Component Of Ground Reaction Force Research Articles

Comparison of the Accuracy of Ground Reaction Force Component Estimation between Supervised Machine Learning and Deep Learning Methods Using Pressure Insoles.

The three Ground Reaction Force (GRF) components can be estimated using pressure insole sensors. In this paper, we compare the accuracy of estimating GRF components for both feet using six methods: three Deep Learning (DL) methods (Artificial Neural Network, Long Short-Term Memory, and Convolutional Neural Network) and three Supervised Machine Learning (SML) methods (Least Squares, Support Vector Regression, and Random Forest (RF)). Data were collected from nine subjects across six activities: normal and slow walking, static with and without carrying a load, and two Manual Material Handling activities. This study has two main contributions: first, the estimation of GRF components (Fx, Fy, and Fz) during the six activities, two of which have never been studied; second, the comparison of the accuracy of GRF component estimation between the six methods for each activity. RF provided the most accurate estimation for static situations, with mean RMSE values of RMSE_Fx = 1.65 N, RMSE_Fy = 1.35 N, and RMSE_Fz = 7.97 N for the mean absolute values measured by the force plate (reference) RMSE_Fx = 14.10 N, RMSE_Fy = 3.83 N, and RMSE_Fz = 397.45 N. In our study, we found that RF, an SML method, surpassed the experimented DL methods.

Unraveling stroke gait deviations with movement analytics, more than meets the eye: a case control study.

This study aimed to identify and quantify the kinematic and kinetic gait deviations in post-stroke hemiplegic patients with matched healthy controls using Statistical Parametric Mapping (SPM). Fifteen chronic stroke patients [4 females, 11 males; age 53.7 (standard deviation 12.2) years; body mass 65.4 (10.4) kg; standing height 168.5 (9.6) cm] and 15 matched healthy controls [4 females, 11 males; age 52.9 (11.7) years; body weight 66.5 (10.7) years; standing height 168.3 (8.8) cm] were recruited. In a 10-m walking task, joint angles, ground reaction forces (GRF), and joint moments were collected, analyzed, and compared using SPM for an entire gait cycle. Generally, when comparing the stroke patients' affected (hemiplegic) and less-affected (contralateral) limbs with the control group, SPM identified significant differences in the late stance phase and early swing phase in the joint angles and moments in bilateral limbs (all p < 0.005). In addition, the vertical and anteroposterior components of GRF were significantly different in various periods of the stance phase (all p < 0.005), while the mediolateral component showed no differences between the two groups. SPM was able to detect abnormal gait patterns in both the affected and less-affected limbs of stroke patients with significant differences when compared with matched controls. The findings draw attention to significant quantifiable gait deviations in the less-affected post-stroke limb with the potential impact to inform gait retraining strategies for clinicians and physiotherapists.

Investigation of joint contact forces during walking in the subjects with toe in gait due to increasing in femoral head anteversion angle.

Toe-in gait is a pathology in which the child walks and turns the foot inward instead of pointing straight ahead. The alignment of the lower limb structure changes in this disease, increasing the incidence of knee and hip osteoarthritis. This study aimed to determine the kinematic and joint loading in subjects walking with a toe-in gait pattern. This study selected two groups of subjects: normal subjects and those with toe-in gait due to an increased femoral head anteversion angle (each group consisted of 15 subjects). A Qualisys motion analysis system and a Kistler force plate were used to record the motions and forces applied to the leg while walking. OpenSim software (version 3.3) was used to analyze the range of motion, moments, muscle forces, and joint contact forces in both groups of subjects. The mean values of stride length for normal subjects (1.1 ± 0.141 m) and those with toe-in gait (0.94 ± 0.183 m) differed significantly. The mediolateral component of the ground reaction force decreased significantly in the toe-in gait group compared to normal subjects (p-value = 0.05). The peak force of most of the hip joint muscles increased significantly in those with toe-in gait compared to normal subjects (p-value < 0.05). The results of this study showed that those with toe-in gait, due to an increase in femoral head anteversion angle, only had a change in rotation of the pelvic and hip joint. There was no significant difference between walking speed and most ground reaction force components between normal subjects and those with toe-in gait. As the peaks of most of the hip joint muscles increased significantly in the toe-in gait group, this increased joint contact forces (especially the anteroposterior component of the hip joint and the mediolateral component of the knee joint), which may ultimately increase the incidence of hip and knee joint osteoarthritis.

