Morten Enemark Lund

On validation of multibody musculoskeletal models

We review the opportunities to validate multibody musculoskeletal models in view of the current transition of musculoskeletal modelling from a research topic to a practical simulation tool in product design, healthcare and other important applications. This transition creates a new need for justification that the models are adequate representations of the systems they simulate. The need for a consistent terminology and established standards is identified and knowledge from fields with a more progressed state-of-the-art in verification and validation is introduced. A number of practical steps for improvement of the validation of multibody musculoskeletal models are pointed out and directions for future research in the field are proposed. It is hoped that a more structured approach to model validation can help to improve the credibility of musculoskeletal models.

Inductive tongue control of powered wheelchairs

Alternative and effective methods for controlling powered wheelchairs are important to individuals with tetraplegia and similar impairments whom are unable to use the standard joystick. This paper describes a system where tongue movements are used to control a powered wheelchair thus providing users, with high level spinal cord injuries, full control of their wheelchair. The system is based on an inductive tongue control system developed at Center for Sensory-Motor Interaction (SMI), Aalborg University. The system emulates a standard analog joystick in order to interface the wheelchair, thus ensuring that the system works with almost any wheelchair. The total embedment of the tongue interface into the mouth makes the control practically invisible. A fuzzy system combining 8 sensors for directional control allows for multidirectional control of the wheelchair. Preliminary test results show navigation abilities, which are highly competitive when compared to other tongue control system.

Fully integrated wireless inductive tongue computer interface for disabled people

This work describes a novel fully integrated inductive tongue computer interface for disabled people. The interface consists of an oral unit placed in the mouth, including inductive sensors, related electronics, a system for wireless transmission and a rechargeable battery. The system is activated using an activation unit placed on the tongue, and incorporates 18 inductive sensors, arranged in both a key area and a mouse-pad area. The systems functionality was demonstrated in a pilot experiment, where a typing rate of up to 70 characters/minute was obtained with an error rate of 3%. Future work will include tests with disabled subjects.

Scaling of musculoskeletal models from static and dynamic trials

Subject-specific scaling of cadaver-based musculoskeletal models is important for accurate musculoskeletal analysis within multiple areas such as ergonomics, orthopaedics and occupational health. We present two procedures to scale ‘generic’ musculoskeletal models to match segment lengths and joint parameters to a specific subject and compare the results to a simpler approach based on linear, segment-wise scaling. By incorporating data from functional and standing reference trials, the new scaling approaches reduce the model sensitivity to assumed model marker positions. For validation, we applied all three scaling methods to an inverse dynamics-based musculoskeletal model and compared predicted knee joint contact forces to those measured with an instrumented prosthesis during gait. Additionally, a Monte Carlo study was used to investigate the sensitivity of the knee joint contact force to random adjustments of the assumed model marker positions (+/− one marker diameter). The model based on linear scaling showed the highest variation in the knee joint contact force of 1.44 body weight (BW) around contra-lateral heel strike, and a variation in root mean square deviation (RMSD) of 0.36 BW. The proposed methods reduced the variation to 1.0 BW (RMSD 0.26 BW) for the anatomical landmark based method and 0.47 BW (RMSD 0.06 BW) for the functional based method. Variation in model predictions due to uncertainty in marker positions is a trait of all marker-based musculoskeletal modelling approaches. The presented methods solve part of this problem and rely less on manual identification of anatomical landmarks in the model. The work represents a step towards a more consistent methodology in musculoskeletal modelling.

Clinical evaluation of wireless inductive tongue computer interface for control of computers and assistive devices

Typing performance of a full alphabet keyboard and a joystick type of mouse (with on-screen keyboard) provided by a wireless integrated tongue control system (TCS) has been investigated. The speed and accuracy have been measured in a form of a throughput defining the true correct words per minute [cwpm]. Training character sequences were typed in a dedicated interface that provided visual feedback of activated sensors, a map of the alphabet associated, and the task character. Testing sentences were typed in Word, with limited visual feedback, using non-predictive typing (map of characters in alphabetic order associated to sensors) and predictive typing (LetterWise) for TCS keyboard, and non-predictive typing for TCS mouse. Two subjects participated for four and three consecutive days, respectively, two sessions per day. Maximal throughput of 2.94, 2.46, and 2.06, 1.68 [cwpm] were obtained with TCS keyboard by subject 1 and 2 with predictive and non-predictive typing respectively. Maximal throughput of 2.09 and 1.71 [cwpm] was obtained with TCS mouse by subject 1 and 2, respectively. Same experimental protocol has been planned for a larger number of subjects.

FAST CHANGES IN DIRECTION DURING HUMAN LOCOMOTION ARE EXECUTED BY IMPULSIVE ACTIVATION OF MOTOR MODULES

This study investigated the modular control of complex locomotor tasks that require fast changes in direction, i.e., cutting manoeuvres. It was hypothesized that such tasks are accomplished by an impulsive (burst-like) activation of a few motor modules, as observed during walking and running. It was further hypothesized that the performance in cutting manoeuvres would be associated to the relative timing of the activation impulses. Twenty-two healthy men performed 90° side-step cutting manoeuvres while electromyography (EMG) activity from 16 muscles of the supporting limb and trunk, kinematics, and ground reaction forces were recorded. Motor modules and their respective temporal activations were extracted from the EMG signals by non-negative matrix factorization. The kinematic analysis provided the velocity of the centre of mass and the external work absorbed during the load acceptance (negative work, external work during absorption (W-Abs)) and propulsion phases (positive work, external work during propulsion (W-Prp)) of the cutting manoeuvres. Five motor modules explained the EMG activity of all muscles and were driven in an impulsive way, with timing related to the initial contact (M2), load acceptance (M3), and propulsion (M4). The variability in timing between impulses across subjects was greater for cutting manoeuvres than for running. The timing difference between M2 and M3 in the cutting manoeuvres was significantly associated to W-Abs (r(2)=0.45) whereas the timing between M3 and M4 was associated to W-Prp (r(2)=0.43). These results suggest that complex locomotor tasks can be achieved by impulsive activation of muscle groups, and that performance is associated to the specific timing of the activation impulses.

Inductive pointing device for tongue control system for computers and assistive devices

Experimental results for pointing tasks using a tongue control system are reported in this paper. Ten untrained subjects participated in the experiment. Both typing and pointing tasks were performed, in three short-term training sessions, in consecutive days, by each subject. The system provided a key pad (14 sensors) and a mouse pad (10 sensors with joystick functionality) whose placements were interchanged (front, back) in half of the subjects. The pointing tasks consisted of selecting and tracking a target circle (of 50, 75 and 100 pixels diameter) that occurred randomly in each of the 16 positions uniformly distributed along the perimeter of a layout circle of 250 pixels diameter. The throughput was of 0.808 bits per second and the time on target was of 0.164 of the total tracking time. The pads layout, the subjects, the sessions, the target diameters, and the angle of the tracking direction had a statistically significant effect on the two performance measures. Long term training is required to assess the improvement of the user capability.

Effects of Perturbations to Balance on Neuromechanics of Fast Changes in Direction during Locomotion

This study investigated whether the modular control of changes in direction while running is influenced by perturbations to balance. Twenty-two healthy men performed 90° side-step unperturbed cutting manoeuvres while running (UPT) as well as manoeuvres perturbed at initial contact (PTB, 10 cm translation of a moveable force platform). Surface EMG activity from 16 muscles of the supporting limb and trunk, kinematics, and ground reaction forces were recorded. Motor modules composed by muscle weightings and their respective activation signals were extracted from the EMG signals by non-negative matrix factorization. Knee joint moments, co-contraction ratios and co-contraction indexes (hamstrings/quadriceps) and motor modules were compared between UPT and PTB. Five motor modules were enough to reconstruct UPT and PTB EMG activity (variance accounted for UPT  = 92±5%, PTB = 90±6%). Moreover, higher similarities between muscle weightings from UPT and PTB (similarity = 0.83±0.08) were observed in comparison to the similarities between the activation signals that drive the temporal properties of the motor modules (similarity = 0.71±0.18). In addition, the reconstruction of PTB EMG from fixed muscle weightings from UPT resulted in higher reconstruction quality (82±6%) when compared to reconstruction of PTB EMG from fixed activation signals from UPT (59±11%). Perturbations at initial contact reduced knee abduction moments (7%), as well as co-contraction ratio (11%) and co-contraction index (12%) shortly after the perturbation onset. These changes in co-contraction ratio and co-contraction index were caused by a reduced activation of hamstrings that was also verified in the activation signals of the specific motor module related to initial contact. Our results suggested that perturbations to balance influence modular control of cutting manoeuvres, especially the temporal properties of muscle recruitment, due to altered afferent inputs to the motor patterns. Furthermore, reduced knee stability during perturbed events may be related to overall control of lower limb muscles.

A framework for mouse and keyboard emulation in a tongue control system

Effective human input devices for computer control are very important to quadriplegics and others with severe disabilities. This paper describes a framework for computer control without need for special PC software or drivers. The framework is based on a tongue control system recently developed at Center for Sensory-Motor Interaction (SMI), Aalborg University. The framework provides emulation of a standard USB keyboard and mouse, and allows tongue control of any computer using standard USB drivers available in all modern operating systems.

Development and functional demonstration of a wireless intraoral inductive tongue computer interface for severely disabled persons

Abstract Purpose: Individuals with tetraplegia depend on alternative interfaces in order to control computers and other electronic equipment. Current interfaces are often limited in the number of available control commands, and may compromise the social identity of an individual due to their undesirable appearance. The purpose of this study was to implement an alternative computer interface, which was fully embedded into the oral cavity and which provided multiple control commands. Methods: The development of a wireless, intraoral, inductive tongue computer was described. The interface encompassed a 10-key keypad area and a mouse pad area. This system was embedded wirelessly into the oral cavity of the user. The functionality of the system was demonstrated in two tetraplegic individuals and two able-bodied individuals Results: The system was invisible during use and allowed the user to type on a computer using either the keypad area or the mouse pad. The maximal typing rate was 1.8 s for repetitively typing a correct character with the keypad area and 1.4 s for repetitively typing a correct character with the mouse pad area. Conclusion: The results suggest that this inductive tongue computer interface provides an esthetically acceptable and functionally efficient environmental control for a severely disabled user. Implications for Rehabilitation New Design, Implementation and detection methods for intra oral assistive devices. Demonstration of wireless, powering and encapsulation techniques suitable for intra oral embedment of assistive devices. Demonstration of the functionality of a rechargeable and fully embedded intra oral tongue controlled computer input device.