Dataset Information

Optical Myography: Detecting Finger Movements by Looking at the Forearm.

ABSTRACT: One of the crucial problems found in the scientific community of assistive/rehabilitation robotics nowadays is that of automatically detecting what a disabled subject (for instance, a hand amputee) wants to do, exactly when she wants to do it, and strictly for the time she wants to do it. This problem, commonly called "intent detection," has traditionally been tackled using surface electromyography, a technique which suffers from a number of drawbacks, including the changes in the signal induced by sweat and muscle fatigue. With the advent of realistic, physically plausible augmented- and virtual-reality environments for rehabilitation, this approach does not suffice anymore. In this paper, we explore a novel method to solve the problem, which we call Optical Myography (OMG). The idea is to visually inspect the human forearm (or stump) to reconstruct what fingers are moving and to what extent. In a psychophysical experiment involving ten intact subjects, we used visual fiducial markers (AprilTags) and a standard web camera to visualize the deformations of the surface of the forearm, which then were mapped to the intended finger motions. As ground truth, a visual stimulus was used, avoiding the need for finger sensors (force/position sensors, datagloves, etc.). Two machine-learning approaches, a linear and a non-linear one, were comparatively tested in settings of increasing realism. The results indicate an average error in the range of 0.05-0.22 (root mean square error normalized over the signal range), in line with similar results obtained with more mature techniques such as electromyography. If further successfully tested in the large, this approach could lead to vision-based intent detection of amputees, with the main application of letting such disabled persons dexterously and reliably interact in an augmented-/virtual-reality setup.

SUBMITTER: Nissler C

PROVIDER: S-EPMC4827323 | biostudies-literature | 2016

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Optical Myography: Detecting Finger Movements by Looking at the Forearm.

Nissler Christian C Mouriki Nikoleta N Castellini Claudio C

Frontiers in neurorobotics 20160411

One of the crucial problems found in the scientific community of assistive/rehabilitation robotics nowadays is that of automatically detecting what a disabled subject (for instance, a hand amputee) wants to do, exactly when she wants to do it, and strictly for the time she wants to do it. This problem, commonly called "intent detection," has traditionally been tackled using surface electromyography, a technique which suffers from a number of drawbacks, including the changes in the signal induced ...[more]

PMID: 27148039

Similar Datasets

Project description:It is still challenging to achieve a complex grasp or fine finger control by using surface functional electrical stimulation (FES), which usually requires a precise electrode configuration under laboratory or clinical settings. The goals of this study are as follows: 1) to study the possibility of selectively activating individual fingers; 2) to investigate whether the current activation threshold and selective range of individual fingers are affected by two factors: changes in the electrode position and forearm rotation (pronation, neutral and supination); and 3) to explore a theoretical model for guidance of the electrode placement used for selective activation of individual fingers. A coordinate system with more than 400 grid points was established over the forearm skin surface. A searching procedure was used to traverse all grid points to identify the stimulation points for finger extension/flexion by applying monophasic stimulation pulses. Some of the stimulation points for finger extension and flexion were selected and tested in their respective two different forearm postures according to the number and the type of the activated fingers and the strength of finger action response to the electrical stimulation at the stimulation point. The activation thresholds and current ranges of the selectively activated finger at each stimulation point were determined by visual analysis. The stimulation points were divided into three groups ("Low", "Medium" and "High") according to the thresholds of the 1st activated fingers. The angles produced by the selectively activated finger within selective current ranges were measured and analyzed. Selective stimulation of extension/flexion is possible for most fingers. Small changes in electrode position and forearm rotation have no significant effect on the threshold amplitude and the current range for the selective activation of most fingers (p > 0.05). The current range is the largest (more than 2 mA) for selective activation of the thumb, followed by those for the index, ring, middle and little fingers. The stimulation points in the "Low" group for all five fingers lead to noticeable finger angles at low current intensity, especially for the index, middle, and ring fingers. The slopes of the finger angle variation in the "Low" group for digits 2~4 are inversely proportional to the current intensity, whereas the slopes of the finger angle variation in other groups and in all groups for the thumb and little finger are proportional to the current intensity. It is possible to selectively activate the extension/flexion of most fingers by stimulating the forearm muscles. The physiological characteristics of each finger should be considered when placing the negative electrode for selective stimulation of individual fingers. The electrode placement used for the selective activation of individual fingers should not be confined to the location with the lowest activation threshold.