Effect of thoracolumbosacral braces on running ground reaction force components in male individuals with kyphosis

Background & aimsBraces are one of the methods for kyphosis treatment, but they can relocate the center of gravity of the trunk, affecting the ground reaction force (GRF) during running. Therefore, this study aimed to investigate the effects of two types of thoracolumbosacral braces on running GRF components in individuals with kyphosis. Materials & methodsParticipants were 15 males diagnosed with kyphosis who volunteered in this quasi-experimental study. Each subject performed the barefoot running trials on the force plate with one simple brace, with a sensor brace, and without the brace condition. The ground reaction forces components were calculated in the stance phase. Statistical analysis was done with repeated measures test with a significant level of 0.05. ResultsPeak medial ground reaction force when running with a sensor brace was lower than running with a simple brace (p = 0.017). Free moments were similar during three running conditions (p > 0.05). ConclusionLower maximum medial ground reaction force while using a sensor brace may possibly demonstrate the beneficial effects of a sensor brace in individuals with kyphosis.

Ground reaction force analysis in flexible and rigid flatfoot subjects

BackgroundFlatfoot is a structural and functional deformity of the foot that might change ground reaction force variables of gait. Evaluating the components of ground reaction force in three dimensions during gait is considered clinically important. This study aimed to investigate the components of ground reaction force, impulse, and loading rate during gait in people with flexible and rigid flatfoot compared to healthy subjects. 20 young women with flatfoot in two experimental groups (10 with rigid flatfoot and 10 with flexible flatfoot) and 10 healthy women in the control group participated in this study. Ground reaction force components during gait were measured using two force plates. The peak of ground reaction forces, impulse, and loading rate were then extracted. Data were processed and analyzed using MATLAB and SPSS software. One-way ANOVA with a significant level (P˂0.05) was used for statistical analysis. The results showed that peak braking force was higher in the rigid flatfoot group than in the control group (p = 0.016) and the flexible flatfoot group (p = 0.003). The posterior force loading rate was significantly higher in the rigid flatfoot group than in the flexible flatfoot group (P = 0.04). There was no significant difference in vertical loading rate between groups (P˃0.05). Since the maximal posterior ground reaction force was higher in the subjects with rigid flatfoot than in those with flexible flatfoot and healthy subjects, the increase in posterior ground reaction force is associated with an increase in anterior shear force at the knee.

Impact of PCA Pre-Normalization Methods on Ground Reaction Force Estimation Accuracy.

Ground reaction force (GRF) components can be estimated using insole pressure sensors. Principal component analysis in conjunction with machine learning (PCA-ML) methods are widely used for this task. PCA reduces dimensionality and requires pre-normalization. In this paper, we evaluated the impact of twelve pre-normalization methods using three PCA-ML methods on the accuracy of GRF component estimation. Accuracy was assessed using laboratory data from gold-standard force plate measurements. Data were collected from nine subjects during slow- and normal-speed walking activities. We tested the ANN (artificial neural network) and LS (least square) methods while also exploring support vector regression (SVR), a method not previously examined in the literature, to the best of our knowledge. In the context of our work, our results suggest that the same normalization method can produce the worst or the best accuracy results, depending on the ML method. For example, the body weight normalization method yields good results for PCA-ANN but the worst performance for PCA-SVR. For PCA-ANN and PCA-LS, the vector standardization normalization method is recommended. For PCA-SVR, the mean method is recommended. The final message is not to define a normalization method a priori independently of the ML method.

