Search

Background: Accurately predicting the walking independence of stroke patients is important. Our objective was to determine and compare the performance of logistic regression (LR) and three machine learning models (eXtreme Gradient Boosting (XGBoost), Support Vector Machines (SVM), and Random Forest (RF)) in predicting walking independence at discharge in stroke patients, as well as to explore the variables that predict prognosis. Methods: 778 (80% for the training set and 20% for the test set) stroke patients admitted to China Rehabilitation Research Center between February 2020 and January 2023 were retrospectively included. The training set was used for training models. The test set was used to validate and compare the performance of the four models in terms of area under the curve (AUC), accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and F1 score. Results: Among the three ML models, the AUC of the XGBoost model is significantly higher than that of the SVM and RF models (P < 0.001, P = 0.024, respectively). There was no significant difference in the AUCs between the XGBoost model and the LR model (0.891 vs. 0.880, P = 0.560). The XGBoost model demonstrated superior accuracy (87.82% vs. 86.54%), sensitivity (50.00% vs. 39.39%), PPV (73.68% vs. 73.33%), NPV (89.78% vs. 87.94%), and F1 score (59.57% vs. 51.16%), with only slightly lower specificity (96.09% vs. 96.88%). Together, the XGBoost model and the stepwise LR model identified age, FMA-LE at admission, FAC at admission, and lower limb spasticity as key factors influencing independent walking. Conclusion: Overall, the XGBoost model performed best in predicting independent walking after stroke. The XGBoost and LR models together confirm that age, admission FMA-LE, admission FAC, and lower extremity spasticity are the key factors influencing independent walking in stroke patients at hospital discharge. Trial registration: Not applicable. [ABSTRACT FROM AUTHOR]

Background: Accurately predicting the walking independence of stroke patients is important. Our objective was to determine and compare the performance of logistic regression (LR) and three machine learning models (eXtreme Gradient Boosting (XGBoost), Support Vector Machines (SVM), and Random Forest (RF)) in predicting walking independence at discharge in stroke patients, as well as to explore the variables that predict prognosis. Methods: 778 (80% for the training set and 20% for the test set) stroke patients admitted to China Rehabilitation Research Center between February 2020 and January 2023 were retrospectively included. The training set was used for training models. The test set was used to validate and compare the performance of the four models in terms of area under the curve (AUC), accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and F1 score. Results: Among the three ML models, the AUC of the XGBoost model is significantly higher than that of the SVM and RF models (P < 0.001, P = 0.024, respectively). There was no significant difference in the AUCs between the XGBoost model and the LR model (0.891 vs. 0.880, P = 0.560). The XGBoost model demonstrated superior accuracy (87.82% vs. 86.54%), sensitivity (50.00% vs. 39.39%), PPV (73.68% vs. 73.33%), NPV (89.78% vs. 87.94%), and F1 score (59.57% vs. 51.16%), with only slightly lower specificity (96.09% vs. 96.88%). Together, the XGBoost model and the stepwise LR model identified age, FMA-LE at admission, FAC at admission, and lower limb spasticity as key factors influencing independent walking. Conclusion: Overall, the XGBoost model performed best in predicting independent walking after stroke. The XGBoost and LR models together confirm that age, admission FMA-LE, admission FAC, and lower extremity spasticity are the key factors influencing independent walking in stroke patients at hospital discharge. Trial registration: Not applicable. [ABSTRACT FROM AUTHOR]

Background: Pavlik harness is the most widely used orthosis in the treatment of developmental dysplasia of the hip. The aim of this study was to evaluate the effect of the Pavlik harness on the development of "unaided sitting" and "independent walking" in infants with developmental dysplasia of the hip. Methods: This prospective study, conducted from 2017 to 2020, included infants undergoing Pavlik harness therapy. Inclusion criteria comprised gestational age > 37 weeks, treatment initiation before 6 months of age, and no prior treatment for developmental dysplasia of the hip. We assessed treatment initiation age, treatment duration, and the age of achieving unaided sitting and independent walking. Results: In the patient group, unaided sitting commenced at a mean age of 6.8 ± 1.6 (range: 4–11) months, while independent walking began at a mean age of 12.7 ± 1.8 (range: 9–18) months. By 15 months, 92% of the patients achieved independent walking. In the control group, unaided sitting occurred at a mean age of 6.1 ± 1.1 (range: 4–8) months, and independent walking at 11.8 ± 1.6 (range: 9–18) months. A significant positive correlation was observed between the duration of Pavlik harness usage and the age of unaided sitting (p < 0.001) and independent walking (p < 0.001). Conclusion: Our study indicates that Pavlik harness treatment for developmental dysplasia of the hip is generally safe and does not lead to clinically significant delays in unaided sitting and independent walking. However, some minor delays may occur due to extended orthosis use. Level of evidence: level III—prospective cohort study. [ABSTRACT FROM AUTHOR]