Project description:BACKGROUND:We hypothesized that a selective neural electrical stimulation of radial and median nerves enables the activation of functional movements in the paralyzed hand of individuals with tetraplegia. Compared to previous approaches for which up to 12 muscles were targeted through individual muscular stimulations, we focused on minimizing the number of implanted electrodes however providing almost all the needed and useful hand movements for subjects with complete tetraplegia. METHODS:We performed acute experiments during scheduled surgeries of the upper limb with eligible subjects. We scanned a set of multicontact neural stimulation cuff electrode configurations, pre-computed through modeling simulations. We reported the obtained isolated and functional movements that were considered useful for the subject (different grasping movements). RESULTS:In eight subjects, we demonstrated that selective stimulation based on multicontact cuff electrodes and optimized current spreading over the active contacts provided isolated, compound, functional and strong movements; most importantly 3 out of 4 had isolated fingers or thumb flexion, one patient performed a Key Grip, another one the Power and Hook Grips, and the 2 last all the 3 Grips. Several configurations were needed to target different areas within the nerve to obtain all the envisioned movements. We further confirmed that the upper limb nerves have muscle specific fascicles, which makes it possible to activate isolated movements. CONCLUSIONS:The future goal is to provide patients with functional restoration of object grasping and releasing with a minimally invasive solution: only two cuff electrodes above the elbow. Ethics Committee / ANSM clearance prior to the beginning of the study (inclusion period 2016-2018): CPP Sud Méditerranée, #ID-RCB:2014-A01752-45, first acceptance 10th of February 2015, amended 12th of January 2016. TRIAL REGISTRATION:(www.clinicaltrials.gov): #NCT03721861, Retrospectively registered on 26th of October 2018.

Project description:BackgroundWhat can theories regarding memory-related gaze preference contribute to the field of deception detection? While abundant research has examined the ability to detect concealed information through physiological responses, only recently has the scientific community started to explore how eye tracking can be utilized for that purpose. However, previous attempts to detect deception through eye movements have led to relatively low detection ability in comparison to physiological measures. In the current study, we demonstrate that the modulation of gaze behavior by familiarity, changes considerably when participants perform a visual detection task in comparison to a short-term memory task (that was used in a previous study). Thus, we highlight the importance of theory-based selection of task demands for improving the ability to detect concealed information using eye-movement measures.ResultsDuring visual exploration of four faces (some familiar and some unfamiliar) gaze was allocated preferably on familiar faces, manifested by more fixations. However, this preference tendency vanished once participants were instructed to convey countermeasures and conceal their familiarity by deploying gaze equally to all faces. This gaze behavior during the visual detection task differed significantly from the one observed during a short-term memory task used in a previous study in which a preference towards unfamiliar faces was evident even when countermeasures were applied.ConclusionsDifferent tasks elicit different patterns of gaze behavior towards familiar and unfamiliar faces. Moreover, the ability to voluntarily control gaze behavior is tightly related to task demands. Adequate ability to control gaze was observed in the current visual detection task when memorizing the faces was not required for a successful accomplishment of the task. Thus, applied settings would benefit from a short-term memory task which is much more robust to countermeasure efforts. Beyond shedding light on theories of gaze preference, these findings provide a backbone for future research in the field of deception detection via eye movements.

Dataset Information

Optical Myography: Detecting Finger Movements by Looking at the Forearm.

Publications

Optical Myography: Detecting Finger Movements by Looking at the Forearm.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets