7th Dutch Bio-Medical Engineering Conference
January 24th & 25th 2019, Egmond aan Zee, the Netherlands
10:30   Neuromuscular – lower extremities 1
Chair: Wouter Serdijn
15 mins
Effect of Perturbation Amplitude and Number of Repetitions on System Identification of Human Balance Control During Stance
Ingrid Schut, Jantsje Pasma, Alfred Schouten, Herman van der Kooij
Abstract: Human balance is a complex interplay of several underlying mechanisms. To unravel these underlying mechanisms system identification techniques are applied in combination with dedicated perturbations, like support surface translations. However, it remains unclear what the minimal recording time and the optimal perturbation amplitude are. In this study we investigated the effect of these settings on the identification of the neuromuscular controller (NMC). 12 healthy subjects were asked to stand as normal as possible on a treadmill while small continuous belt translations were applied in the form of a periodic multisine signal. The perturbation amplitude varied over seven conditions, where 6 segments of the multisine signal were applied for each amplitude, resulting in a trial lengths of 120 sec. For one of the amplitudes 24 segments were applied. The frequency response functions (FRF) of the NMC as well as the nonlinear distortions were estimated based on the recorded external perturbation torque, body sway and ankle torque. For low amplitudes the FRF could not reliably be estimated due to a low signal-to-noise ratio (SNR). For high amplitudes the relative contribution of the non-linear contributions increased. Increasing the number of repetitions resulted in a lower variability of the FRF. The perturbation amplitude of the support surface translations must be high enough to evoke a sufficient reaction of the human body and also low enough to allow for the use of linear system identification techniques. The number of repetitions of the perturbation signal must be high enough to reduce the amount of noise in the estimation of the NMC. The results of this study showed the optimal range of perturbation amplitudes is around 10 cm (peak-to-peak) and at least 6 repetitions long.
15 mins
Provoking Neuromuscular Responses in the Healthy Knee with Various Gait Perturbations
Jim Schrijvers, Josien van den Noort, Martin van der Esch, Jaap Harlaar
Abstract: Purpose: In patients with disorders of the knee, abnormal loading and instability could be related to impaired neuromuscular control of the knee. Controlled perturbations during gait are used to provoke neuromuscular control. Studies investigating this are often limited to one perturbation type with a fixed intensity. This retains the question how neuromuscular control will respond to various gait perturbations. Therefore, the aim was to compare the effect of four perturbation types, each with five intensities, on the neuromuscular control of the knee during gait in healthy subjects. Methods: A pilot study was performed with eight healthy young subjects (24.4±1.7y) walking (1.2m/s) on an R-mill instrumented treadmill. After unperturbed walking, 20 walking trials were performed with perturbations applied during the stance phase of the right leg. Four perturbation types were tested: acceleration & deceleration of one belt and left & right sway of the treadmill. Five intensities were used per perturbation type. Electromyography was recorded of the medial and lateral muscles surrounding the knee. Repeated measures ANOVA were performed between the perturbation types and intensities using peak, root mean squared (RMS), summation over gait cycle (SUM) and co-contraction index (CCI) values of the electromyography signals. Results: Peak, RMS, SUM and CCI values of all muscles combined (medial + lateral) changed with increasing perturbation intensity (p<0.05) or with the use of different perturbation types (p<0.05). No differences were observed in the peak and CCI values in the lateral muscles with increasing perturbation intensity. Moreover, no differences were present between perturbation types in the peak and CCI values of the lateral and medial muscles. At the individual muscle level, some changes in peak, SUM and RMS values were present between the various gait perturbations. Conclusion: Subtle changes in neuromuscular control of the knee occur in response to various gait perturbations. Increasing the perturbation intensity seems to raise the overall muscle activation and using various perturbation types appeared to provoke different neuromuscular responses. The results of this study could be used as a reference to further explore gait perturbations in patients with knee disorders.
15 mins
Performance-Based Adaptive Assistance for Robotic Gait Training
Simone Fricke, Cristina Bayón, Herman van der Kooij, Edwin van Asseldonk
Abstract: Robotic gait training can improve walking function after neurological conditions (e.g. stroke, SCI). However, the effect of robotic gait training might largely depend on how exactly therapy is provided. It is expected that robotic gait training is more effective if it is tailored to the patient’s needs and if the gait trainer only assists the user when it is really necessary (assist-as-needed, AAN). For the LOPES II gait trainer [1], the gait cycle is divided into different subtasks (weight shift, step length, step width, step height, stability during stance and prepositioning) and the physical therapist may separately adjust the assistance for these subtasks to only assist the user when it is needed. However, training can be effected by therapist’s subjective decisions and assistance often remains constant during a session. To address these issues, we developed an adaptive controller that automatically adjusts the amount of robotic assistance for the different subtasks of walking based on user’s performance [2]. In addition to using this controller as a therapeutic tool, it might also be used as a monitoring tool to evaluate progress during therapy. The new controller calculates the user’s performance for each subtask by comparing the measured joint angles to a reference trajectory at specific points of the gait cycle. The performance is averaged over the previous three steps and the robotic assistance is adjusted based on that performance. If the performance is better than a threshold plus a tolerance, the assistance decreases by 10%. If it is worse than the threshold minus the tolerance, the assistance increases by 10% and in other cases the assistance remains constant. Pilot experiments in stroke survivors and people with spinal cord injury showed the feasibility of the controller to provide appropriate support during walking, adapting the robotic assistance for each subtasks of walking. However, further experiments need to be performed to investigate the feasibility of the controller as a therapeutic and monitoring tool in people with walking impairments. REFERENCES [1] Meuleman J, Van Asseldonk E, Van Oort G, Rietman H, Van Der Kooij H. LOPES II -Design and Evaluation of an Admittance Controlled Gait Training Robot with Shadow-Leg Approach. IEEE Trans Neural Syst Rehabil Eng. 2016;24(3):352–63. [2] Bayón C, Fricke S, Rocon E, van der Kooij H, van Asseldonk E. Performance-Based Adaptive Assistance for Diverse Subtasks of Walking in a Robotic Gait Trainer: Description of a New Controller and Preliminary Results. In: IEEE International Conference on Biomedical Robotics and Biomechatronics (BioRob). 2018.
15 mins
Hydra: Hydrostatic Actuation for Wearable Robotics
Kyrian Staman, Herman van der Kooij
Abstract: Development of high force density actuation technology is a constant endeavor among many fields of research, especially in wearable robotics for human enhancement, rehabilitation, or function restoration. The principle of remote actuation places the heavy components on the fixed world or a location where the added mass and inertia is of least influence to the wearer (usually the back). A hydrostatic actuator concept is proposed that eliminates conventional hydraulic systems' fluid supply and valves, potentially making it lighter, more efficient, and simpler. It also avoids the configuration-dependent friction of Bowden cable transmissions. Test setups are used to identify a nonlinear friction model for commercial hydraulic cylinders and evaluate a control strategy to compensate this friction. Controlled forces up to 300 N (3 N error) and velocities up to 0.5 m/s demonstrate the concept's performance to be sufficient for many wearable applications. With the obtained performance, models, and insights new, small scale hydraulic cylinders are designed to create novel actuation systems for wearable robotics. Part of this work is published in [1]. Contribution to the conference will either focus on design of small scale hydraulic components and their application to wearable robotics or on identification, control, and performance using two test setups. Reference: [1] K. Staman, A. J. Veale and H. van der Kooij, “The PREHydrA: A Passive Return, High Force Density, Electro-Hydrostatic Actuator Concept for Wearable Robotics,” IEEE Robot. Autom. Lett. vol. 3, no. 4, pp. 3569-3574, Oct. 2018. doi: 10.1109/LRA.2018.2854367
15 mins
Analysis of the Factors of Patellofemoral Instability Using Musculoskeletal Modelling and Simulation
Marco Marra, Sebastiaan van de Groes, Dennis Janssen, Nico Verdonschot, Bart Koopman