Stroke rehabilitation with mechanical assistance improves outcomes by facilitating repetition and relieving the care burden of therapy staff. Here, we tested the Medical Care Pit (MCP) walking assistance training device in the rehabilitation of eight acute stroke patients (median age 60.7 ± 16.3 years) who had recently suffered ischemic (three) or hemorrhagic (five) stroke (14.1 ± 6.5 days). Patients received standard rehabilitation approximately 5 days per week (weekdays only), plus MCP therapy twice a week, totaling four MCP sessions over 2 weeks. Fugl–Meyer Assessment-Lower Extremities (FMA-LE), Functional Ambulation Category (FAC), and other gait-associated parameters were measured. Over the 10.5 ± 1.6 days of therapy, MCP qualitatively assisted in gait analysis and real-time patient feedback while independent walking scores significantly improved (FAC 2.2 ± 0.8 to 3.1 ± 1.3, p = 0.020). FMA-LE scores also slightly improved but not to significance (p = 0.106). Objective burden on patients, as measured by modified Borg scale, was significantly improved (2.7 ± 1.6 to 2.0 ± 1.6, p = 0.014). In terms of questionnaires, anxiety scores for the physical therapist regarding gait training and falling with MCP significantly decreased (3.8 ± 2.3 to 1.0 ± 1.6; p = 0.027 and 3.1 ± 2.2 to 0.8 ± 1.3; p = 0.045) from the first to fourth sessions. Taken together, MCP, in addition to the usual rehabilitation program, was effective in gait rehabilitation for independent walking and relieved burdens on the patients. Such walking support systems may be an important part of acute stroke rehabilitation. [ABSTRACT FROM AUTHOR]

Background: Onset of walking in infants leads to regular cyclic loading of the plantar foot surface for the first time. This is a critical period for evolving motor skills and foot structure and function. Plantar pressure literature typically studies gait only once walking is established and under conditions that artificially constrain the walking direction and bouts compared to how infants move in the real-world. We therefore do not know how the foot is loaded when self-directed walking is first achieved and whether it changes as walking is practiced. Research question How do pressures on the plantar foot in real-world walking change from new to confident walking? Methods Fifty-seven infants participated in a two-site longitudinal study. Bespoke child-friendly spaces incorporated large pressure platforms and video. Data was collected at two milestones: new (403 days) and confident (481 days) walking. Steps were defined as walking straight or turning medially/laterally. Pressure variables were calculated for eight-foot regions and compared between milestones. Results Confident walking resulted in more steps (median: 18 v 35) and almost twice as many turning steps. During straight-line steps, confident walking increased peak pressures in the medial heel (median: 99.3 v 106.7kPa, p < .05) and lateral forefoot (median: 53.9 v 65.3kPa, p < .001) and reduced medial toe pressure (median: 98.1 v 80.0kPa, p < .05). Relative medial midfoot contact area reduced (median: 12.4 v 11.2%, p < .05) as absolute foot contact increased. A faster transition across stance and a reduced relative contact time in the forefoot were recorded in confident walking. Significance Pressures change rapidly as walking is initiated with significant differences in foot loading evident within an average 77 days. Importantly, these changes differ in straight and turning walking. Continued reliance on assessment of straight-line walking during early stages of ambulation likely fails to characterise 26% of steps experienced by infant feet. [ABSTRACT FROM AUTHOR]

Introduction: Trochanteric fractures are one of the most common fragility fractures, and the number of the patients is increasing worldwide. Identifying prognostic factors is important to manage and minimize the social losses caused by postoperative walking impairment. The purpose of this retrospective study is to clarify the association between early postoperative Barthel index score and postoperative long-term walking ability among patients with trochanteric fractures who could walk independently before injury. Materials and methods: Consecutive patients with trochanteric fractures aged ≥ 65 years who were walking independently before injury were included. Patients were divided into two groups according to the Barthel index score measured within 1 week after surgery; the cut-off value was set at 20 points. The prevalence of walking impairment after 1 year was compared between these two groups after adjusting for confounding factors in a complete case analysis and using the data introduced by the multiple imputation method. Results: Among the 758 eligible patients, 254 patients (34%) had their walking ability evaluated after 1 year from injury. The patients in the lower Barthel index group showed a significantly impaired walking ability in both analyses (adjusted odds ratio 5.5 and 2.4, 95% confidence intervals 2.4–13 and 1.5–3.8, respectively). Conclusions: The present results suggested that the Barthel index score measured in the early postoperative period after trochanteric fractures was associated with the level of long-term walking impairment in patients who could walk independently before injury. [ABSTRACT FROM AUTHOR]

