typically developing children and young adults ages 5-30 utilizing the APDM Mobility Lab® system.

METHODS

Participants completed the i-WALK test at both self-selected (SS) and fast as possible (FAP) walking speeds. Spatiotemporal gait and turning parameters included stride length, stride length variability, gait speed, cadence, stance, swing, and double support times, and foot strike, toe-off, and toe-out angles, turn duration, peak turn velocity and number of steps to turn.

RESULTS

Absolute stride length and gait speed increased with age. Normalized gait speed, absolute and normalized cadence, and stride length variability decreased with age. Normalized stride length and all parameters of gait cycle phase and foot position remained unaffected by age except for greater FSA in children 7-8. Foot position parameters in children 5-6 were excluded due to aberrant values and high standard deviations. Turns were faster in children ages 5-13 and 7-13 in the SS and FAP conditions, respectively. There were no differences in number of steps to turn. Similar trends were observed in the FAP condition except: normalized gait speed did not demonstrate a relationship with age and children ages 5-8 demonstrated increased stance and double support times and decreased swing time compared to children 11-13 and young adults (ages 5-6 only). Females ages 5-6 demonstrated increased stride length variability in the SS condition; males ages 7-8 and 14-30 ha d increased absolute stride length in the FAP condition. Similarities and differences were found between our values and previous literature.

SIGNIFICANCE

This normative database can be used by clinicians and researchers to compare abnormal gait patterns and responses to interventions.

Use of an instrumented dual-task timed up and go test in children with traumatic brain injury

BACKGROUND

Wearable sensors have allowed researchers to instrument tests of gait-related mobility, including the widely used timed 'up-and-go' test (TUG). Currently, there is a lack of instrumented test data on whether children with moderate to severe traumatic brain injury (TBI) perform differently on the TUG compared to typically developed (TD) controls during a cognitive-motor task.

RESEARCH QUESTION

The aim was to explore the effects of a cognitive-motor task on TUG subcomponents among children with TBI compared to TD children.

METHODS

This observational cross-sectional study included 12 children with moderate to severe TBI (6 males and 6 females, age 10.5 ± 1.5 years of age) and 10 age and sex-matched TD controls (5 males and 5 females, 10.4 ± 1.3 years of age). Each participant completed 6 trials of the TUG wearing a single inertial measurement unit sensor at a self-selected walking pace while listening to an array of 10 randomly presented single digits during each TUG trial.

RESULTS

Total time to complete the TUG was not significantly different between groups. The cognitive-motor task led to significantly lower mean turn and peak turn angular velocity values during the turn-around-the-cone and turn-before-sitting TUG subcomponents in children with TBI compared to the TD controls (p ≤ 0.05). Additionally, the cognitive-motor task led to significantly lower values for maximum torso flexion to extension angle, peak flexion and extension angular velocity and peak vertical acceleration for the sit-to-stand subcomponent (p < 0.05). Peak flexion angular velocity during the stand-to-sit subcomponent was lower for the TBI group compared to the children with TD (p < 0.05).

SIGNIFICANCE

The study provides new insights into the performance of complex gait-related mobility tasks in the context of an instrumented TUG among children with moderate to severe TBI. Our results highlight the potential benefits of outfitting pediatric inpatients with an IMU while completing the TUG.