KEKynan Eng

Kynan Eng

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Kynan Eng has a total of 192 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • DHP19: Dynamic Vision Sensor 3D Human Pose Dataset

    by Enrico Calabrese, Gemma Taverni, Christopher Awai Easthope, Sophie Skriabine, Federico Corradi, Luca Longinotti, Kynan Eng, Tobi Delbruck

    Human pose estimation has dramatically improved thanks to the continuous developments in deep learning. However, marker-free human pose estimation based on standard frame-based cameras is still slow and power hungry for real-time feedback interaction because of the huge number of operations necessary for large Convolutional Neural Network (CNN) inference. Event-based cameras such as the Dynamic Vision Sensor (DVS) quickly output sparse moving-edge information. Their sparse and rapid output is ideal for driving low-latency CNNs, thus potentially allowing real-time interaction for human pose estimators. Although the application of CNNs to standard frame-based cameras for human pose estimation is well established, their application to event-based cameras is still under study. This paper proposes a novel benchmark dataset of human body movements, the Dynamic Vision Sensor Human Pose dataset (DHP19). It consists of recordings from 4 synchronized 346x260 pixel DVS cameras, for a set of 33 movements with 17 subjects. DHP19 also includes a 3D pose estimation model that achieves an average 3D pose estimation error of about 8 cm, despite the sparse and reduced input data from the DVS.


  • Open Access

    Effect of a four-week virtual reality-based training versus conventional therapy on upper limb motor function after stroke: A multicenter parallel group randomized trial

    by Corina Schuster-Amft, Kynan Eng, Zorica Suica, Irene Thaler, Sandra Signer, Isabelle Lehmann, Ludwig Schmid, Michael A McCaskey, Miura Hawkins, Martin L Verra, et al.

    Virtual reality-based training has found increasing use in neurorehabilitation to improve upper limb training and facilitate motor recovery. The aim of this study was to directly compare virtual reality-based training with conventional therapy. In a multi-center, parallel-group randomized controlled trial, patients at least 6 months after stroke onset were allocated either to an experimental group (virtual reality-based training) or a control group receiving conventional therapy (16x45 minutes within 4 weeks). The virtual reality-based training system replicated patients´ upper limb movements in real-time to manipulate virtual objects. Blinded assessors tested patients twice before, once during, and twice after the intervention up to 2-month follow-up for dexterity (primary outcome: Box and Block Test), bimanual upper limb function (Chedoke-McMaster Arm and Hand Activity Inventory), and subjective perceived changes (Stroke Impact Scale). 54 eligible patients (70 screened) participated (15 females, mean age 61.3 years, range 20–81 years, time since stroke 3.0±SD 3 years). 22 patients were allocated to the experimental group and 32 to the control group (3 drop-outs). Patients in the experimental and control group improved: Box and Block Test mean 21.5±SD 16 baseline to mean 24.1±SD 17 follow-up; Chedoke-McMaster Arm and Hand Activity Inventory mean 66.0±SD 21 baseline to mean 70.2±SD 19 follow-up. An intention-to-treat analysis found no between-group differences. Patients in the experimental and control group showed similar effects, with most improvements occurring in the first two weeks and persisting until the end of the two-month follow-up period. The study population had moderate to severely impaired motor function at entry (Box and Block Test mean 21.5±SD 16). Patients, who were less impaired (Box and Block Test range 18 to 72) showed higher improvements in favor of the experimental group. This result could suggest that virtual reality-based training might be more applicable for such patients than for more severely impaired patients. ClinicalTrials.gov NCT01774669.


  • Open Access

    Emotion-awareness for intelligent vehicle assistants

    by Hans-Jörg Vögel, Christian Süß, Thomas Hubregtsen, Viviane Ghaderi, Ronee Chadowitz, Elisabeth André, Nicholas Cummins, Björn Schuller, Jérôme Härri, Raphaël Troncy, et al.


