Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

Wereldwijd is een beroerte de meest frequente oorzaak van blijvende ernstige beperkingen, waarbij niet goed kunnen gebruiken van de hand en de arm één van de meest voorkomende gevolgen is. Omdat voor de meeste dagelijkse activiteiten beide handen nodig zijn, leidt een beroerte vaak tot afhankelijkheid voor deze activiteiten. Levensbalans staat voor het hebben van een evenwichtig patroon van activiteiten en kan ook worden verstoord door een beroerte. Het algemene doel van dit doctoraat het bimanuele handelen en levensbalans te onderzoeken bij personen tijdens het eerste jaar na het optreden van een beroerte, aangezien hierover nog weinig gekend is. Meer specifiek focust dit doctoraat zich op het verder nagaan van de kwaliteiten van een test die het bimanuele handelen in kaart brengt, het nagaan van de kwaliteiten van een vragenlijst die levensbalans meet, het onderzoeken van welke factoren het bimanuele handelen en levensbalans kunnen voorspellen op zes maanden en een jaar na de beroerte en tot slot wordt ook de invloed van het bimanuele handelen op levensbalans nagegaan.Worldwide, stroke is the most frequent cause of long-term disability, with loss of hand and arm use being one of the most common consequences. Because most daily activities require the use both hands working together, a stroke often results in dependency for daily activities. Life balance stands for having a balanced pattern of activities and may be compromised after a stroke. As little is known about bimanual performance and life balance in persons after stroke, the general aim of this PhD is to investigate bimanual performance and life balance within the first year after stroke. More specific, this PhD project focuses on further examining the features of a test that assesses bimanual performance, examining the qualities of a questionnaire that measures life balance, investigating which factors can predict bimanual performance and life balance at six months and a year after the stroke and finally, to investigate the influence of bimanual performance on life balance.

Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

Physics-based simulations of human movement have the potential to predict the functional outcome of treatments, thereby allowing clinicians to personalize and optimize clinical decision-making. These predictive simulations would be particularly valuable for treatments with modest and unpredictable outcomes, such as orthopedic surgeries aiming to improve walking abilities of children with cerebral palsy (CP). Yet computational and modeling challenges still limit the applicability of such simulations in clinical practice. This dissertation aimed to address some of these limitations by improving model-based predictions of walking in children with CP, paving the way for treatment outcome predictions.The central hypothesis of this dissertation was that predictive simulations can be exploited to explore the differential effects of motor control and mechanical deficits on the walking abilities of children with CP. We carried out five studies addressing three objectives to investigate this hypothesis.The first objective was to develop methods and models to characterize the mechanical and motor control deficits of children with CP. Mechanical deficits include altered muscle-tendon properties represented through Hill-type muscle-tendon parameters, whereas motor control deficits include muscle spasticity described as exaggerated muscle reflex activity.In study 1, we proposed a computationally efficient approach based on optimal control and musculoskeletal modeling to estimate subject-specific Hill-type muscle-tendon parameters. The approach identifies parameters that minimize the difference between experimental and model-based joint torques during functional movements. We applied this optimization procedure to estimate optimal fiber lengths and tendon slack lengths of several knee actuators of healthy adults. In addition, we evaluated which functional movements contained sufficient information to provide valid (i.e., representative of the subjects) parameter estimates. We showed that our estimation procedure identified subject-specific parameters that improved the accuracy of joint torque simulations as compared to generic parameters. Further, we found that only specific functional movements contained enough information for a valid estimation, underlining the importance of the experimental data selection.In study 2, we developed and compared three models of spasticity based on delayed sensory feedback from muscle states. The models relied on feedback from muscle length and velocity, feedback from muscle length, velocity, and acceleration, and feedback from muscle force and force rate, respectively. It is thought that the muscle spindle proprioceptive receptors encode information about muscle length and velocity changes. Yet recent experimental findings suggest that spindle firing might be more directly related to muscle force and force rate. Hence, we hypothesized that the force-based model would better explain spastic activity than the length-based models. We developed an optimal control formulation to identify feedback gains that minimized the difference between feedback and measured muscle activity during passive muscle stretches eliciting spastic responses. We found that the force-based model better explained spastic activity than the length-based models, confirming our hypothesis. In addition, we showed that only the force-based model was able to predict muscle activity in agreement with pathological activity measured during gait, opening perspectives for studying the controversial influence of spasticity during gait.The second objective was to develop methods to generate computationally efficient predictive simulations of human movement. We formulated predictive simulations as optimal control problems. Solving such problems is, however, challenging. On the one hand, the equations describing the dynamics of the musculoskeletal system are stiff and nonlinear. On the other hand, using complex musculoskeletal models results in large and computationally expensive problems.In study 3, we developed a framework to generate computationally efficient predictive simulations with complex three-dimensional musculoskeletal models. Key to this efficiency is the combination of direct collocation methods, implicit differential equations, and algorithmic differentiation. We evaluated the computational benefits of using algorithmic differentiation against finite differences to compute derivatives when generating predictive simulations of walking. As expected, algorithmic differentiation drastically reduced the computational time as compared to finite differences. The proposed simulation framework allows overcoming computational roadblocks that have long limited the use of optimal control simulations for biomechanical applications.The third objective was to exploit predictive simulations to i) evaluate candidate walking control strategies in healthy and impaired individuals and ii) explore the differential effects of motor control and mechanical deficits on the walking pattern of a child with CP. In predictive simulations formulated as optimal control problems, the performance criterion determines the control strategy of the motor task. Yet it remains unclear what such criterion would be for human gait. Further, it is unknown whether the same performance criterion can explain the range of gaits adopted by humans in different contexts.In study 4, we identified a multi-objective performance criterion combining energy and effort considerations that produces physiologically realistic predictive simulations of walking based on a three-dimensional musculoskeletal model. We then showed that the same criterion predicts the walk-to-run transition and clinical gait deficiencies caused by muscle weakness and prosthesis use, suggesting that diverse gaits can emerge from the same control strategy. This finding is in line with prior experimental and simulation studies based on two-dimensional models. The ability to predict the mechanics and energetics of diverse gaits suggests that our neuro-musculoskeletal model and simulation framework are generalizable enough for studying healthy and pathological gaits.In study 5, we tested the ability of our simulations to identify the main treatment targets for a child with CP. We generated predictive simulations of walking with a medical imaging based three-dimensional neuro-musculoskeletal model of a child with CP and explored the influence of altered muscle-tendon properties, reduced neuromuscular control complexity, and spasticity on walking performance. We modeled altered muscle-tendon properties through personalized Hill-type muscle-tendon parameters, reduced neuromuscular control complexity through muscle synergies, and spasticity through delayed muscle force and force rate feedback. We found that, in the presence of aberrant musculoskeletal geometries, altered muscle-tendon properties were the primary driver of the crouch gait pattern observed for this child, which is in line with the clinical examination. This suggests that muscle-tendon properties should be the primary target of treatments aiming to restore a more upright gait pattern for this child, which is in accordance with the orthopedic intervention that was performed.The ability of our framework for personalized modeling and predictive simulation to distinguish the contribution of different impairments on walking performance opens the door for modeling the effect of treatments, with the aim of designing optimal and personalized interventions for neuro-musculoskeletal disorders.

Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

Due to dopamine loss in the basal ganglia, patients with Parkinson's disease (PD) have problems performing automatic tasks like walking. The capacity to perform tasks automatically can be evaluated by assessing dual-task (DT) performance, which is defined as "the simultaneous performance of two tasks with distinct goals". Under DT situations, people with PD show a larger deterioration of performance than age-matched healthy control subjects, since conscious attention needs to be given to the tasks performed. Up to now, it remains unclear whether DT performance in people with PD can be improved by training and whether training can be offered in a safe way without increasing fall risk. The purpose of this doctoral thesis was therefore to gain more insight into the scope and effectiveness of DT training in people with PD. The basis of this research project was the DUALITY trial, in which two centers (KU Leuven, Belgium and Radboud University Medical Center, Nijmegen) were involved.In a first study, we examined whether DT motor and DT cognitive outcome measures could be used reliably in longitudinal studies in patients with PD. We included data from 62 patients who underwent two DT tests repeated after a 6 week interval. We found that DT gait outcomes, particularly DT gait velocity, showed an excellent reliability, while proportional DT interference measures were less stable. We also showed that a functional dual task, such as using a mobile phone while walking, was a reliable and ecologically valid task with the potential to be used in the clinic.In a second study, we determined which factors contributed to DT motor and DT cognitive performance in different types of dual tasks. In this analysis all patients participating in the DUALITY trial took part (N = 121). As expected, both motor and cognitive factors played a significant role in the determination of DT performance. We found that single-task gait performance together with a test of verbal fluency and task switching contributed most to DT gait performance. Determining factors remained constant in different types of dual tasks, while additional factors emerged in the Mobile Phone and Stroop task. Single-task cognitive performance was the most important determinant of DT cognitive performance.The third study provided a summary of the recent literature on the effects of DT interventions in PD and an overview on current concepts of DT interference. Different strategies for improving DT performance were proposed, based on studies in older adults. Current evidence was inconclusive on whether DT training was useful in PD. However, two types of training strategies theoretically provided the best option: integrated task training (IDT) would increase the efficiency of brain areas thought to play an executive role in combining two tasks, whereas consecutive task training (CTT) would increase task automaticity, resulting in a higher level of residual brain capacity. Therefore, in the fourth study, we compared the effectiveness and safety of these two different training strategies, following the experimental set-up of the DUALITY trial. In this trial, 121 patients with PD were recruited and randomized into one of the two training groups. The first training group practiced gait and cognitive tasks separately (consecutive), while the other training group practiced gait tasks in combination with cognitive tasks (integrated). Participants received six weeks of training, twice a week in the presence of a physiotherapist and twice a week unsupervised. Before the intervention, DT performance was tested twice before and after a six week control period (test 1 and test 2). Effects of training on DT performance were assessed immediately after the intervention (test 3) and after 12 weeks of follow-up (test 4). The primary outcome was DT gait velocity during the Stroop task, which was an untrained task. Both training strategies proved to be effective in improving DT gait outcomes after training and no differences between strategies could be found. Training did not only improve trained dual tasks, but effects also transferred to untrained tasks and were retained at 12-weeks of follow-up. A second important finding was that fall risk did not increase during and after DT training. Effects on DT cognitive outcomes however were less consistent.In study five, we examined which factors determined DT training effects. Comparable to study 2, motor and cognitive factors, which were assessed at baseline, determined the training effects. Better DT performance before training related to less improvement after training, probably indicating that less room for improvement was available in these patients. Better cognitive performance on the ScopaCog related to larger effects from training, especially in the integrated task training group.In conclusion, this thesis provided insight into general mechanisms of dual tasking in PD and also generated a better understanding into the specific mechanisms underlying DT training interventions in people with PD. We found that DT measures can be used reliably in people with PD and are mostly determined by single-task performance and a test of executive function. Both consecutive and integrated strategies of DT training can be used effectively and safely in a large sample of people with early to mid stage PD. The effects suggested that task automatization, as well as a degree of task integration is possible in this patient group. The findings of this thesis offer new avenues for further research into the concept of dual tasking and DT training in PD and offer a framework for implementing DT interventions in clinical practice.

Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

In life, performance of everyday activities always requires some degree of trunk control. The development of trunk control is a complex process and therefore vulnerable for adverse conditions during early life. Cerebral palsy (CP) is defined as a group of disorders of the development of movement and posture, causing activity limitations that are attributed to non-progressive disturbances occurring in the developing fetal and infant brain. Empirical findings show that children with CP often have impaired trunk control, which affects their ability to maintain a stable position in sitting and standing as well as their performance of functional activities such as reaching and walking. This may in turn impact on the independency and quality of life of these children. However, despite its clinical relevance, current insights into trunk deficits in children with CP is limited, which can be partially explained by a paucity of available measurement tools for both clinical and objective evaluation of trunk control. Adequate assessment is necessary to gain a thorough understanding of their movement deficits, and to provide a sound base for well-targeted treatment planning. At the start of this doctoral project, a comprehensive clinical measurement scale to evaluate both static and dynamic aspects of trunk control and their relation to functional activities was nonexistent. Moreover, whilst three-dimensional (3D) movement analysis has proven a suitable method to obtain a detailed and objective description of upper and lower limb movements in children with CP, its application for trunk movement analysis remained very limited. The few available methods to assess trunk kinematics were dissimilar with regard to the biomechanical model, marker sets, calculation procedures and study population. The scope of this doctoral project was to contribute to the understanding of impaired trunk control in children with CP. The first part of this doctoral project focused on the development of a clinical and an objective measurement method for trunk control in children with CP (Chapters 1 and 2). In the second part, we aimed to gain insights into clinical and objective parameters of impaired trunk control (Chapters 3 and 4), and into the functional relationship between the trunk and the lower limbs during gait (Chapter 5).The first part of this doctoral project addressed the need for appropriate measurement tools for trunk control evaluation in children with CP. A review of the literature revealed a lack of a suitable clinical measurement tool for trunk evaluation in children with CP. In stroke rehabilitation, the Trunk Impairment Scale (TIS) appeared a promising tool to measure both static and dynamic aspects of trunk control in sitting. However, the specific clinical features of impaired trunk control in children with CP necessitated the development of a new scale, based on the TIS, the Trunk Control Measurement Scale (TCMS) (Chapter 1). The TCMS assesses trunk control in sitting and consists of three subscales. The first subscale static sitting balance measures static trunk control during movements of upper and lower limbs. The other two subscales selective movement control and dynamic reaching, evaluate dynamic aspects of trunk control, i.e. active trunk movements within and beyond the base of support. The psychometric properties of the TCMS were evaluated in 26 children with spastic CP and 30 typically developing (TD) children between 8 and 15 years. Results showed excellent inter-rater and test-retest reliability. Also, several aspects of validity (internal consistency, construct validity, discriminant ability) of the TCMS were established. These findings support the use of the TCMS as an evaluative tool.Besides the evaluation of trunk control in sitting, we were also interested to study trunk movements during functional activities, such as gait. Despite the additional value of 3D movement analysis in the assessment of pathological movement patterns, no suitable model to study trunk movements in children with CP was available in literature. Therefore, a biomechanical model was composed for evaluation of head and trunk kinematics during gait in children with CP (Chapter 2). This model consists of five rigid segments, i.e. head, thorax, pelvis, shoulder line and spine. Reliability of the model was verified in 10 children with spastic diplegia between 5 and 15 years. Results showed overall good reliability of all segments within one session and over time, except for the head due to standardization difficulties. Range of motion (ROM) was found the most reliable parameter. The findings of this study provided a sound base for the clinical utility of the model.The second part of this doctoral project was set out to improve our understanding of trunk deficits in children