Automatic determination of synergies by radial basis function artificial neural networks for the control of a neural prosthesis

Transhumeral Arm Reaching Motion Prediction through Deep Reinforcement Learning-Based Synthetic Motion Cloning

The lack of intuitive controllability remains a primary challenge in enabling transhumeral amputees to control a prosthesis for arm reaching with residual limb kinematics. Recent advancements in prosthetic arm control have focused on leveraging the predictive capabilities of artificial neural networks (ANNs) to automate elbow joint motion and wrist pronation-supination during target reaching tasks. However, large quantities of human motion data collected from different subjects for various activities of daily living (ADL) tasks are required to train these ANNs. For example, the reaching motion can be altered when the height of the desk is changed; however, it is cumbersome to conduct human experiments for all conditions. This paper proposes a framework for cloning motion datasets using deep reinforcement learning (DRL) to cater to training data requirements. DRL algorithms have been demonstrated to create human-like synergistic motion in humanoid agents to handle redundancy and optimize movements. In our study, we collected real motion data from six individuals performing multi-directional arm reaching tasks in the horizontal plane. We generated synthetic motion data that mimicked similar arm reaching tasks by utilizing a physics simulation and DRL-based arm manipulation. We then trained a CNN-LSTM network with different configurations of training motion data, including DRL, real, and hybrid datasets, to test the efficacy of the cloned motion data. The results of our evaluation showcase the effectiveness of the cloned motion data in training the ANN to predict natural elbow motion accurately across multiple subjects. Furthermore, motion data augmentation through combining real and cloned motion datasets has demonstrated the enhanced robustness of the ANN by supplementing and diversifying the limited training data. These findings have significant implications for creating synthetic dataset resources for various arm movements and fostering strategies for automatized prosthetic elbow motion.

Synergy-Space Recurrent Neural Network for Transferable Forearm Motion Prediction from Residual Limb Motion

Transhumeral amputees experience considerable difficulties with controlling a multifunctional prosthesis (powered hand, wrist, and elbow) due to the lack of available muscles to provide electromyographic (EMG) signals. The residual limb motion strategy has become a popular alternative for transhumeral prosthesis control. It provides an intuitive way to estimate the motion of the prosthesis based on the residual shoulder motion, especially for target reaching tasks. Conventionally, a predictive model, typically an artificial neural network (ANN), is directly trained and relied upon to map the shoulder-elbow kinematics using the data from able-bodied subjects without extracting any prior synergistic information. However, it is essential to explicitly identify effective synergies and make them transferable across amputee users for higher accuracy and robustness. To overcome this limitation of the conventional ANN learning approach, this study explicitly combines the kinematic synergies with a recurrent neural network (RNN) to propose a synergy-space neural network for estimating forearm motions (i.e., elbow joint flexion-extension and pronation-supination angles) based on residual shoulder motions. We tested 36 training strategies for each of the 14 subjects, comparing the proposed synergy-space and conventional neural network learning approaches, and we statistically evaluated the results using Pearson's correlation method and the analysis of variance (ANOVA) test. The offline cross-subject analysis indicates that the synergy-space neural network exhibits superior robustness to inter-individual variability, demonstrating the potential of this approach as a transferable and generalized control strategy for transhumeral prosthesis control.

A novel framework for designing a multi-DoF prosthetic wrist control using machine learning

Prosthetic arms can significantly increase the upper limb function of individuals with upper limb loss, however despite the development of various multi-DoF prosthetic arms the rate of prosthesis abandonment is still high. One of the major challenges is to design a multi-DoF controller that has high precision, robustness, and intuitiveness for daily use. The present study demonstrates a novel framework for developing a controller leveraging machine learning algorithms and movement synergies to implement natural control of a 2-DoF prosthetic wrist for activities of daily living (ADL). The data was collected during ADL tasks of ten individuals with a wrist brace emulating the absence of wrist function. Using this data, the neural network classifies the movement and then random forest regression computes the desired velocity of the prosthetic wrist. The models were trained/tested with ADLs where their robustness was tested using cross-validation and holdout data sets. The proposed framework demonstrated high accuracy (F-1 score of 99% for the classifier and Pearson's correlation of 0.98 for the regression). Additionally, the interpretable nature of random forest regression was used to verify the targeted movement synergies. The present work provides a novel and effective framework to develop an intuitive control for multi-DoF prosthetic devices.

