Correlation between the Gait Deviation Index and skeletal muscle mass in children with spastic cerebral palsy

Machine learning approach to gait deviation prediction based on isokinetic data acquired from biometric sensors

BACKGROUND

Analyzing gait deviation is one of the crucial factors during the diagnosis and treatment of children with Cerebral Palsy (CP). The typical diagnostic procedure requires an expensive and complicated three-dimensional gait analysis system based on visual sensors. In this work, we focus on predicting well-known gait pathology scores using only information collected from the BS4P, the affordable isokinetic dynamometer. Using such equipment, it is possible to determine gait pathological indices such as the gait deviation index (GDI) or the Gillette gait index (GGI).

RESEARCH QUESTION

Are there correlations between the results of examining patients with CP on the Biodex Pro 4 device and the gait quality metrics (GDI and GGI)?

METHODS

The isokinetic data acquired from biometric sensors (74 records) were analyzed using big data methods. We used several Machine Learning methods to find the correlation between gait deviation and isokinetic data: Adaptive Boosting Regression, K-nearest Neighbor, Decision Tree Regression, Random Forest Regression, and Gradient Boost Regression.

RESULTS

In this paper, we provided a detailed comparison of different machine learning regression models in predicting gait quality in patients with CP based only on the data gathered from affordable Biodex 4 Pro device. The best result was obtained using the gradient boosting regression model with Mean Absolute Percentage Error of 6%. However, it was not possible to precisely predict the GGI index using this method.

SIGNIFICANCE

The results obtained showed promising results in the evaluation of gait index scores, which gives the possibility of diagnosing patients with CP without the use of expensive optometric systems. Evaluating gait metrics using the approach proposed in this paper could be very helpful for both physicians and physiotherapists in assessing the condition of patients with CP, as well as other diseases related to gait problems.

Estimation of Gross Motor Functions in Children with Cerebral Palsy Using Zebris FDM-T Treadmill

A standardized observational instrument designed to measure change in gross motor function over time in children with cerebral palsy is the Gross Motor Function Measure (GMFM). The process of evaluating a value for the GMFM index can be time consuming. It typically takes 45 to 60 min for the patient to complete all tasks, sometimes in two or more sessions. The diagnostic procedure requires trained and specialized therapists. The paper presents the estimation of the GMFM measure for patients with cerebral palsy based on the results of the Zebris FDM-T treadmill. For this purpose, the regression analysis was used. Estimations based on the Generalized Linear Regression were assessed using different error metrics. The results obtained showed that the GMFM score can be estimated with acceptable accuracy. Because the Zebris FDM-T is a widely used device in gait rehabilitation, our method has the potential to be widely adopted for objective diagnostics of children with cerebral palsy.

Relationship between 3D lower limb bone morphology and 3D gait variables in children with uni and bilateral Cerebral Palsy

BACKGROUND

Medical and surgical interventions to prevent or reduce bone deformities and improve gait in children with cerebral palsy (CP) are based on empirical evidence that there is a relationship between bone deformities and gait deviations.

RESEARCH QUESTION

What is the relationship between tibial-femoral bone morphology and kinematic gait variables in ambulant children with CP?

METHODS

A retrospective analysis was conducted on data from 121 children with uni- (n = 64, mean age 9.9 (SD 3.4) years) and bi- lateral (n = 57, mean age 10.4 (SD 3.6) years) CP who had undergone 3D gait analysis and biplanar X-rays (EOS® system). The limbs were split as DIP (the more impaired limb of children with bilateral CP), HEMI (the impaired limb of unilateral CP) and REF (the unimpaired limb of unilateral CP). Multi-variable Linear Regressions were performed between 23 kinematic variables, the Gait Deviation Index (GDI) and a model composed of nine 3D bone variables for each limb type.

RESULTS

When the whole sample was pooled, 72% of R² values were poor, 16% were fair, and 12% were moderate. Lower limb bone morphology models explained less than 1% of GDI variability. Correlations between tibial-femoral rotational parameters and hip rotation were mostly poor. Mean foot progression angle was the only kinematic parameter that was fairly to moderately correlated with bone variables in the 3 limb types. A tibial-femoral bone model explained 48% of the variability of mean foot progression angle in the REF limbs, 31% in the HEMI limbs and 25% in the DIP limbs.

SIGNIFICANCE

Tibial-femoral bone morphology was only weakly related to kinematic gait variables, in contrast with common clinical assumptions. These results suggest that factors other than bone morphology influence gait quality and thus a thorough clinical examination and gait analysis is required prior to making treatment decisions.

Three-dimensional gait quantitative analysis in postoperative rehabilitation of lumbar degenerative diseases: a self-controlled before-after study

OBJECTIVE

To evaluate the efficacy of three-dimensional (3-D) gait quantitative analysis in the surgical treatment and postoperative rehabilitation in patients with lumbar degenerative diseases.

METHODS

This is a prospective study with self-controll before -after. A total of 48 patients with lumbar degenerative diseases and treated in our hospital were enrolled in the observation group, 40 healthy individuals were included in the control group. Gait analysis was carried out with 3-D motion acquisition and analysis system. The 3-D gait quantitative parameters of the two groups were compared preoperatively. These include time-distance parameters (gait speed, stride frequency, stride length, support phase), hip joint flexion angle and gait deviation index (GDI). The 3-D gait quantitative parameters in the observation group were analyzed post operation and during rehabilitation. Pearson correlation coefficient was used to evaluate the correlation between 3-D gait quantitative parameters and the patient's visual analog score (VAS), Japanese Orthopedic Association (JOA) score and Oswestry Disability Index (ODI).

