On Nonlinear Regression for Trends in Split-Belt Treadmill Training (original) (raw)
Related papers
Medical & biological engineering & computing, 2018
This paper applies a nonparametric modelling method with kernel-based regularization to estimate the carbon dioxide production during jogging exercises. The kernel selection and regularization strategies have been discussed; several commonly used kernels are compared regarding the goodness-of-fit, sensitivity, and stability. Based on that, the most appropriate kernel is then selected for the construction of the regularization term. Both the onset and offset of the jogging exercises are investigated. We compare the identified nonparametric models, which include both impulse response models and step response models for the two periods, as well as the relationship between oxygen consumption and carbon dioxide production. The result statistically indicates that the steady-state gain of the carbon dioxide production in the onset of exercise is bigger than that in the offset while the response time of both onset and offset are similar. Compared with oxygen consumption, the response speed ...
Application of nonlinear models to studies in the ergonomics area
Brazilian Journal of …, 2010
This paper presents nonlinear models applied to studies in the ergonomics area. In general, these models evaluate the nature of the relationship between a group ofIndependent Variables (IVs) and one Dependent Variable (DV). In this study the ...
Estimation of oxygen consumption for moderate exercises by using a Hammerstein model
This paper aims to establish block-structured nonlinear model (Hammerstein model) to predict oxygen uptake during moderate treadmill exercises. In order to model the steady state relationship between oxygen uptake (oxygen consumption) and walking speed, six healthy male subjects walked on a motor driven treadmill at six different speed (2,3,4,5,6, and 7 km/h). The averaged oxygen uptake of exercisers at steady state was measured by a mixing chamber based gas analyzer(AEI Moxus Metabolic Cart). Based on these reliable experiment data, a nonlinear static function was obtained by using Support Vector Regression. In order to capture the dynamics of oxygen uptake, a suitable Pseudo Random Binary Signal (PRBS) input was designed and implemented on a computer controlled treadmill. Breath by breath analysis of all exercisers' dynamic responses (PRBS responses) to treadmill walking was performed. A useful ARX model is identified to justify the measured oxygen uptake dynamics within the aerobic range. Finally, a Hammerstein is achieved, which is useful for the control system design of oxygen uptake regulation during treadmill exercises.
Development of a Regression Model for the Treadmill Ground Reaction Force Components
Sports and Exercise Medicine - Open Journal, 2015
Treadmills allow for collecting multiple steps in a small area, and continuous testing for a long period of time with multiple speeds. These factors prove to be useful for biomechanics research laboratories that are usually equipped with floor embedded platforms. Acquiring instrumented treadmills with a built in force plate(s)-may not be financially feasible for many general purpose biomechanics laboratories; additionally instrumented treadmills only measures the vertical component of ground reaction force. The purpose of this study was to examine the components of Ground Reaction Force (GRF) in treadmill walking which were measured by placing a treadmill over floor-mounted force plates and to develop a set of regression equations to be used in associating treadmill's GRF components with GRF components obtained during overground walking. The GRF measured during the treadmill walk was compared to GRF measured in overground walking. A total of twelve male subjects participated in this study. The analysis of the data did not reveal statistical differences in the anterior-posterior component of the GRF (APGRFP1 and APGRFP2) and in the early-stance and mid-stance peaks of the vertical GRF (VGRFP1 and VGRFP2) between treadmill and overground walking. Statistical differences between treadmill and overground walking were found during latestance for vertical ground reaction force (VGRFP3) and medial lateral ground reaction force (MLGRFP2) (p<0.05). During push-off-occurring in late-stance-vertical ground reaction force peaks (VGRFP3) were less in treadmill walking than in overground walking by 5-6% (p<0.05). The Medial-lateral ground reaction forces peaks (MLGRFP2) were also less in treadmill walking than in overground walking by 1-2% (p<0.05). In addition, five regression equations were developed for treadmill's GRF.
The validity of endurance running performance on the Curve 3(TM) non-motorised treadmill
Journal of sports sciences, 2014
This study aimed to test the validity of a non-motorised treadmill (NMT) for the measurement of self-paced overground endurance running performance. Ten male runners performed randomised 5-km running time trials on a NMT and an outdoor athletics track. A range of physiological and perceptual responses was measured, and foot strike was classified subjectively. Performance time was strongly correlated (r = 0.82, ICC = 0.86) between running modes, despite running time being significantly longer on the NMT (1264 ± 124 s vs. 1536 ± 130 s for overground and NMT, respectively; P < 0.001). End blood lactate concentration and rating of perceived exertion were significantly higher on the NMT compared to overground. Integrated electromyography was significantly lower on the NMT for three muscles (P < 0.05), and mean stride rate was also significantly lower on the NMT (P = 0.04). Cardiorespiratory responses of heart rate, oxygen uptake and expired air volume demonstrated strong correlatio...