Comparison of the Vertical and Horizontal Components of the Ground Reaction Force of Patients with Anterior Cruciate Ligament Reconstruction and Healthy Individuals during Walking

Introduction: The aim of this research was to compare the vertical, medial-lateral and anterior-posterior components of ground reaction force after anterior cruciate ligament reconstruction with the healthy individuals during walking.&#x0D; Methods: This was an analytic cross sectional research. The participants of this research included 46 football and volleyball athletes, who were divided into two groups of individuals with anterior cruciate ligament reconstruction (ACLR) and healthy individuals. A force plate was employed to evaluate the components of the ground reaction force while walking.The data were statistically analysed using independent t test and SPSS version 16 software.&#x0D; Results: The results related to the vertical component of ground reaction force showed that the participants with anterior cruciate ligament reconstruction applied more force to the ground during heel strike and toe-off compared to healthy individuals, although the difference between the two groups was not significant (p=0.09 and p=0.35), but this difference was greater during heel strike. Likewise, the results related to the horizontal components of the ground reaction force showed that participants with anterior cruciate ligament reconstruction applied more force to the ground in all horizontal components of the ground reaction force, but the biggest difference between the two groups was in the first 20% of the stance phase and in the posterior (p&lt;0.007) and lateral (p&lt;0.02) directions. This difference between the two groups was significant.&#x0D; Conclusion: The results of the research showed that the most important difference between the two groups occurred in the first 20% of the stance phase. Although among the 6 measured components, only the posterior (p &lt; 0.007) and medial (p &lt; 0.02) ground force components were significant.

Long-term effects of running exercises on natural grass, artificial turf, and synthetic surfaces on ground reaction force components in individuals with overpronated feet: A randomized controlled trial

BackgroundOverpronated feet are associated with alteration of lower limb mechanics. Also, quantifying ground reaction force-related changes due to exercise on different surfaces is important for understanding the potential risk of injuries. Research questionWhat are the long-term effects of running exercises on natural grass, artificial turf, and synthetic surfaces on ground reaction force components in individuals with overpronated feet? MethodsOf 82 contacted individuals with overpronated feet aged 18–30, 22 were excluded because they did not meet inclusion criteria (20), and two declined to participate. Sixty individuals were included in randomization, the control (n = 15), hard court or synthetic surface (n = 15), natural grass (n = 15), and artificial turf (n = 15) groups. There was a sample loss of 8 individuals because of personal problems (2 in each group). The intervention groups performed running exercises on natural grass, artificial turf, and synthetic surfaces over eight weeks, three sessions per week. No training or test-related injuries were reported throughout the study. A force plate was embedded midway through the 18-running concrete path to collect ground reaction force data while running on stable ground before and after interventions. ResultsFindings demonstrated significant group-by-time interactions for vertical loading rate (p = 0.016, ETA=0.297). Post-hoc analyses showed increased loading rate amplitude in two natural grass and artificial turf groups (but not in the control and synthetic surface groups) post-intervention. SignificanceRunning training on natural grass and artificial turf surfaces may place individuals with overpronated feet at a higher risk of injury while running on a stable surface.

Automatic gender and unilateral load state recognition for biometric purposes.

Automatic recognition of a person's gender as well as his or her unilateral load state are issues that are often analyzed and utilized by a wide range of applications. For years, scientists have recognized human gait patterns for purposes connected to medical diagnoses, rehabilitation, sport, or biometrics. The present paper makes use of ground reaction forces (GRF) generated during human gait to recognize gender or the unilateral load state of a walking person as well as the combination of both of those characteristics. To solve the above-stated problem parameters calculated on the basis of all GRF components such as mean, variance, standard deviation of data, peak-to-peak amplitude, skewness, kurtosis, and Hurst exponent as well as leading classification algorithms including kNN, artificial neural networks, decision trees, and random forests, were utilized. Data were collected by means of Kistler's force plates during a study carried out at the Bialystok University of Technology on a sample of 214 people with a total of 7,316 recorded gait cycles. The best results were obtained with the use of the kNN classifier which recognized the gender of the participant with an accuracy of 99.37%, the unilateral load state with an accuracy reaching 95.74%, and the combination of those two states with an accuracy of 95.31% which, when compared to results achieved by other authors are some of the most accurate. The study has shown that the given set of parameters in combination with the kNN classifying algorithm allows for an effective automatic recognition of a person's gender as well as the presence of an asymmetrical load in the form of a hand-carried briefcase. The presented method can be used as a first stage in biometrics systems.

