Ximena Omlin

Real-Time Closed-Loop Control of Cognitive Load in Neurological Patients During Robot-Assisted Gait Training

Cognitively challenging training sessions during robot-assisted gait training after stroke were shown to be key requirements for the success of rehabilitation. Despite a broad variability of cognitive impairments amongst the stroke population, current rehabilitation environments do not adapt to the cognitive capabilities of the patient, as cognitive load cannot be objectively assessed in real-time. We provided healthy subjects and stroke patients with a virtual task during robot-assisted gait training, which allowed modulating cognitive load by adapting the difficulty level of the task. We quantified the cognitive load of stroke patients by using psychophysiological measurements and performance data. In open-loop experiments with healthy subjects and stroke patients, we obtained training data for a linear, adaptive classifier that estimated the current cognitive load of patients in real-time. We verified our classification results via questionnaires and obtained 88% correct classification in healthy subjects and 75% in patients. Using the pre-trained, adaptive classifier, we closed the cognitive control loop around healthy subjects and stroke patients by automatically adapting the difficulty level of the virtual task in real-time such that patients were neither cognitively overloaded nor under-challenged.

Controlling patient participation during robot-assisted gait training.

BackgroundThe overall goal of this paper was to investigate approaches to controlling active participation in stroke patients during robot-assisted gait therapy. Although active physical participation during gait rehabilitation after stroke was shown to improve therapy outcome, some patients can behave passively during rehabilitation, not maximally benefiting from the gait training. Up to now, there has not been an effective method for forcing patient activity to the desired level that would most benefit stroke patients with a broad variety of cognitive and biomechanical impairments.MethodsPatient activity was quantified in two ways: by heart rate (HR), a physiological parameter that reflected physical effort during body weight supported treadmill training, and by a weighted sum of the interaction torques (WIT) between robot and patient, recorded from hip and knee joints of both legs. We recorded data in three experiments, each with five stroke patients, and controlled HR and WIT to a desired temporal profile. Depending on the patients cognitive capabilities, two different approaches were taken: either by allowing voluntary patient effort via visual instructions or by forcing the patient to vary physical effort by adapting the treadmill speed.ResultsWe successfully controlled patient activity quantified by WIT and by HR to a desired level. The setup was thereby individually adaptable to the specific cognitive and biomechanical needs of each patient.ConclusionBased on the three successful approaches to controlling patient participation, we propose a metric which enables clinicians to select the best strategy for each patient, according to the patients physical and cognitive capabilities. Our framework will enable therapists to challenge the patient to more activity by automatically controlling the patient effort to a desired level. We expect that the increase in activity will lead to improved rehabilitation outcome.

Predicting Targets of Human Reaching Motions Using Different Sensing Technologies

Rapid recognition of voluntary motions is crucial in human-computer interaction, but few studies compare the predictive abilities of different sensing technologies. This paper thus compares performances of different technologies when predicting targets of human reaching motions: electroencephalography (EEG), electrooculography, camera-based eye tracking, electromyography (EMG), hand position, and the users preferences. Supervised machine learning is used to make predictions at different points in time (before and during limb motion) with each individual sensing modality. Different modalities are then combined using an algorithm that takes into account the different times at which modalities provide useful information. Results show that EEG can make predictions before limb motion onset, but requires subject-specific training and exhibits decreased performance as the number of possible targets increases. EMG and hand position give high accuracy, but only once the motion has begun. Eye tracking is robust and exhibits high accuracy at the very onset of limb motion. Several advantages of combining different modalities are also shown, including advantages of combining measurements with contextual data. Finally, some recommendations are given for sensing modalities with regard to different criteria and applications. The information could aid human-computer interaction designers in selecting and evaluating appropriate equipment for their applications.

Model-based Heart rate prediction during Lokomat walking

We implemented a model for prediction of heart rate during Lokomat walking. Using this model, we can predict potential overstressing of the patient and adapt the physical load accordingly. Current models for treadmill based heart rate control neglect the fact that the interaction torques between Lokomat and human can have a significant effect on heart rate. Tests with five healthy subjects lead to a model of sixth order with walking speed and power expenditure as inputs and heart rate prediction as output. Recordings with five different subjects were used for model validation. Future work includes model identification and predictive heart rate control with spinal cord injured and stroke patients.

Voluntary gait speed adaptation for robot-assisted treadmill training

Robot-assisted gait training currently lacks the possibility of the robot to automatically adapt to the patients needs and demands (so called “bio-cooperative control strategies”). It is desired to give the patient voluntary control over training parameters such as gait speed or joint trajectories. We implemented a control algorithm for the driven gait orthosis Lokomat that allows severely disabled stroke patients a limited and safe allowance of influence on their gait speed. To exercise gait symmetry, our algorithm can be configured such that only activity in the paretic leg will cause changes in treadmill speed. The algorithm was successfully tested with eight healthy subjects and six stroke patients.