RESULTS

Compared with the healthy group, the time-distance parameters and the kinematics parameters of lower extremity joints in the observation group were significantly decreased (both P<0.001). The gait index indicated that there were significant gait abnormalities in the observation group (P<0.001). Two weeks after operation, the patient's VAS score, JOA score and ODI were significantly improved compared to the results preoperatively, as well as the 3-D gait quantitative parameters (all P<0.05). Further improvement was then observed after 12 weeks of rehabilitation training (all P<0.05), and the patient's gait was close to normal. Pearson correlation analysis showed that the improvement of the 3-D gait quantitative parameters positively correlated with VAS score, JOA score and ODI (all P<0.001).

CONCLUSION

The 3-D gait quantitative analysis can effectively evaluate the effect of operation and rehabilitation training.

Effect of an interval rehabilitation program with home-based, vibration-assisted training on the development of muscle and bone in children with cerebral palsy - an observational study

Objectives In children with cerebral palsy (CP), the most common cause of physical impairment in childhood, less muscle and bone growth has been reported, when compared with typically developing children. The aim of this study was to evaluate the effect of an intensive rehabilitation program including physiotherapy in combination with 6 months of home-based, vibration-assisted training on muscle and bone growth in children with CP. Methods We included children with CP, who participated in a rehabilitation program utilizing whole-body vibration (WBV). Muscle mass was quantified by appendicular lean mass index (App-LMI) and bone mass by total-body-less-head bone mineral content (TBLH-BMC) assessed by Dual-energy X-ray absorptiometry (DXA) at the beginning of rehabilitation and one year later. To assess the functional muscle-bone unit, the relation of TBLH-BMC to TBLH lean body mass (TBLH-LBM) was used. Results The study population included 128 children (52 females, mean age 11.9 ± 2.7). App-LMI assessed in kg/m2 increased significantly after rehabilitation. The age-adjusted Z-score for App-LMI showed no significant change. TBLH-BMC assessed in gram increased significantly. The Z-scores for TBLH-BMC decreased lesser than expected by the evaluation of the cross-sectional data at the beginning of rehabilitation. The parameter T B L H - B M C T B L H - L B M $\frac{TBLH-BMC}{TBLH-LBM}$ did not change relevantly after 12 months. Conclusions Muscle growth and to a lesser extent bone growth could be increased in children with CP. The intensive rehabilitation program including WBV seemed to have no direct effect on the bone, but the observed anabolic effect on the bone, may only been mediated through the muscle.

The Appendicular Lean Mass Index Is a Suitable Surrogate for Muscle Mass in Children with Cerebral Palsy

BACKGROUND

Densitometrically measured lean body mass (LBM) is often used to quantify skeletal muscle mass in children with cerebral palsy (CP). Since LBM depends on the individual's height, the evaluation of $\frac{{{\rm{LBM}}}}{{heigh{t^2}}}\ $ (lean BMI) is often recommended. However, LBM includes not only skeletal muscle mass but also the mass of skin, internal organs, tendons, and other components. This limitation applies to a far lesser extent to the appendicular lean mass index (LMIapp).

OBJECTIVES

The aim of the study was to evaluate skeletal muscle mass in children with CP using total lean BMI (LMItot) and LMIapp.

METHODS

The present study was a monocentric retrospective analysis of prospectively collected data among children and adolescents with CP participating in a rehabilitation program. In total, 329 children with CP [148 females; Gross Motor Function Classification Scale (GMFCS) I, 32 children; GMFCS II, 73 children; GMFCS III, 133 children; GMFCS IV, 78 children; and GMFCS V, 13 children] were eligible for analysis. The mean age was 12.3 ± 2.75 y. Pediatric reference centiles for age-adjusted LMIapp were generated using data from NHANES 1999-2004. Low skeletal muscle mass was defined as a z score for DXA determined LMItot and LMIapp less than or equal to -2.0.

RESULTS

The z scores for LMIapp were significantly lower than LMItot in children with CP, GMFCS levels II-V (P < 0.001), with the exception of GMFCS level I (P = 0.121), where no significant difference was found. The prevalence of low LMItot (16.1%; 95% CI: 16.1, 20.1%) was significantly lower (P < 0.001) than the prevalence of LMIapp (42.2%; 95% CI: 36.9, 47.9%) in the study population.

CONCLUSIONS

The prevalence of low skeletal muscle mass in children with CP might be underestimated by LMItot. LMIapp is more suitable for the evaluation of skeletal muscle mass in children with CP.

Differences of the ankle plantar flexor length in typically developing children and children with spastic hemiplegic cerebral palsy

The purpose of this study is to analyze the lengths of the medial and lateral gastrocnemius and soleus muscles in children with spastic hemiplegic cerebral palsy to quantitatively assess the structural differences in skeletal muscles. This study included 10 children with spastic cerebral palsy and 10 children with typically development. To assess the changes in the length of the ankle plantar flexor due to cerebral palsy, we utilized both gait analysis and software for interactive musculoskeletal modeling to model skeletal muscle length. With this model, the differences in the lengths of the medial and lateral gastrocnemius and soleus muscles were assessed at different knee (0°, 45°, and 90°) and ankle (-10°, 0°, 15°, and 30°) angles. Muscle length on the paretic group was shorter than the typically developing and nonparetic group for all three muscles (medial and lateral gastrocnemius and soleus muscles) for knee and ankle angles. These results were not statistically significant. Normalized muscle lengths in the dynamic/static status revealed a significant difference in the length of the lateral gastrocnemius muscle between the cerebral palsy and typically developing group. I observed muscle shortening on the paretic side of the children with cerebral palsy. This finding suggests that the recovery of plantar flexor length is the most important issue that must be resolved for normal gait and motor function.