Predicting vertical and shear ground reaction forces during walking and jogging using wearable plantar pressure insoles

BackgroundThe development of plantar pressure insoles has made them a potential replacement for force plates. These wearable devices can measure multiple steps and might be used outside of the lab environment for rehabilitation and evaluation of sport performance. However, they can only measure the vertical force which does not completely represent the vertical ground reaction force. In addition, they are not able to measure shear forces which play an import role in the dynamic performance of individuals. Indirect approaches might be implemented to improve the accuracy of the force estimated by plantar pressure systems. Research questionThe aim of this study was to predict the vertical and shear components of ground reaction force from plantar pressure data using recurrent neural networks. MethodsGround reaction force and plantar pressure data were collected from 16 healthy individuals during 10 trials of walking and five trials of jogging using Bertec force plates at 1000 Hz and FScan plantar pressure insoles at 100 Hz. A long short-term memory neural network was built to consider the time dependency of pressure and force data in predictions. The data were split into three subsets of train, to train the model, evaluate, to optimize the model hyperparameters, and test sets, to assess the accuracy of the model predictions. ResultsThe results of this study showed that our long short-term memory model could accurately predict the shear and vertical force components during walking and jogging. The predictions were more accurate during walking compared to jogging. In addition, the predictions of mediolateral force had higher error and lower correlation compared to vertical and anteroposterior components. SignificanceThe long short-term memory model developed in this study may be an acceptable option for accurate estimation of ground reaction force during outdoor activities which can have significant impacts in rehabilitation, sport performance, and gaming.

Lower-Extremity Kinematics of Soccer Players with Chronic Ankle Instability during Running: A Case-Control Study

The purpose of this study was to clarify the characteristics of lower-extremity kinematics during the running of soccer players with chronic ankle instability (CAI) in comparison to those without CAI. Twenty-two male college soccer players participated in this study. Twelve players were assigned to the CAI group and ten players to the non-CAI group, and players were diagnosed according to the Cumberland Ankle Instability Tool. Kinematic data of the hip, knee, ankle, foot, and ground reaction force components during the stance phase of running were obtained using a three-dimensional motion analysis system. The results revealed that soccer players with CAI who landed with ankle inversion and other characteristic kinematics in their lower extremity during the stance phase of running were similar to those without CAI. These results show that running kinematics in soccer players are not affected by the presence or absence of CAI. Future studies based on the results of this study may contribute to the analysis of the risk of developing CAI during soccer and may also help prevent lateral ankle sprains.

Modeling biological individuality using machine learning: A study on human gait

Human gait is a complex and unique biological process that can offer valuable insights into an individual’s health and well-being. In this work, we leverage a machine learning-based approach to model individual gait signatures and identify factors contributing to inter-individual variability in gait patterns. We provide a comprehensive analysis of gait individuality by (1) demonstrating the uniqueness of gait signatures in a large-scale dataset and (2) highlighting the gait characteristics that are most distinctive to each individual. We utilized the data from three publicly available datasets comprising 5368 bilateral ground reaction force recordings during level overground walking from 671 distinct healthy individuals. Our results show that individuals can be identified with a prediction accuracy of 99.3% by using the bilateral signals of all three ground reaction force components, with only 10 out of 1342 recordings in our test data being misclassified. This indicates that the combination of bilateral ground reaction force signals with all three components provides a more comprehensive and accurate representation of an individual’s gait signature. The highest accuracy was achieved by (linear) Support Vector Machines (99.3%), followed by Random Forests (98.7%), Convolutional Neural Networks (95.8%), and Decision Trees (82.8%). The proposed approach provides a powerful tool to better understand biological individuality and has potential applications in personalized healthcare, clinical diagnosis, and therapeutic interventions.

The Effect of Combined Training on Older Men’s Walking and Turning Kinetics

Aims: This study aimed to investigate the effect of combined turning and balance training on ground reaction force (GRF) characteristics during walking and turning in older men.Methods: Thirty older men were selected and randomly divided into two equal experimental (EG: age = 65.47 ± 2.50 years; BMI = 24.57 ± 0.68 kg/m2) and control (CG: age = 65.80 ± 2.34 years; BMI = 24.72 ± 0.64 kg/m2) groups. EG participants completed 12 weeks of training. Kinetic data were recorded using the force plate.Results: After the training, there was a significant difference between the EG and CG in the peak of GRF components and related time parameters as well as the required coefficient of friction in all directions. There was a significant decrease and increase in the stance time and loading rate respectively after the training in EG.Conclusions: The results showed improvement in gait and turning performance of older men. Therefore, combined exercises with emphasis on the turning components for older men are recommended.

Visualising knee loading patterns during kneeling and the development of a laboratory-based adaptation of the Aberdeen Weight-Bearing Test (Knee)

BackgroundInformation regarding the loading of key anatomical structures of the knee during kneeling would enhance existing functional tests, yet current visualisation methods are limited and require further development. Aims(1) Develop a knee loading visualisation technique to investigate loading patterns of the knee during kneeling; and (2) determine the utility of the technique in combination with vertical ground reaction forces and centre of pressure data in the lab-based Aberdeen Weight-Bearing Test (Knee) by assessing their reliability. MethodsFourteen healthy participants conducted kneeling tasks with and without knee pads across two testing sessions. Eight force-sensitive resistors were affixed to the right knee throughout different kneeling tasks: upright kneeling, and reaching forward, back, left, and right. A photo of the force-sensitive resistor configuration was used to generate participant-specific heat maps of knee loading. Two in-ground force platforms were used to measure vertical ground reaction forces and centre of pressure. ResultsThe inferior patella tendon showed the highest proportion of activation during both bare and knee pad kneeling for all kneeling tasks. Knee pads reduced the repeatability of knee loading patterns. Force-sensitive resistor activation and vertical ground reaction force components of the lab-based Aberdeen Weight-Bearing Test (Knee) were shown to be reliable, whereas the centre of pressure data was unreliable. ConclusionWe have developed a lab-based technique for visualising knee loading using force-sensitive resistors. The combination of force-sensitive resistor activation and vertical ground reaction force data provides valuable insights into both the magnitude and locations of applied loads throughout kneeling.

Ground Reaction Force And Electromyograms Of Lower Limb Muscles During Fast Walking

Endurance exercise has been known to be a potent procedure to avoid obesity and its related cardiovascular diseases and to improve glycemic control. Several types of exercise modalities (e.g., walking, running, swimming) are currently utilized, but walking would be most prevalent and accessible. Although general walking is performed around 4-5 km/h, walking at faster speed (fast walking) has been increasing much attention. In previous studies, five months of fast walking increased maximal oxygen uptake and lowered blood pressure (Nemoto et al., 2007). However, the impact of fast walking on mechanical variables as well as muscle activity has not been fully clarified. PURPOSE: To compare ground reaction force and muscle activity between fast walking and running at the equivalent speed. METHODS: Eight males (age; 21 ± 1 yr, height; 172.1 ± 1.7 cm, bodyweight; 62.1 ± 7.0 kg,) participated in the present study. All subjects performed four minutes of slow walking (at 45% of the maximum walking speed, SW), fast walking (at 85% of the maximum walking speed, FW) and running at the equivalent speed from fast walking (Run). During each condition, ground reaction force (GRF) and electromyography (EMG) of seven muscles (gluteus maximus, biceps femoris, rectus femoris, vastus lateralis, gastrocnemius medialis, soleus, and tibialis anterior) were evaluated. We analyzed GRF, average muscle activity (aEMG) and integrated volume (iEMG) during contact phase. RESULTS: In the vertical component of GRF, the load of impact (the maximal and the average values) in FW was lower than those of Run (maximal; 150 ± 25 %BW vs. 219 ± 33 %BW, average; 89 ± 12 %BW vs. 119 ± 19 %BW, p < 0.05). In soleus, aEMG during the ground contact phase in FW (25 ± 9 %MVC) was greater than in SW (13 ± 4 %MVC, p < 0.05). However, no difference was found between FW and Run, but iEMG presented higher value in FW (13 ± 5 %MVC vs. 9 ± 4 %MVC, p < 0.05). In tibialis anterior, aEMG was higher in FW (12 ± 6 %MVC) than in SW (6 ± 4 %MVC) and Run (6 ± 3 %MVC, p < 0.05). iEMG was higher in FW (6 ± 3 %MVC) than in Run (2 ± 1 %MVC, p < 0.05), with no difference between the SW and FW. CONCLUSIONS: The ground reaction force during fast walking was lower than running at the same speed, but the muscle activities of lower limb muscles were comparable.

Knee adduction moment is correlated with the increase in medial meniscus extrusion by dynamic ultrasound in knee osteoarthritis

BackgroundAn increase in medial meniscus extrusion (MME) due to abnormal biomechanical stress leads to knee osteoarthritis (OA) progression. MME evaluation during walking is a key method of detecting dynamic changes in the meniscus, and in combination with motion analysis, can provide a deeper understanding of the mechanisms involved in the increase of MME. ObjectiveTo validate the feasibility of MME dynamic evaluation in combination with a motion analysis system based on the correlation between the increase in MME and biomechanical factors. MethodsTwenty-three knees from 23 patients with mild to moderate knee OA were analysed in this study. The medial meniscus during walking was evaluated by ultrasound. The increase in MME was calculated as the difference between the minimum and maximum MME during walking. A three-dimensional motion analysis system was synchronised with the ultrasound and then, biomechanical factors such as knee moment and ground reaction force were evaluated. ResultsThe wave patterns of the mediolateral and vertical components of ground reaction forces and knee adduction moment were similar to those in the MME based on a high cross-correlation coefficient (>0.8). The increase in MME was significantly correlated with the peak value of the knee adduction moment (r = 0.54, P = 0.0073) but not with the mediolateral and vertical components of the ground reaction force. ConclusionThe findings show that knee adduction moment is correlated with an increase in MME during walking and indicates the validity and feasibility of the dynamic evaluation of MME in combination with a motion analysis system.

Biomechanics of running: A special reference to the comparisons of wearing boots and running shoes.

Boots are often used in sports, occupations, and rehabilitation. However, there are few studies on the biomechanical alterations after wearing boots. The current study aimed to compare the effects of running shoes and boots on running biomechanics. Kinematics and ground reaction forces were recorded from 17 healthy males during running at 3.3 m/s with shoe and boot conditions. Temporal distance gait variables, ground reaction force components as well as lower limb joints angle, moment, and power were compared using Paired t-test and Statistical Parametric Mapping package for time-series analysis. Running with boots was associated with greater stride, step, flight, and swing times, greater flight length, and smaller cadence (p<0.05). The only effect of boots on lower limb joints kinematics during running was a reduction in ankle range of motion (p<0.05). Significantly greater hip flexor, abductor, and internal rotator moments, greater knee extensor and abductor moments, and ankle plantar flexor moments were observed at push-off phase of running as well as greater ankle dorsiflexor moment at early-stance in boot condition (p<0.05). Also, knee joint positive power was greater with a significant temporal shift in boot condition, suggesting a compensatory mechanism in response to limited ankle range of motion and the inability of the ankle joint to generate the required power. Our findings showed that running with boots is physically more demanding and is associated with a greater net contribution of muscles spanning hip and knee joints in order to generate more power and compensate for the ankle joint limitations, consequently, may increase the risk of both musculoskeletal injuries and degenerative joint diseases.

The effect of backpack carriage in different weights on ground reaction force components, loading rate and free moment of Schoolchildren's gait

BackgroundThe objective of this study was to investigate the effect of backpack carriage in different weights on ground reaction force components, loading rate and free moment of schoolchildren's gait. MethodsTwenty-one healthy primary school children (8–11 years old) of second, third, fourth and fifth grade participated in this study where quasi-experimental design was administered. The weight of the backpacks used was 7.5, 10, 12.5 and 15 percent of the participant's body weight. Gait was evaluated using a Kistler force plate at a frequency of 1000 Hz. The GRF along anterior-posterior(y), and medial-lateral (x) and vertical (z) axes, time to peak (TTP), the rate of loading, free moment were calculated. A repeated-measures ANOVA was used to compare the data between subjects. ResultsThe amplitudes of Fy2 and Fx2 were affected by the carrying load with 15 percent body weight. The amplitudes of Fy1, Fz1 and Fz3 were also affected by the carrying load with 12 and 15 percent body weight. ConclusionsThe carrying load with %12 and 15% of the body weight altered schoolchildren's gait performance. These altered variables should be considered as a risk factor for carrying a backpack.