with spastic CP, by use of the previously developed methodology. In Chapter 3, clinical characteristics of impaired trunk control were investigated with the TCMS in a large cohort of children with spastic CP (N=100). Also, differences in these characteristics between subgroups, based on topography (hemiplegia, diplegia, quadriplegia) and severity of the functional impairment (GMFCS levels I-IV), were studied. Results showed clearly impaired trunk control in children with spastic CP, however to a various extent, depending on the topography and severity of the functional impairment. Children with hemiplegia and diplegia mainly showed impaired dynamic trunk control, while children with quadriplegia showed distinct deficits in both static as well as dynamic trunk control. Marked differences were found between GMFCS levels, with larger deficits found in more severely impaired children. These findings provided first specific guidelines for targeted treatment interventions to improve trunk control.Chapter 4 aimed to identify objective parameters of impaired trunk control during gait. Head and trunk kinematic parameters were compared between 20 children with spastic diplegia and 20 age-matched TD children. Children with CP were divided into two subgroups according to their GMFCS level (I-II). Discrete (ROM, mean position) and continuous kinematic parameters (waveforms) were compared between groups. Additionally, a newly developed measure for the overall trunk deficit during gait, the Trunk Profile Score (TPS), was proposed. This index reflects the magnitude of deviation of trunk motion relative to a norm-based dataset of TD children. Results indicated clear head and trunk deficits during gait in children with spastic diplegia, predominantly in more severely impaired children. ROM in the sagittal plane differentiated best between GMFCS levels. The results of this study further established the discriminant ability of the trunk model.The functional relation between trunk and lower limb motion during gait was further explored in Chapter 5. In particular, we wanted to investigate to what extent trunk deficits during gait reflected the presence of underlying impaired trunk control, or needed to be considered as compensatory movements induced by lower limb impairments. We therefore studied the relationship between trunk performance in sitting, measured with the TCMS, and trunk and lower limb kinematic parameters during gait in 20 children with spastic diplegia. Several meaningful correlations were found between TCMS scores and increased trunk deficits during gait, while lower limb and trunk deficits were only related to a limited extent. The findings of this exploratory study thus provided first evidence that the observed trunk deficits during gait should not be solely considered compensatory to lower limb impairments, but also partially reflect an underlying trunk deficit. More in-depth study on a larger cohort will provide a better understanding of underlying trunk deficits in children with CP, which may facilitate well-targeted therapy planning.In conclusion, this doctoral project made a fundamental contribution to the understanding of impaired trunk control in children with spastic CP by providing reliable, clinically feasible and comprehensive tools to assess trunk control in a clinical and biomechanical context. Further elaboration of the current methodology to other motor types and severity levels is recommended. Also, future research is needed to gain insights into the neurological determinants and early development of trunk control. These insights may provide valuable and specific clues for therapeutic interventions aimed at improvement of trunk control in children with CP, which may ultimately improve their functionality.

Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

Freezing of Gait (FOG) is a debilitating phenomenon, common in advanced Parkinson's Disease (PD). Developing FOG is a major risk for falls and is associated with higher anxiety and depression and reduced quality of life. The pathophysiology of FOG is poorly understood and prospective studies designed to explore the mechanisms of FOG development are scarce. Furthermore, given the partial response to medication, the role of preventative or remedial therapy is promising. This project contains two parts - the first a prospective study to investigate the markers of FOG development using a multi-level and multi-domain approach while the second part is a training study which will target a potential marker of FOG development i.e. gait asymmetry to examine whether split-belt training can remediate FOG behaviour as a proof of concept for clinical trials. We expect that this innovative study will provide much needed insights into the development of FOG as well as guide future research aiming to ameliorate the burden of patients with PD.