Dimensionality Reduction and Motion Clustering During Activities of Daily Living: Decoupling Hand Location and Orientation

This article is the second in a two-part series analyzing human arm and hand motion during a wide range of unstructured tasks. In this work, we track the hand of healthy individuals as they perform a variety of activities of daily living (ADLs) in three ways decoupled from hand orientation: end-point locations of the hand trajectory, whole path trajectories of the hand, and straight-line paths generated using start and end points of the hand. These data are examined by a clustering procedure to reduce the wide range of hand use to a smaller representative set. Hand orientations are subsequently analyzed for the end-point location clustering results and subsets of orientations are identified in three reference frames: global, torso, and forearm. Data driven methods that are used include dynamic time warping (DTW), DTW barycenter averaging (DBA), and agglomerative hierarchical clustering with Ward's linkage. Analysis of the end-point locations, path trajectory, and straight-line path trajectory identified 5, 5, and 7 ADL task categories, respectively, while hand orientation analysis identified up to 4 subsets of orientations for each task location, discretized and classified to the facets of a rhombicuboctahedron. Together these provide insight into our hand usage in daily life and inform an implementation in prosthetic or robotic devices using sequential control.

Synergistic Elbow Control for a Myoelectric Transhumeral Prosthesis

This paper presents a control approach for a myoelectric transhumeral prostheses that supplements a typical two-site electromyograms (EMG) input with the inertial measurement of arm motion. The inertial measurement is employed to: 1) provide synergistic movement between the prosthetic elbow joint and intact upper arm and 2) to switch control between the myoelectric elbow and hand. In order to assess the prospective efficacy of the control method, experiments were conducted on six healthy subjects who performed a series of pick-and-place tasks within a virtual environment. The assessments compared the time required to complete the pick-and-place tasks using the proposed coordinated control approach, with the time required using a sequential control approach (i.e., the conventional approach used in commercial devices). Subjects on average completed the pick-and-place tasks 34% faster with the coordinated control approach, relative to the conventional sequential EMG method, with no difference in compensatory torso motion.

Movement-Based Control for Upper-Limb Prosthetics: Is the Regression Technique the Key to a Robust and Accurate Control?

Due to the limitations of myoelectric control (such as dependence on muscular fatigue and on electrodes shift, difficulty in decoding complex patterns or in dealing with simultaneous movements), there is a renewal of interest in the movement-based control approaches for prosthetics. The latter use residual limb movements rather than muscular activity as command inputs, in order to develop more natural and intuitive control techniques. Among those, several research works rely on the interjoint coordinations that naturally exist in human upper limb movements. These relationships are modeled to control the distal joints (e.g., elbow) based on the motions of proximal ones (e.g., shoulder). The regression techniques, used to model the coordinations, are various [Artificial Neural Networks, Principal Components Analysis (PCA), etc.] and yet, analysis of their performance and impact on the prosthesis control is missing in the literature. Is there one technique really more efficient than the others to model interjoint coordinations? To answer this question, we conducted an experimental campaign to compare the performance of three common regression techniques in the control of the elbow joint on a transhumeral prosthesis. Ten non-disabled subjects performed a reaching task, while wearing an elbow prosthesis which was driven by several interjoint coordination models obtained through different regression techniques. The models of the shoulder-elbow kinematic relationship were built from the recordings of fifteen different non-disabled subjects that performed a similar reaching task with their healthy arm. Among Radial Basis Function Networks (RBFN), Locally Weighted Regression (LWR), and PCA, RBFN was found to be the most robust, based on the analysis of several criteria including the quality of generated movements but also the compensatory strategies exhibited by users. Yet, RBFN does not significantly outperform LWR and PCA. The regression technique seems not to be the most significant factor for improvement of interjoint coordinations-based control. By characterizing the impact of the modeling techniques through closed-loop experiments with human users instead of purely offline simulations, this work could also help in improving movement-based control approaches and in bringing them closer to a real use by patients.

