Predicting gait kinematics in youth with cerebral palsy using clinically informed machine learning algorithms.

Validation, characterization, and utility of markerless motion capture in a large cohort of pediatric patients with complex gait patterns.

Torsional abnormalities of the femur, such as femoral retrotorsion, can significantly impact gait and joint biomechanics in children and adolescents. However, the resulting gait patterns as well as the relationship between static imaging measurements and dynamic gait deviations remain unclear. This study aimed to investigate biomechanical differences in gait between young individuals with true femoral retrotorsion and typically developing peers, examining correlations between gait deviations, clinical measurements, and imaging data. A multicenter retrospective study was conducted, including 31 patients with unilateral or bilateral femoral retrotorsion and a control group of typically developing children. Femoral torsion was assessed using MRI or CT, while 3D motion capture was used for gait analysis. Clinical hip rotation was measured, and statistical analyses, including correlation assessments, were performed to evaluate relationships between imaging, clinical, and gait parameters. Patients exhibited significantly increased hip external rotation and out-toeing gait compared with controls. However, no correlation was found between femoral torsion measured by imaging and hip rotation during gait or foot progression angle. In contrast, the midpoint of clinical hip rotation correlated moderately with both hip rotation and foot progression angle during gait. In addition, patients had higher BMI and demonstrated compensatory knee valgus, despite no radiologic evidence of knee malalignment. Children and adolescents with femoral retrotorsion exhibit distinct gait deviations, but static imaging alone does not reliably predict dynamic gait patterns. This emphasizes the need for dynamic assessment in clinical decision-making.

Automated gait classification: Comparison of automated algorithms to expert classification.

Comparative effects of bilateral and unilateral ankle-foot orthoses on balance and gait in children with spastic unilateral cerebral palsy.

Relationships between knee and hip adduction moment, static leg torsion and dynamic frontal and transversal plane gait parameters in children and adolescents.

SPD-Net: A semantic partitioned transformer with dynamic graph network for improved skeleton-based gait recognition.

The coronal plane alignment of the knee (CPAK) classification provides an objective framework to describe alignment changes after total knee arthroplasty (TKA). Although preserving the native phenotype improves intraoperative soft-tissue balance in varus and neutral knees, its functional benefits remain uncertain. We therefore evaluated whether preserving the preoperative CPAK phenotype in varus and neutral knees is associated with superior early postoperative gait patterns following TKA. We prospectively enrolled 44 patients who had varus and neutral coronal alignment who underwent robotic-assisted TKA performed by five fellowship-trained surgeons. Spatiotemporal gait parameters were evaluated preoperatively and at six weeks postoperatively using a pressure mat that captured velocity, cadence, step length, and support time. Patients were grouped according to whether their postoperative alignment preserved or altered their native CPAK phenotype, as measured in long-leg biplane radiographs. Between-group differences were assessed using analysis of covariance adjusted for confounding and baseline gait values. Of the 44 patients, 21 (47.7%) had their native CPAK phenotype preserved, and 23 (52.3%) experienced a phenotype change. Preserving CPAK was associated with significantly faster postoperative walking velocity (+11.7 cm/s, P = 0.01), higher cadence (+7.0 steps/minute, P = 0.02), longer single-limb support (+1.5%, P = 0.003), and shorter step time (P = 0.02) compared with patients whose CPAK phenotype changed. Preoperative-to-postoperative gait changes analysis further confirmed this pattern: the changed group demonstrated significant reductions in walking velocity (P = 0.04) and cadence (P = 0.01), along with increased step time (P = 0.02), whereas the unchanged group showed significant improvements in step length (P = 0.02), greater single-limb support (P < 0.01), and reduced double-limb support (P < 0.01). Preserving the native CPAK phenotype after TKA is associated with improved early gait metrics, suggesting a potential biomechanical advantage of restoring constitutional alignment. These findings should be confirmed in larger cohorts with longer-term follow-up.

BackgroundGait patterns in children with myelomeningocele (MMC) at various neurological levels have been described, both with and without orthotic support. Although the neurological level of the lesion serves as an important predictor of ambulatory potential, the expected walking ability is not always achieved, as additional factors such as spasticity may influence gait negatively.The aim of this study was to retrospectively compare gait patterns as assessed in childhood with those observed in adulthood.MethodsOf 59 individuals with MMC aged 18 years or older, 29 had undergone three-dimensional gait analysis in childhood (Ch-GA). These data were retrospectively analysed and compared with findings from a subsequent adult gait analysis (Ad-GA). The mean (standard deviation) age at the time of Ch-GA was 11.6 (4.1) years and at Ad-GA 25.9 (3.9) years. The median (range) interval between assessments was 15.0 years (5.1–17.2).ResultsTwenty-two participants maintained independent, non-assisted walking (Group A), 5 had transitioned from independent walking to using a walking aid (Group B), and 2 used a walking aid at both Ch-GA and Ad-GA (Group C), with individualized orthotic prescriptions provided at both time points. In Group A, two of eleven kinematic variables and six of eleven kinetic variables in the hip, knee, and ankle showed deterioration, and walking speed had decreased. Functional ambulation declined from 18 community ambulators and 4 household ambulators (Ha) in childhood to 8 and 14, respectively, in adulthood. In Group B, analysed with only Gait Deviation Index (GDI), values were unchanged, but all temporospatial gait parameters had deteriorated. Functional ambulation decreased from five individuals classified as Ha to two Ha and three non-functional ambulators. The two individuals in Group C, who used a walker at both assessments, largely maintained the same GDI values and temporospatial parameters as in childhood.ConclusionLargely consistent with our original expectations, the findings indicate that gait patterns remain relatively stable from childhood to adulthood in individuals with MMC when supported by appropriate rehabilitation interventions, though some deterioration of gait and ambulation occurred. The results reflect gait-related changes that can be expected during growth in this population.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12883-026-05094-y.

Gait impairment is a hallmark symptom of Parkinson's Disease (PD). Traditional clinical assessments cannot capture real-world motor fluctuations, as they are sparsely performed. We validated the use of nearables, passive sensing technologies, including Kinect RGB-D cameras and ultra-wideband (UWB) radar, for continuous, objective assessment of gait fluctuations in PD within a home-like setting. Fifteen PD patients with mild symptoms and fourteen age- and sex-matched healthy controls (HC) performed 4-metre walking tasks in a living lab facility. Patients repeated the task during "ON" and "OFF" states of their daily medication cycle. Gait features, including stride length, stride time, and gait speed, were extracted from Kinect, radar, and a ground-truth smart floor. Data were analysed to assess inter-sensor agreements and group-level differences. Stride time demonstrated the highest agreement between devices (r = 0.903), while stride length was weaker (r = 0.779). Nevertheless, stride length from both Kinect and radar distinguished PD OFF from HC (camera q = 0.020; radar q = 0.005), and radar additionally differentiated ON from OFF (q = 0.020). Neither device differentiated PD ON from HC, indicating medication reduced observable gait differences. Although some spatial metrics show device discrepancies, both systems demonstrate sensitivity to gait patterns and medication-dependent changes, supporting their use for longitudinal, real-world monitoring of motor symptoms.

Individuals with transfemoral amputation (TFA) often experience gait adaptations when walking under environmental and cognitive challenges. This study examined the influence of passive prosthetic knee mechanisms on self-reported mobility and gait biomechanics during single-task and dual-task walking over an uneven compliant surface. Participants were grouped as able-bodied individuals (AB; n = 15), individuals with TFA using polycentric knees (TFAP; n = 6), and individuals with TFA using fluid-controlled knees (TFAFC; pneumatic/hydraulic; n = 5). The dual-task condition involved walking while performing an auditory Stroop task. Outcome measures included self-report questionnaires, joint kinematics, joint kinetics, and ground reaction forces. Biomechanical variables were analysed using linear mixed-effects models to examine group, task, and group × task interaction effects. Self-reported outcomes indicated reduced functional mobility in individuals with TFA compared with AB. Biomechanical findings showed altered gait patterns in both TFA groups compared with AB, particularly reduced knee flexion and ankle plantarflexion. For kinetic outcomes, significant group effects were observed for selected joint moment and power variables, although Bonferroni-adjusted pairwise differences were identified only for maximum stance hip extension moment (HM1), with lower values in TFAFC compared with AB. A significant task effect was observed for maximum stance ankle dorsiflexion (AA2) during dual-task walking across all groups. A significant group × task interaction was found for maximum hip eccentric power during late stance (HP2), with increased hip power during dual-task walking in TFAP. Overall, passive prosthetic knee mechanisms showed limited differences, although task-specific adaptations were evident under combined uneven surface and cognitive demands.

Following the footsteps: wearable sensors quantify principal component informed kinetic gait outcomes before and after knee arthroplasty.

A modeling approach to understanding poor stability in people with vestibular hypofunction.

Inter-joint coordination and coordination variability during gait differentiate copers from noncopers in males with anterior cruciate ligament deficiency.

A predictive simulation framework for personalised in silico gait retraining in knee osteoarthritis.