Frontiers in Neurology, 2022
With increasing age, gait changes often occur, leading to mobility problems and thus a higher risk of falling. Interest in training at home or at retirement homes has led to the development of “mobile treadmills.” A difference in treadmill surface length may influence walking parameters (i.e., step length) and therefore may affect muscle activation. This led to the question: Does the treadmill size affect the muscle activation, i.e., with the length of the walking surface. The study aimed to investigate the influence of treadmill size, i.e., length of the walking surface, on gait pattern and to determine differences in the amplitude of muscle activation using a participant-specific musculoskeletal model (AnyBody Technology A/S, Aalborg, Denmark). For a prospective, randomized study gait parameters were collected from 47 healthy participants (aged 50.19 ± 20.58 years) while walking on two different treadmills, a small mobile treadmill (walking surface length 100 cm) and a conventiona...
A nonlinear dynamic model for heart rate response to treadmill walking exercise
A dynamic model of the heart rate response to treadmill walking exercise is presented. The model is a feedback interconnected system; the subsystem in the forward path represents the neural response to exercise, while the subsystem in the feedback path describes the peripheral local response. The parameters of the model were estimated from 5 healthy adult male subjects, each undertaking 3 sets of walking exercise at different speeds. Simulated responses from the model closely match the experimental data both in the exercise and the recovery phases. The model will be useful in explaining the cardiovascular response to exercise and in the design of exercise protocols for individuals.
ISBS Proceedings Archive, 2020
This study aims to investigate the efficacy of a stacking approach to estimate parameters in treadmill running. Nineteen participants ran on a treadmill at self-selected pace. Ground reaction force and kinematic data were collected. Stacking in machine learning was used to estimate the peak vertical ground reaction force and step time. Good agreement was observed in the test data set for predicted and measured values of the peak vertical ground reaction force component and step time where the ICC values were 0.85 and 0.99 respectively. This suggests stacking may be a feasible approach to estimate kinetic and kinematic parameters during treadmill running.
Journal of sports sciences, 2018
The purpose of this study was to assess metabolic and kinematic parameters (contact and flight time, step length and frequency) while walking at the preferred speed (1.44 ± 0.22 m · s) and while performing an incremental running test (up to exhaustion) on a motorised treadmill (MT) and on a curved non-motorised treadmill (CNMT). Twenty-five volunteers (24.1 ± 3.4 years; 64.7 ± 11.2 kg) participated in the study. Maximal running speed on MT was significantly larger (P < 0.001) than on CNMT (4.31 ± 0.50 vs. 3.75 ± 0.39 m · s) but no differences in heart rate or oxygen uptake [Formula: see text] were observed at this speed. The energy cost of walking (C) and running (C) were significantly greater (P < 0.001) on CNMT than on MT (37 and 17%, respectively). No major differences in kinematic parameters were observed at paired, submaximal, running speeds (2.22-3.89 m · s) but [Formula: see text] was systematically larger in CNMT (of about 340 ml · min · kg). This systematic difference...
The West Indian medical journal, 2013
In the United States of America, 6.2 million individuals are using elliptical motion trainers in fitness centres. However, graded exercise test protocols to estimate peak oxygen consumption (VO(2peak)) using elliptical motion trainers have not been developed for the general population. Fifty-nine subjects (mean age: 23.5 +/- 4.1 years) were randomly divided into a validation (VAL: n = 39) or cross-validation (XVAL: n = 20) group. Peak oxygen consumption (ml x kg(-1) x min(-1)) was measured via indirect calorimetry on an elliptical motion trainer for both groups. Subjects exercised at 150 strides x min(-1) against a resistance of four and a crossramp of 8%. The resistance was increased every two minutes by two units until exhaustion. For the VAL group, a stepwise regression analysis was used to predict VO(2peak) from resistance, maximal heart rate (HR(max)), body mass index (BMI), height and gender (female = 0, male = 1). The prediction equation derived from this study was VO(2peak) ...