Can We Achieve Intuitive Prosthetic Elbow Control Based on Healthy Upper Limb Motor Strategies?

Most transhumeral amputees report that their prosthetic device lacks functionality, citing the control strategy as a major limitation. Indeed, they are required to control several degrees of freedom with muscle groups primarily used for elbow actuation. As a result, most of them choose to have a one-degree-of-freedom myoelectric hand for grasping objects, a myoelectric wrist for pronation/supination, and a body-powered elbow. Unlike healthy upper limb movements, the prosthetic elbow joint angle, adjusted prior to the motion, is not involved in the overall upper limb movements, causing the rest of the body to compensate for the lack of mobility of the prosthesis. A promising solution to improve upper limb prosthesis control exploits the residual limb mobility: like in healthy movements, shoulder and prosthetic elbow motions are coupled using inter-joint coordination models. The present study aims to test this approach. A transhumeral amputated individual used a prosthesis with a residual limb motion-driven elbow to point at targets. The prosthetic elbow motion was derived from IMU-based shoulder measurements and a generic model of inter-joint coordinations built from healthy individuals data. For comparison, the participant also performed the task while the prosthetic elbow was implemented with his own myoelectric control strategy. The results show that although the transhumeral amputated participant achieved the pointing task with a better precision when the elbow was myoelectrically-controlled, he had to develop large compensatory trunk movements. Automatic elbow control reduced trunk displacements, and enabled a more natural body behavior with synchronous shoulder and elbow motions. However, due to socket impairments, the residual limb amplitudes were not as large as those of healthy shoulder movements. Therefore, this work also investigates if a control strategy whereby prosthetic joints are automatized according to healthy individuals' coordination models can lead to an intuitive and natural prosthetic control.

IMU-Based Wrist Rotation Control of a Transradial Myoelectric Prosthesis

This paper describes a control method intended to facilitate improved control of a myoelectric prosthesis containing a wrist rotator. Rather than exclusively utilizing electromyogram (EMG) for the control of all myoelectric components (e.g., a hand and a wrist), the proposed controller utilizes inertial measurement (from six-axis inertial measurement unit (IMU)) to sense upper arm abduction/adduction, and uses this input to command a wrist rotation velocity. As such, the controller essentially substitutes shoulder abduction/adduction in place of agonist/antagonist EMG to control wrist angular velocity, which preserves EMG for control of the hand (or other arm components). As a preliminary assessment of efficacy, the control method was implemented on a transradial prosthesis prototype with a powered wrist rotator and hand, and experimentally assessed on five able-bodied subjects who wore the prototype using an able-bodied adaptor and one transradial amputee subject while performing assessments representative of activities of daily living. The assessments compared the (timed) performance of the combined EMG/ IMU-based control method with a (conventional) sequential EMG control approach. Results of the assessment indicate that the able-bodied subjects were able to perform the tasks 33% faster on average with the EMG/IMU-based method, relative to a conventional sequential EMG method. The same assessment was subsequently conducted using a single transradial amputee subject, which resulted in similar performance trends, although with a somewhat lessened effect size-specifically, the amputee subject was on average 22% faster in performing tasks with the IMU-based controller.

Candidates for synergies: linear discriminants versus principal components

Movement primitives or synergies have been extracted from human hand movements using several matrix factorization, dimensionality reduction, and classification methods. Principal component analysis (PCA) is widely used to obtain the first few significant eigenvectors of covariance that explain most of the variance of the data. Linear discriminant analysis (LDA) is also used as a supervised learning method to classify the hand postures corresponding to the objects grasped. Synergies obtained using PCA are principal component vectors aligned with dominant variances. On the other hand, synergies obtained using LDA are linear discriminant vectors that separate the groups of variances. In this paper, time varying kinematic synergies in the human hand grasping movements were extracted using these two diametrically opposite methods and were evaluated in reconstructing natural and American sign language (ASL) postural movements. We used an unsupervised LDA (ULDA) to extract linear discriminants. The results suggest that PCA outperformed LDA. The uniqueness, advantages, and disadvantages of each of these methods in representing high-dimensional hand movements in reduced dimensions were discussed.

Microsoft kinect-based artificial perception system for control of functional electrical stimulation assisted grasping

We present a computer vision algorithm that incorporates a heuristic model which mimics a biological control system for the estimation of control signals used in functional electrical stimulation (FES) assisted grasping. The developed processing software acquires the data from Microsoft Kinect camera and implements real-time hand tracking and object analysis. This information can be used to identify temporal synchrony and spatial synergies modalities for FES control. Therefore, the algorithm acts as artificial perception which mimics human visual perception by identifying the position and shape of the object with respect to the position of the hand in real time during the planning phase of the grasp. This artificial perception used within the heuristically developed model allows selection of the appropriate grasp and prehension. The experiments demonstrate that correct grasp modality was selected in more than 90% of tested scenarios/objects. The system is portable, and the components are low in cost and robust; hence, it can be used for the FES in clinical or even home environment. The main application of the system is envisioned for functional electrical therapy, that is, intensive exercise assisted with FES.

A novel five degree of freedom user command controller in people with spinal cord injury and non-injured for full upper extremity neuroprostheses, wearable powered orthoses and prosthetics

An independent lifestyle requires the ability to place the hand in the complete workspace in concert with hand grasp and release. A novel user command controller monitoring head position for purpose of controlling hand location and orientation is proposed and demonstrated. The controller detected five degrees of freedom which were applied to upper limb movements including forearm and hand placement in three-dimensional space. The controller was evaluated by having subjects complete tracking tasks manipulating a simulated on-screen upper limb representation. Thirteen of the eighteen subjects assessed using the controller had sustained a spinal cord injury at or above the sixth cervical vertebra. Two of the injured subjects with decreased cervical spine mobility were unable to operate the controller. The remaining subjects performed the tracking tasks effectively after minimal training. This simple five-degree of freedom controller has been proposed for the use by those disabled by upper limb amputation, paralysis, weakness or hypertonicity.

Life-like Control for Neural Prostheses: "Proximal Controls Distal"

We describe the model and implementation of the hierarchical hybrid method for controlling of the lower-arm (pronation/supination and elbow flexion/extension) in humans with disabilities. The control follows the strategy found in ablebodied humans where the movement is planned based on the task and the most distal part of the arm; yet, the command starts from the most proximal segment. The controller uses a black box of the movement and relies on temporal and spatial synergies. The driving signals are the shoulder flexion/extension velocity and acceleration, the outputs are four stimulation patterns for the control of elbow flexion/extension and pronation/supination. The operation is discrete at the voluntary and coordination levels, and continuous at the actuator level. The repertoire of movement that were considered was limited to a set of typical daily activities within the normal workspace in the sitting position only. The main application of this control is the therapeutic electrical stimulation in post-stroke hemiplegic patients.

Artificial neural networks based prediction of cerebral palsy in infants with central coordination disturbance

BACKGROUND

In a previous study we demonstrated that heart variability parameters (HRV) could be helpful clinically as well as a prognostic tool in infants with central coordination disturbance (CCD). In recent years, outcome predictions using artificial neural networks (ANN) have been developed in many areas of health care research, but there are no published studies considered ANN models for prediction of cerebral palsy (CP) development.

OBJECTIVE

To compare the results of an ANN analysis with results of regression analysis, using the same data set and the same clinical and HRV parameters.

METHODS

The study included 35 infants with CCD and 37 healthy age and sex-matched controls. Time-domain HRV indices were analyzed from 24h electrocardiography recordings. Clinical parameters and selected time domain HRV parameters are used to predict CP by logistic regression, and then an ANN analysis was applied to the same data set. Input variables were age, gender, postural responses, heart rate parameters (minimum, maximum and average), and time domain parameters of HRV (SDNN, SDANN and RMSSD). For each of one the pairs of ANN and clinical predictors, the area under the receiver operating characteristic (ROC) curves with test accuracy parameters were calculated and compared.

RESULTS

In the observed dataset, ANN model overall correctly classified all infants, compared with 86.11% correct classification for the logistic regression model, and compared with 67.65% and 77.14% for SDANN and SDNN respectively.

CONCLUSIONS

ANN model, based on clinical and HRV data can predict development of CP in patients with CCD with accuracy greater than 90%. Our results strongly indicate that a well-validated ANN may have a role in the clinical prediction of CP in infants with CCD.

Active upper limb prosthesis based on natural movement trajectories

The motion of the current prostheses is sequential and does not allow natural movements. In this work, complex natural motion patterns from a healthy upper limb were characterized in order to be emulated for a trans-humeral prosthesis with three degrees of freedom at the elbow. Firstly, it was necessary to define the prosthesis workspace, which means to establish a relationship using an artificial neural network (ANN), between the arm-forearm (3-D) angles allowed by the prosthesis, and its actuators length. The 3-D angles were measured between the forearm and each axis of the reference system attached at the elbow. Secondly, five activities of daily living (ADLs) were analyzed by means of the elbow flexion (EF), the forearm prono-supination (FPS) and the 3-D angles, from healthy subjects, by using a video-based motion analysis system. The 3-D angles were fed to the prosthesis model (ANN) in order to analyze which ADLs could be emulated by the prosthesis. As a result, a prosthesis kinematics approximation was obtained. In conclusion, in spite of the innovative mechanical configuration of the actuators, it was possible to carry out only three of the five ADLs considered. Future work will include improvement of the mechanical configuration of the prosthesis to have greater range of motion.

Optimization and evaluation of a proportional derivative controller for planar arm movement

In most clinical applications of functional electrical stimulation (FES), the timing and amplitude of electrical stimuli have been controlled by open-loop pattern generators. The control of upper extremity reaching movements, however, will require feedback control to achieve the required precision. Here we present three controllers using proportional derivative (PD) feedback to stimulate six arm muscles, using two joint angle sensors. Controllers were first optimized and then evaluated on a computational arm model that includes musculoskeletal dynamics. Feedback gains were optimized by minimizing a weighted sum of position errors and muscle forces. Generalizability of the controllers was evaluated by performing movements for which the controller was not optimized, and robustness was tested via model simulations with randomly weakened muscles. Robustness was further evaluated by adding joint friction and doubling the arm mass. After optimization with a properly weighted cost function, all PD controllers performed fast, accurate, and robust reaching movements in simulation. Oscillatory behavior was seen after improper tuning. Performance improved slightly as the complexity of the feedback gain matrix increased.

Identifying representative synergy matrices for describing muscular activation patterns during multidirectional reaching in the horizontal plane