ABSTRACT Flexible electronic skin (E‐skin) represents a core component of wearable health monitoring systems, offering substantial potential in intelligent healthcare applications. However, the development of E‐skin materials that integrate high mechanical strength, excellent conductivity, and sustainability remains a formidable challenge. Herein, inspired by human skin, we develop a high‐strength and highly conductive bio‐based polyurethane urea eutectic gel (BPUU‐E), fabricated by incorporating bio‐based polyurethane urea (BPUU) into a zinc chloride/ethylene glycol (ZnCl 2 /EG)‐based metal salt deep eutectic solvent (MDES). Leveraging the “trident” coordination sites within BPUU molecular chains, a dual dynamic crosslinking network of “multiple hydrogen bonds—metal coordination bonds” is constructed, endowing the material with excellent mechanical properties (fracture strength up to 83.2 MPa, fracture toughness of 556.6 MJ m −3 ) and ionic conductivity (43.6 mS m −1 ). The flexible strain sensor based on BPUU‐E can stably monitor human joint movements and dynamic injury repair. Furthermore, integrating a sensing array and combining the Internet of Things and machine learning, enables high‐precision recognition of gait patterns. This study presents an innovative strategy for developing flexible E‐skin that possesses high strength, high conductivity, sustainability, and has significant application prospects in the field of intelligent wearable healthcare.

Osteogenesis imperfecta (OI) is characterized by substantial genetic and functional heterogeneity, yet low-burden objective tools for repeated gait monitoring in routine outpatient care remain limited. This study investigated whether routine smartphone videos combined with markerless pose estimation could capture genotype-associated gait patterns in children with IFITM5- and WNT1-related OI, with COL1-related OI included as a cross-sectional reference cohort. This single-center retrospective cohort study screened 237 pediatric patients with genetically confirmed IFITM5- or WNT1-related OI between 2014 and 2024. The main analytic cohort included 71 patients: 48 with IFITM5 mutations and 23 with WNT1 mutations. During the same period, 107 patients with COL1-related OI were screened, of whom 26 were included as a cross-sectional reference cohort. MediaPipe Pose was used to extract spatiotemporal and postural gait parameters from outpatient walking videos. Algorithmic consistency under simulated video perturbations was assessed, and gait phenotypes were evaluated using multivariable models, linear mixed-effects models, Kaplan-Meier analysis, and exploratory machine-learning analysis. Step_Width and Trunk_Lean showed high algorithmic consistency under perturbation testing, with ICC ranges of 0.966-0.980 and 0.923-0.983, respectively. At baseline, WNT1 patients showed greater Step_Width (0.265 ± 0.073 vs. 0.203 ± 0.062m), Trunk_Lean (4.53° ± 2.10° vs. 2.72° ± 1.63°), and Normalized_Step_Width (0.377 ± 0.103 vs. 0.270 ± 0.083) than IFITM5 patients (all P < 0.001). These differences remained after adjustment for age, sex, documented medication exposure, and surgical history, with adjusted differences of 0.046m for Step_Width, 1.612° for Trunk_Lean, and 0.087 for Normalized_Step_Width. In the COL1-related reference analysis, WNT1 patients retained higher Step_Width and Normalized_Step_Width than COL1-related OI patients, whereas Trunk_Lean did not differ significantly between these groups. Longitudinal analysis showed no significant age-related gait changes in the WNT1 group, whereas IFITM5 patients showed mild annual decreases in Step_Width (- 0.0026m/year) and Normalized_Step_Width (- 0.0046/year). Surgery-free survival declined earlier in the WNT1 group than in the IFITM5 group (log-rank P = 0.007). The exploratory genotype classification model achieved an AUC of 0.804 (95% CI, 0.678-0.914). Routine outpatient smartphone videos combined with markerless pose estimation can capture genotype-associated gait patterns in pediatric OI. WNT1-related OI was characterized by a wider base of support and coronal-plane compensation, whereas IFITM5-related OI showed mild longitudinal changes in spatial gait metrics. Video-derived gait parameters may provide low-burden adjunctive measures for outpatient functional monitoring, although prospective reference-standard validation and larger multi-genotype cohorts are needed.