Model-based heart rate control during robot-assisted gait training

In recent years, gait robots have become increasingly common for gait rehabilitation in non-ambulatory stroke patients. Cardiovascular treadmill training, which has been shown to provide great benefit to stroke survivors, cannot be performed with non-ambulatory patients. We therefore integrated cardiovascular training in robot-assisted gait therapy to combine the benefits of both training modi. We developed a model of human heart rate as a function of exercise parameters during robot-assisted gait training and applied it for automatic control purposes. This structural model of the physiological processes describes the change in heart rate caused by treadmill speed and the power exchanged between robot and subject. We performed physiological parameter estimation for each tested individual and designed a model-based feedback controller to guide heart rate to a desired time profile. Five healthy subjects and eight stroke patients were recorded for model parameter identification, which was successfully used for heart rate control of three healthy subjects. We showed that a model-based control approach can take into account patient-specific limitations of treadmill speed as well as individual power expenditure.

Workload Estimation in Physical Human–Robot Interaction Using Physiological Measurements

This paper uses physiological measurements to estimate human workload and effort in physical human–robot interaction. Ten subjects performed 19 consecutive task periods using the ARMin robot while difficulty was varied along two scales. Three physiological modalities were measured: electroencephalography, autonomic nervous system (ANS) responses (electrocardiography, skin conductance, respiration, skin temperature) and eye tracking. After each task period, reference workload and effort values were collected using the NASA Task Load Index. Machine learning was used to estimate workload and effort from physiological data. All three physiological modalities performed significantly better than random, particularly using nonlinear estimation algorithms. The most important ANS responses were respiration and skin conductance, while the most important electroencephalographic information was obtained from frontal and central sites. However, all three physiological modalities were outperformed by task performance and movement data. This suggests that future studies should try to demonstrate advantages of physiological measurements over other information sources.

A review on bio-cooperative control in gait rehabilitation

While gait rehabilitation robots have become increasingly common to automate treadmill training, their efficacy is still controversial. Current robots lack the ability to react compliantly to the users voluntary effort and cognitive intention. Bio-cooperative control concepts allow integrating the patient into the control loop as part of the plant rather than seeing him as a source of disturbance. Closed loop control is thereby performed on a physiological and psychological level. In this paper, we review the concept of bio-cooperative control implemented with neurological patients during robot-assisted gait rehabilitation. We highlight its clinical importance and review our work on control strategies that allow bio-cooperative control. We finish by discussing the future potential of bio-cooperative control in rehabilitation robotics.

Effect of Rocking Movements on Respiration

For centuries, rocking has been used to promote sleep in babies or toddlers. Recent research suggested that relaxation could play a role in facilitating the transition from waking to sleep during rocking. Breathing techniques are often used to promote relaxation. However, studies investigating head motions and body rotations showed that vestibular stimulation might elicit a vestibulo-respiratory response, leading to an increase in respiration frequency. An increase in respiration frequency would not be considered to promote relaxation in the first place. On the other hand, a coordination of respiration to rhythmic vestibular stimulation has been observed. Therefore, this study aimed to investigate the effect of different movement frequencies and amplitudes on respiration frequency. Furthermore, we tested whether subjects adapt their respiration to movement frequencies below their spontaneous respiration frequency at rest, which could be beneficial for relaxation. Twenty-one healthy subjects (24–42 years, 12 males) were investigated using an actuated bed, moving along a lateral translation. Following movement frequencies were applied: +30%, +15%, -15%, and -30% of subjects’ rest respiration frequency during baseline (no movement). Furthermore, two different movement amplitudes were tested (Amplitudes: 15 cm, 7.5 cm; movement frequency: 0.3 Hz). In addition, five subjects (25–28 years, 2 males) were stimulated with their individual rest respiration frequency. Rocking movements along a lateral translation caused a vestibulo-respiratory adaptation leading to an increase in respiration frequency. The increase was independent of the applied movement frequencies or amplitudes but did not occur when stimulating with subjects’ rest respiration frequency. Furthermore, no synchronization of the respiration frequency to the movement frequency was observed. In particular, subjects did not lower their respiration frequency below their resting frequency. Hence, it was not feasible to influence respiration in a manner that might be considered beneficial for relaxation.

The Effect of a Slowly Rocking Bed on Sleep

Rocking movements appear to affect human sleep. Recent research suggested a facilitated transition from wake to sleep and a boosting of slow oscillations and sleep spindles due to lateral rocking movements during an afternoon nap. This study aimed at investigating the effect of vestibular stimulation on sleep onset, nocturnal sleep and its potential to increase sleep spindles and slow waves, which could influence memory performance. Polysomnography was recorded in 18 males (age: 20–28 years) during three nights: movement until sleep onset (C1), movement for 2 hours (C2), and one baseline (B) without motion. Sleep dependent changes in memory performance were assessed with a word-pair learning task. Although subjects preferred nights with vestibular stimulation, a facilitated sleep onset or a boost in slow oscillations was not observed. N2 sleep and the total number of sleep spindles increased during the 2 h with vestibular stimulation (C2) but not over the entire night. Memory performance increased over night but did not differ between conditions. The lack of an effect might be due to the already high sleep efficiency (96%) and sleep quality of our subjects during baseline. Nocturnal sleep in good sleepers might not benefit from the potential facilitating effects of vestibular stimulation.