This longitudinal study investigated the impact of the first independent steps on harmonic gait development in unilateral cerebral palsy (CP) and typically developing (TD) children. We analysed the gait ratio values (GR) by comparing the duration of stride/stance, stance/swing and swing/double support phases. Our investigation focused on identifying a potential trend towards the golden ratio value of 1.618, which has been observed in the locomotion of healthy adults as a characteristic of harmonic walking. Locomotor ability was assessed in both groups at different developmental stages: before and after the emergence of independent walking. Results revealed that an exponential fit was observed only after the first unsupported steps were taken. TD children achieved harmonic walking within a relatively short period (approximately one month) compared to children with CP, who took about seven months to develop harmonic walking. Converging values for stride/stance and stance/swing gait ratios, averaged on the two legs, closely approached the golden ratio in TD children (R2 = 0.9) with no difference in the analysis of the left vs right leg separately. In contrast, children with CP exhibited a trend for stride/stance and stance/swing (R2 = 0.7), with distinct trends observed for the most affected leg which did not reach the golden ratio value for the stride/stance ratio (GR = 1.5), while the least affected leg exceeded it (GR = 1.7). On the contrary, the opposite trend was observed for the stance/swing ratio. These findings indicate an overall harmonic walking in children with CP despite the presence of asymmetry between the two legs. These results underscore the crucial role of the first independent steps in the progressive development of harmonic gait over time. • We hypothesized first steps are linked to the development of gait harmony in both Typically Developing (TD) and children with Cerebral Palsy (CP). • Gait ratios (GR) phases were investigated as a function of the time since onset of independent steps and compared to the golden ratio (Phi ≈ 1.618...). • An exponential fit was applied to inspect the development of harmonicity of gait ratios towards the established golden ratio of adult locomotion. • A Good fit model (R2 > 0.5) was found after the first steps in TD group for all the GR indices, while in CP group only for GR1 (stance-to-swing ratio). • CP group exhibited different GR indices between legs suggesting a compensatory behaviour underlying a functional walking pattern. [ABSTRACT FROM AUTHOR]

Abstract: Background: One objective of a neonatal follow-up program is to examine and predict gross motor outcome of infants born preterm. Aims: To assess the concurrent validity of the Test of Infant Motor Performance (TIMP) and the Alberta Infant Motor Scale (AIMS), the ability to predict gross motor outcome around 15months corrected age (CA), and to explore factors associated with the age of independent walking. Methods: 95 infants, born at a gestational age <30weeks, were assessed around 3, 6 and 15months CA. At 3months CA, correlations of raw-scores, Z-scores, and diagnostic agreement between TIMP and AIMS were determined. AIMS-score at 15months CA and parental-reported walking age were outcome measures for regression analyses. Results: The correlation between TIMP and AIMS raw-scores was 0.82, and between Z-scores 0.71. A cut-off Z-score of −1.0 on the TIMP had 92% diagnostic agreement (κ=0.67) with an AIMS-score

Abstract: Productive vocabulary was measured every four months in 16- to 28-months-olds. Attainment of motor-milestones was also measured. An earlier age of sitting and walking predicted a higher intercept and a larger slope (growth) of productive vocabulary respectively, suggesting that attainment of walking propels linguistic development. [Copyright &y& Elsevier]

Abstract: The onset of crawling marks a motor, cognitive and social milestone. The present study investigated whether independent walking marks a second milestone for social behaviors. In Experiment 1, the social and exploratory behaviors of crawling infants were observed while crawling and in a baby-walker, resulting in no differences based on posture. In Experiment 2, the social behaviors of independently walking infants were compared to age-matched crawling infants in a baby-walker. Independently walking infants spent significantly more time interacting with the toys and with their mothers, and also made more vocalizations and more directed gestures compared to infants in the walker. Experiment 3 tracked infants’ social behaviors longitudinally across the transition from crawling and walking. Even when controlled for age, the transition to independent walking marked increased interaction time with mothers, as well as more sophisticated interactions, including directing mothers’ attention to particular objects. The results suggest a developmental progression linking social interactions with milestones in locomotor development. [Copyright &y& Elsevier]

Aims: The aims of this study is to (i) determine the age of sitting unsupported and independent walking in preterm infants with birth weight under 1500 g (very low birth weight, VLBW); (ii) estimate differences between VLBW children and a reference population and (iii) estimate the association between clinical characteristics and late age at sitting and walking. Methods: A longitudinal study was conducted of a cohort of 876 children with VLBW. The World Health Organization (WHO) motor development study population was used as a reference. Ages for both skills were established by interview with parents. Means were compared with t-test, ANOVA and Bonferroni adjustment where appropriate. Results: The inclusion criteria were complied with 694 patients; 50% of VLBW sat at 7 m corrected age (CA) and walked at 13 m CA. Both motor skills were acquired later (7.3 ± 1.5 and 13.6 ± 2.8 m) compared with the control group (6 ± 1.1 and 12.1 ± 1.8 m). Weight or head circumference at birth below the 10th percentile or the presence of bronchopulmonary dysplasia were associated with delayed acquisition of both skills. Conclusion: Very low birth weight infants typically sit unsupported and walk later than term infants. Tables describing reference values for milestones acquisition for different categories of infants (gestational age, birth weight and other determinants) may contribute to inform the decision making process on access to available resources. [ABSTRACT FROM AUTHOR]