Abstract: Patellofemoral instability affects very young patients (incidence 21–45%), and causes pain, (sub-)luxation and a general sense of instability in the patellofemoral joint. Many factors play a role in this pathology: trochlear dysplasia, patella alta, rupture of the medial patellofemoral ligament (MPFL), tightness of the lateral retinaculum, and vastus medialis obliquuus (VMO) deficiency. The current surgical treatment aims at correcting the patellar tracking, and includes tibial tubercle transfer, trochleoplasty, and MPFL reconstruction. However, the pre-operative information available is scarce, as it is based on static measurements on X-rays and physical examination. The aim of this research is to use musculoskeletal modelling to understand how the abovementioned factors influence patellofemoral instability. CT and MR images of a cadaver specimen were acquired and segmented using Mimics to reconstruct the 3 D models of femur, patella and tibia, and tibiofemoral and patellofemoral articular cartilage. A specimen-specific 12-DoF knee model was developed using the AnyBody Modeling System, and simulated during a simple knee extension using Force Dependent Kinematics (FDK). Four additional cases were created by manual editing of the 3 D shapes, representing four types (A–D) of trochlear dysplasia. In addition, a combination of MPFL rupture, tightness of lateral retinaculum and VMO deficiency was simulated. Compared to the intact case, trochlear dysplasia (Type B) resulted in 6 mm more lateral patellar translation at knee extension, but no clear dislocation was observed. Patellofemoral contact pressures were altered (Types B–D), showing pressure concentrations in the lowest portion of the patellar cartilage. Only the combination of Type B dysplasia, MPFL rupture, lateral tightness and VMO insufficiency showed patellar sub-luxation during the FDK simulations. Patellofemoral instability is a complex phenomenon and these findings confirm its highly multifactorial nature. Whereas trochlear dysplasia alone did not lead to any (sub-)luxation – although it did change both kinematic and kinetic patterns in the patellofemoral joint – the combination of multiple risk factors clearly led to instability. In further research, bone morphology and soft tissues properties will be studied extensively to identify risky combinations, and a clinical tool for pre-operative planning will be implemented, which will help optimizing the surgery to restore joint stability.
15 mins
Eeg Analysis of Freezing of Gait in Local-Moving Experiment
Ying Wang, Mike X Cohen, Jorik Nonnekes, Richard van Wezel
Abstract: Freezing of gait (FOG) is a motor symptom, which is described as feet that seem to be glued on the floor by patients with Parkinson’s disease (PD). Several cueing‐strategies, e.g. rhythmic auditory or visual cues, may help PD patients overcome FOG. The on-demand manner cueing (i.e. cueing only occurring when FOG is detected, and preferably predicted) would further improve the feasibility in daily life. Current studies mainly analyze movement patterns (via 3D gyroscopes or accelerometers) to detect FOG. However, these studies are limited by their ability to at best detect, but not predict FOG. The possibility of using electroencephalography (EEG) to predict FOG (detect the transition between normal walking and FOG) was proposed by Handojoseno [1]. In his study, 4-channel EEG data was acquired and analyzed, which is not yet sufficiently accurate for FOG prediction and detection. Our study therefore focuses on real‐time detection and prediction of FOG, with not only EEG, but also Electrocardiogram (ECG) and motion sensor data. Ultimately, we aim to develop a ‘brainwave’ chip, which can monitor the EEG signals of PD patients. An automatic FOG detection/prediction algorithm applied on this chip is under investigation. This is an explorative observational study. Fifteen patients with idiopathic Parkinson’s disease (Hoehn and Yahr scale [2, 4]) at off-medication state were asked to execute three tasks in place (two minutes per task) at each session: stepping, normal half turning, and rapid half turning. The following data were acquired during the sessions: 64-channel EEG data (ActiCap), motion data from 6 accelerometers (TMSi; applied above ankles, knees, and metacarpophalangeal joints) and 8 footswitches (TMSi; 4 per foot), EMG data (TMSi; 1 sensor per forearm), and 3-lead ECG data. The experiment was videotaped, and two independent raters annotated the presence of FOG via the videos. Our results show that rapid half turning especially evokes FOG in Parkinson patients. The EEG signals correlated with the FOG episodes were time-frequency analyzed. Based on our preliminary results, we hypothesize that EEG signals present different patterns during FOG in each condition (task), and each subject bring differences in the patterns.