  • Open Access

    Home-Based Virtual Reality-Augmented Training Improves Lower Limb Muscle Strength, Balance, and Functional Mobility following Chronic Incomplete Spinal Cord Injury

    by Michael Villiger, Jasmin Liviero, Lea Awai, Rahel Stoop, Pawel Pyk, Ron Clijsen, Armin Curt, Kynan Eng, Marc Bolliger

    Key factors positively influencing rehabilitation and functional recovery after spinal cord injury (SCI) include training variety, intensive movement repetition, and motivating training tasks. Systems supporting these aspects may provide profound gains in rehabilitation, independent of the subject’s treatment location. In the present study, we test the hypotheses that virtual reality (VR)-augmented training at home (i.e., unsupervised) is feasible with subjects with an incomplete SCI (iSCI) and that it improves motor functions such as lower limb muscle strength, balance, and functional mobility. In the study, 12 chronic iSCI subjects used a home-based, mobile version of a lower limb VR training system. The system included motivating training scenarios and combined action observation and execution. Virtual representations of the legs and feet were controlled via movement sensors. The subjects performed home-based training over 4 weeks, with 16–20 sessions of 30–45 min each. The outcome measures assessed were the Lower Extremity Motor Score (LEMS), Berg Balance Scale (BBS), Timed Up and Go (TUG), Spinal Cord Independence Measure mobility, Walking Index for Spinal Cord Injury II, and 10 m and 6 min walking tests. Two pre-treatment assessment time points were chosen for outcome stability: 4 weeks before treatment and immediately before treatment. At post-assessment (i.e., immediately after treatment), high motivation and positive changes were reported by the subjects (adapted Patients’ Global Impression of Change). Significant improvements were shown in lower limb muscle strength (LEMS, P = 0.008), balance (BBS, P = 0.008), and functional mobility (TUG, P = 0.007). At follow-up assessment (i.e., 2–3 months after treatment), functional mobility (TUG) remained significantly improved (P = 0.005) in contrast to the other outcome measures. In summary, unsupervised exercises at home with the VR training system led to beneficial functional training effects in subjects with chronic iSCI, suggesting that it may be useful as a neurorehabilitation tool. Trial registration: Canton of Zurich ethics committee (EK-24/2009, PB_2016-00545), ClinicalTrials.gov: NCT02149186. Registered 24 April 2014.


  • Neural correlates of visuomotor adjustments during scaling of human finger movements

    by Johannes Brand, Lars Michels, Romy Bakker, Marie‐Claude Hepp‐Reymond, Daniel Kiper, Manfred Morari, Kynan Eng

    Visually guided finger movements include online feedback of current effector position to guide target approach. This visual feedback may be scaled or otherwise distorted by unpredictable perturbations. Although adjustments to visual feedback scaling have been studied before, the underlying brain activation differences between upscaling (visual feedback larger than real movement) and downscaling (feedback smaller than real movement) are currently unknown. Brain activation differences between upscaling and downscaling might be expected because within‐trial adjustments during upscaling require corrective backwards accelerations, whereas correcting for downscaling requires forward accelerations. In this behavioural and fMRI study we investigated adjustments during up‐ and downscaling in a target‐directed finger flexion–extension task with real‐time visual feedback. We found that subjects made longer and more complete within‐trial corrections for downscaling perturbations than for upscaling perturbations. The finger task activated primary motor (M1) and somatosensory (S1) areas, premotor and parietal regions, basal ganglia, and cerebellum. General scaling effects were seen in the right pre‐supplementary motor area, dorsal anterior cingulate cortex, inferior parietal lobule, and dorsolateral prefrontal cortex. Stronger activations for down‐ than for upscaling were observed in M1, supplementary motor area (SMA), S1 and anterior cingulate cortex. We argue that these activation differences may reflect differing online correction for upscaling vs. downscaling during finger flexion‐extension.


  • Open Access

    Virtual Hand Feedback Reduces Reaction Time in an Interactive Finger Reaching Task

    by Johannes Brand, Marco Piccirelli, Marie-Claude Hepp-Reymond, Manfred Morari, Lars Michels, Kynan Eng

    Computer interaction via visually guided hand or finger movements is a ubiquitous part of daily computer usage in work or gaming. Surprisingly, however, little is known about the performance effects of using virtual limb representations versus simpler cursors. In this study 26 healthy right-handed adults performed cued index finger flexion-extension movements towards an on-screen target while wearing a data glove. They received each of four different types of real-time visual feedback: a simple circular cursor, a point light pattern indicating finger joint positions, a cartoon hand and a fully shaded virtual hand. We found that participants initiated the movements faster when receiving feedback in the form of a hand than when receiving circular cursor or point light feedback. This overall difference was robust for three out of four hand versus circle pairwise comparisons. The faster movement initiation for hand feedback was accompanied by a larger movement amplitude and a larger movement error. We suggest that the observed effect may be related to priming of hand information during action perception and execution affecting motor planning and execution. The results may have applications in the use of body representations in virtual reality applications.