Background and ObjectivesAutosomal Recessive Spastic Ataxia of Charlevoix-Saguenay (ARSACS) is a neurodegenerative disease caused by pathogenic variants in SACS. ARSACS is characterized by mitochondrial abnormalities and disruptions of the neurofilament cytoskeleton. In other conditions, these features have been linked to activation of Sterile Alpha and TIR Motif Containing 1 (SARM1), an enzyme that can trigger axon degeneration and neuronal death. Inhibition of SARM1 is an attractive therapeutic strategy because SARM1 is inactive in healthy cells and knockout of SARM1 has little or no deleterious effects. We therefore asked whether SARM1 activity contributed to Purkinje cell degeneration and motor defects present in a Sacs−/− mouse model of ARSACS.MethodsWe studied 4 cohorts of mice: (1) Sacs−/−; Sarm1+/+; (2) Sacs+/+; Sarm1+/−; (3) Sacs−/−; Sarm1+/−; and (4) Sacs−/−; Sarm1−/−. In 9-month-old mice, we analyzed protein markers of Purkinje cells (n = 3–4 mice per genotype) and counted surviving Purkinje cells in folia III, IV, and VIII of cerebellar sections (n = 3 mice per genotype, 6 sections per mouse), and tested motor function at 3, 6, and 9 months by quantifying parameters of gait (Digigait) and coordination and balance (Rotarod) (8 male, 8 female mice of each genotype).ResultsProbing of cerebellar extracts showed that the Purkinje cell protein markers Calbindin-1, RGS8, and PCP2 were decreased in Sacs−/− mice but restored to normal levels in Sacs−/−; Sarm1−/− mice. Purkinje cell loss in Sacs−/− mice was most prominent in anterior folia, as previously noted. Sarm1 loss partially mitigated the Purkinje cell death in folium III of 9-month-old Sacs−/− mice. Similarly, longitudinal behavioral assessment of motor functions showed that disturbances in gait pattern (slower cadence, prolonged swing, and stance phases) were partially alleviated. Rotarod tests gave more ambivalent results, as the homozygous loss of Sarm1 was less effective than heterozygous loss in ameliorating the Sacs−/− phenotype.DiscussionWe conclude that SARM1 contributes to neurodegeneration in ARSACS, and its downregulation or inhibition could constitute a significant therapeutical strategy in the treatment of the disease.

Evidence-based guidelines for motor rehabilitation in Hereditary Spastic Paraplegia (HSP) are still lacking but robotic gait training may offer benefits for motor function and walking. The aim of this study is to evaluate the effects of a robot-assisted gait training (RAGT) combined with traditional treatment (TT) on gait pattern, motor skills and quality of life in subjects affected by HSP. 40 patients (33 pure; 20 pediatric) underwent an intensive program with RAGT and TT daily (five-times a week for three consecutive weeks). Patients were evaluated pre and post intervention with gait analysis, gross-motor and timed walking tests and quality of life questionnaires. A three-month follow-up was conducted. A statistically significant improvement was observed in the Six-Minute Walk Test (6MWT). Enhancements have been obtained also in the Ten-Meter Walk Test (10mWT), the Berg Balance Scale (BBS) and the GMFM-88. The Spastic Paraplegia Rating Scale (SPRS) suggested reduction in disease severity. Quality of life's questionnaires showed improved perceived physical health. Gait analysis data did not show significant changes. Intensive treatment with a combination of RAGT and TT improves functional skills and quality of life in HSP. These new findings could guide clinical management and development of future rehabilitation studies.

Total knee arthroplasty (TKA) provides reliable pain relief and implant survival; however, a clinically relevant proportion of patients still report residual symptoms, functional limitations, or an unnatural joint perception. Instrumented gait analysis may objectively assess postoperative function and clarify whether alignment strategies influence gait recovery. This systematic review followed PRISMA 2020 recommendations. PubMed/MEDLINE and the Cochrane Library were searched from January 1990 to November 2024. Eligible studies reported instrumented gait-analysis outcomes after primary fixed-bearing TKA performed with mechanical, kinematic, anatomical, functional, inverse kinematic, restricted kinematic, or other personalized alignment techniques, with at least 12months of follow-up. Revision TKA, unicompartmental knee arthroplasty, case reports, reviews, and studies without an appropriate comparison group were excluded. Six studies met the inclusion criteria, underscoring the limited available evidence. The studies were heterogeneous in design, alignment strategy, gait-analysis protocol, and reported outcomes. Four studies compared kinematic and mechanical alignment, one inverse kinematic and adjusted mechanical alignment, and one gap-balanced TKA with healthy controls. No consistent gait advantage of one alignment strategy was demonstrated. Some studies reported lower knee adduction moment or gait patterns closer to healthy controls after kinematic or inverse kinematic alignment, but findings were inconsistent. Evidence on gait analysis after TKA with different alignment strategies remains limited and heterogeneous. Current data do not allow firm conclusions regarding the superiority of any alignment strategy. Larger prospective studies using standardized gait-analysis protocols are needed.