Machine Learning-Based Diabetic Neuropathy and Previous Foot Ulceration Patients Detection Using Electromyography and Ground Reaction Forces during Gait.

Diabetic neuropathy (DN) is one of the prevalent forms of neuropathy that involves alterations in biomechanical changes in the human gait. Diabetic foot ulceration (DFU) is one of the pervasive types of complications that arise due to DN. In the literature, for the last 50 years, researchers have been trying to observe the biomechanical changes due to DN and DFU by studying muscle electromyography (EMG) and ground reaction forces (GRF). However, the literature is contradictory. In such a scenario, we propose using Machine learning techniques to identify DN and DFU patients by using EMG and GRF data. We collected a dataset from the literature which involves three patient groups: Control (n = 6), DN (n = 6), and previous history of DFU (n = 9) and collected three lower limb muscles EMG (tibialis anterior (TA), vastus lateralis (VL), gastrocnemius lateralis (GL)), and three GRF components (GRFx, GRFy, and GRFz). Raw EMG and GRF signals were preprocessed, and different feature extraction techniques were applied to extract the best features from the signals. The extracted feature list was ranked using four different feature ranking techniques, and highly correlated features were removed. In this study, we considered different combinations of muscles and GRF components to find the best performing feature list for the identification of DN and DFU. We trained eight different conventional ML models: Discriminant analysis classifier (DAC), Ensemble classification model (ECM), Kernel classification model (KCM), k-nearest neighbor model (KNN), Linear classification model (LCM), Naive Bayes classifier (NBC), Support vector machine classifier (SVM), and Binary decision classification tree (BDC), to find the best-performing algorithm and optimized that model. We trained the optimized the ML algorithm for different combinations of muscles and GRF component features, and the performance matrix was evaluated. Our study found the KNN algorithm performed well in identifying DN and DFU, and we optimized it before training. We found the best accuracy of 96.18% for EMG analysis using the top 22 features from the chi-square feature ranking technique for features from GL and VL muscles combined. In the GRF analysis, the model showed 98.68% accuracy using the top 7 features from the Feature selection using neighborhood component analysis for the feature combinations from the GRFx-GRFz signal. In conclusion, our study has shown a potential solution for ML application in DN and DFU patient identification using EMG and GRF parameters. With careful signal preprocessing with strategic feature extraction from the biomechanical parameters, optimization of the ML model can provide a potential solution in the diagnosis and stratification of DN and DFU patients from the EMG and GRF signals.

Effects of Shoe Lifespan on Gait Kinetics in Women With Knee Genu Varum and Healthy Peers

Background and Aims Knee genu varum is one of the lower limb abnormalities associated with biomechanical changes in lower limbs. The present study aims to compare the effects of shoe lifespan on ground reaction force (GRF) components in women with knee genu varum and healthy peers during running. Methods The is randomized clinical trial. Participants were 15 female college students with knee genu varum and 15 healthy peers. A new pair of sport shoes was given to each participant during to wear them for six months. Then, they were asked to run at a preferred speed in three trials with the new shoes (pre-test phase) and three trials with the used shoes (post-test phase) in a path where a force plate was installed. Statistical analysis was conducted by using repeated measures ANOVA. All analyses were done in SPSS v. 23 software. The significance level was set at 0.05. Results The results indicated the significant main (group) effect of shoes on peak anterior-posterior GRF during heel contact and on the anterior-posterior impulse (P&lt;0.018, d&gt;0.8). The values of these variables in the genu varum group were significantly higher than in the healthy group. Moreover, findings showed the significant main effects of shoes on the time to reach the peak GRF during heel strike and on the loading rate (P&lt; 0.02, d &gt; 0.9). The time to reach the peak GRF during heel strike with the used shoes was shorter than with the new shoes (P= 0.015, d=0.9), while the vertical loading rate during running with the worn shoes was greater than with the new shoes (P=0.019, d=0.9). Conclusion Running with shoes having a lifespan &gt;6 months is a risk factor for women with knee genu varum.