  • Open Access

    Relationship between structural brainstem and brain plasticity and lower-limb training in spinal cord injury: a longitudinal pilot study

    by Michael Villiger, Patrick Grabher, Marie-Claude Hepp-Reymond, Daniel Kiper, Armin Curt, Marc Bolliger, Sabina Hotz-Boendermaker, Spyros Kollias, Kynan Eng, Patrick Freund

    Rehabilitative training has shown to improve significantly motor outcomes and functional walking capacity in patients with incomplete spinal cord injury (iSCI). However, whether performance improvements during rehabilitation relate to brain plasticity or whether it is based on functional adaptation of movement strategies remain uncertain. This study assessed training improvement-induced structural brain plasticity in chronic iSCI patients using longitudinal MRI. We used tensor-based morphometry (TBM) to analyze longitudinal brain volume changes associated with intensive virtual reality (VR)-augmented lower limb training in nine traumatic iSCI patients. The MRI data was acquired before and after a 4-week training period (16-20 training sessions). Before training, voxel-based morphometry (VBM) and voxel-based cortical thickness (VBCT) assessed baseline morphometric differences in nine iSCI patients compared to 14 healthy controls. The intense VR-augmented training of limb control improved significantly balance, walking speed, ambulation, and muscle strength in patients. Retention of clinical improvements was confirmed by the 3-4 months follow-up. In patients relative to controls, reductions in VBM of white matter volume within the brainstem and cerebellum and VBCT showed cortical thinning in the primary motor cortex. Over time, TBM revealed significant improvement-induced increases in the left middle temporal and occipital gyrus, left temporal pole and fusiform gyrus, both hippocampi, cerebellum, corpus callosum, and brainstem in iSCI patients. This study demonstrates structural plasticity at the cortical and brainstem level as a consequence of VR-augmented training in iSCI patients. These structural changes may serve as neuroimaging biomarkers of VR-augmented lower limb neurorehabilitation in addition to performance measures to detect improvements in rehabilitative training.


  • Open Access

    Using mixed methods to evaluate efficacy and user expectations of a virtual reality–based training system for upper-limb recovery in patients after stroke: a study protocol for a randomised controlled trial

    by Corina Schuster-Amft, Kynan Eng, Isabelle Lehmann, Ludwig Schmid, Nagisa Kobashi, Irène Thaler, Martin L Verra, Andrea Henneke, Sandra Signer, Michael McCaskey, et al.

    In recent years, virtual reality has been introduced to neurorehabilitation, in particular with the intention of improving upper-limb training options and facilitating motor function recovery. The proposed study incorporates a quantitative part and a qualitative part, termed a mixed-methods approach: (1) a quantitative investigation of the efficacy of virtual reality training compared to conventional therapy in upper-limb motor function are investigated, (2a) a qualitative investigation of patients’ experiences and expectations of virtual reality training and (2b) a qualitative investigation of therapists’ experiences using the virtual reality training system in the therapy setting. At three participating clinics, 60 patients at least 6 months after stroke onset will be randomly allocated to an experimental virtual reality group (EG) or to a control group that will receive conventional physiotherapy or occupational therapy (16 sessions, 45 minutes each, over the course of 4 weeks). Using custom data gloves, patients’ finger and arm movements will be displayed in real time on a monitor, and they will move and manipulate objects in various virtual environments. A blinded assessor will test patients’ motor and cognitive performance twice before, once during, and twice after the 4-week intervention. The primary outcome measure is the Box and Block Test. Secondary outcome measures are the Chedoke-McMaster Stroke Assessments (hand, arm and shoulder pain subscales), the Chedoke-McMaster Arm and Hand Activity Inventory, the Line Bisection Test, the Stroke Impact Scale, the MiniMentalState Examination and the Extended Barthel Index. Semistructured face-to-face interviews will be conducted with patients in the EG after intervention finalization with a focus on the patients’ expectations and experiences regarding the virtual reality training. Therapists’ perspectives on virtual reality training will be reviewed in three focus groups comprising four to six occupational therapists and physiotherapists. The interviews will help to gain a deeper understanding of the phenomena under investigation to provide sound recommendations for the implementation of the virtual reality training system for routine use in neurorehabilitation complementing the quantitative clinical assessments. Cliniclatrials.gov Identifier: NCT01774669 (15 January 2013)


  • Intensive virtual reality-based training for upper limb motor function in chronic stroke: a feasibility study using a single case experimental design and fMRI

    by Corina Schuster-Amft, Andrea Henneke, Birgit Hartog-Keisker, Lisa Holper, Ewa Siekierka, Edith Chevrier, Pawel Pyk, Spyros Kollias, Daniel Kiper, Kynan Eng

    To evaluate feasibility and neurophysiological changes after virtual reality (VR)-based training of upper limb (UL) movements. Single-case A-B-A-design with two male stroke patients (P1:67 y and 50 y, 3.5 and 3 y after onset) with UL motor impairments, 45-min therapy sessions 5×/week over 4 weeks. Patients facing screen, used bimanual data gloves to control virtual arms. Three applications trained bimanual reaching, grasping, hand opening. Assessments during 2-week baseline, weekly during intervention, at 3-month follow-up (FU): Goal Attainment Scale (GAS), Chedoke Arm and Hand Activity Inventory (CAHAI), Chedoke-McMaster Stroke Assessment (CMSA), Extended Barthel Index (EBI), Motor Activity Log (MAL). Functional magnetic resonance imaging scans (FMRI) before, immediately after treatment and at FU. P1 executed 5478 grasps (paretic arm). Improvements in CAHAI (+4) were maintained at FU. GAS changed to +1 post-test and +2 at FU. P2 executed 9835 grasps (paretic arm). CAHAI improvements (+13) were maintained at FU. GAS scores changed to -1 post-test and +1 at FU. MAL scores changed from 3.7 at pre-test to 5.5 post-test and 3.3 at FU. The VR-based intervention was feasible, safe, and intense. Adjustable application settings maintained training challenge and patient motivation. ADL-relevant UL functional improvements persisted at FU and were related to changed cortical activation patterns. Implications for Rehabilitation YouGrabber trains uni- and bimanual upper motor function. Its application is feasible, safe, and intense. The control of the virtual arms can be done in three main ways: (a) normal (b) virtual mirror therapy, or (c) virtual following. The mirroring feature provides an illusion of affected limb movements during the period when the affected upper limb (UL) is resting. The YouGrabber training led to ADL-relevant UL functional improvements that were still assessable 12 weeks after intervention finalization and were related to changed cortical activation patterns.


  • Open Access

    Enhanced Activation of Motor Execution Networks Using Action Observation Combined with Imagination of Lower Limb Movements

    by Michael Villiger, Natalia Estévez, Marie-Claude Hepp-Reymond, Daniel Kiper, Spyros S Kollias, Kynan Eng, Sabina Hotz-Boendermaker

    The combination of first-person observation and motor imagery, i.e. first-person observation of limbs with online motor imagination, is commonly used in interactive 3D computer gaming and in some movie scenes. These scenarios are designed to induce a cognitive process in which a subject imagines himself/herself acting as the agent in the displayed movement situation. Despite the ubiquity of this type of interaction and its therapeutic potential, its relationship to passive observation and imitation during observation has not been directly studied using an interactive paradigm. In the present study we show activation resulting from observation, coupled with online imagination and with online imitation of a goal-directed lower limb movement using functional MRI (fMRI) in a mixed block/event-related design. Healthy volunteers viewed a video (first-person perspective) of a foot kicking a ball. They were instructed to observe-only the action (O), observe and simultaneously imagine performing the action (O-MI), or imitate the action (O-IMIT). We found that when O-MI was compared to O, activation was enhanced in the ventralpremotor cortex bilaterally, left inferior parietal lobule and left insula. The O-MI and O-IMIT conditions shared many activation foci in motor relevant areas as confirmed by conjunction analysis. These results show that (i) combining observation with motor imagery (O-MI) enhances activation compared to observation-only (O) in the relevant foot motor network and in regions responsible for attention, for control of goal-directed movements and for the awareness of causing an action, and (ii) it is possible to extensively activate the motor execution network using O-MI, even in the absence of overt movement. Our results may have implications for the development of novel virtual reality interactions for neurorehabilitation interventions and other applications involving training of motor tasks.