Muscle synergies have been proposed as a simplifying principle of generation of movements based on a low-dimensional control by the CNS. This principle may be useful for movement restoration by, e.g., functional electrical stimulation (FES), if a limited set of synergies can describe several functional tasks. This study investigates the possibility of describing a multijoint reaching task of the upper limb by a linear combination of one set of muscle synergies common to multiple directions. Surface electromyographic (EMG) signals were recorded from 12 muscles of the dominant upper limb of eight healthy men during single-joint movements and a multijoint reaching task in 12 directions in the horizontal plane. The movement kinematics was recorded by a motion analysis system. Muscle synergies were extracted with nonnegative matrix factorization of the EMG envelopes. Synergies were computed either from the single-joint movements to describe the two degrees of freedom independently or from the multijoint movements. On average, the multijoint reaching task could be accurately described in all the directions (coefficient of determination >0.85) by a linear combination of either four synergies extracted from the individual degrees of freedom or three synergies extracted from multijoint movements in at least three reaching directions. These results indicate that a large set of multijoint movements can be generated by a synergy matrix of limited dimensionality and common to all directions if the synergies are extracted from a representative number of directions. The linear combination of synergies may thus be used in strategies for restoring functions, such as FES.

Temporal postural synergies of the hand in rapid grasping tasks

Postural synergies of the hand have been widely proposed in the literature, but only a few attempts were made to visualize temporal postural synergies, i.e., profiles of postural synergies varying over time. This paper aims to derive temporal postural synergies from kinematic synergies extracted from joint angular velocity profiles of rapid grasping movements. The rapid movements constrain the kinematic synergies to combine instantaneously, and thus, the movements can be approximated by a weighted summation of synchronous synergies. After being extracted by using singular value decomposition, the synchronous kinematic synergies were translated into temporal postural synergies, which revealed strategies of enslaving, metacarpal flexion for larger movements, and hierarchical recruitment of joints, adapted by subjects while grasping.

Multivariate prediction of upper limb prosthesis acceptance or rejection

OBJECTIVE

To develop a model for prediction of upper limb prosthesis use or rejection.

DESIGN

A questionnaire exploring factors in prosthesis acceptance was distributed internationally to individuals with upper limb absence through community-based support groups and rehabilitation hospitals.

SUBJECTS

A total of 191 participants (59 prosthesis rejecters and 132 prosthesis wearers) were included in this study.

METHODS

A logistic regression model, a C5.0 decision tree, and a radial basis function neural network were developed and compared in terms of sensitivity (prediction of prosthesis rejecters), specificity (prediction of prosthesis wearers), and overall cross-validation accuracy.

RESULTS

The logistic regression and neural network provided comparable overall accuracies of approximately 84 +/- 3%, specificity of 93%, and sensitivity of 61%. Fitting time-frame emerged as the predominant predictor. Individuals fitted within two years of birth (congenital) or six months of amputation (acquired) were 16 times more likely to continue prosthesis use.

CONCLUSIONS

To increase rates of prosthesis acceptance, clinical directives should focus on timely, client-centred fitting strategies and the development of improved prostheses and healthcare for individuals with high-level or bilateral limb absence. Multivariate analyses are useful in determining the relative importance of the many factors involved in prosthesis acceptance and rejection.

PREDICTION OF ELBOW TRAJECTORY FROM SHOULDER ANGLES USING NEURAL NETWORKS

Patients with transhumeral amputations and C5/C6 quadriplegia may be able to use voluntary shoulder motion as command signals for powered prostheses and functional electrical stimulation, respectively. Spatiotemporal synergies exist between the shoulder and elbow joints for goal-oriented reaching movements as performed by able-bodied subjects. We are using a multi-layer perceptron neural network to discover and embody these synergies. Such a network could be used as a high-level controller that could predict the desired distal arm joint kinematics from the voluntary movements of the shoulder joint of an able-bodied subject. We evaluated this for a task that involved reaching to 16 targets in a horizontal plane. After reaching reasonable offline prediction accuracy for our neural networks, we then deployed the best network to make real-time predictions of the elbow angles and examined its performance on both inter- and intra-subject trials. Finally, we extended the model to utilize the five degrees-of-freedom at the shoulder to control the five degrees-of-freedom required for a prosthetic arm and hand to reach and grasp variously oriented objects in the extrapersonal workspace. Such a system, although very simple, was readily controllable for a reach and grasp task presented to the subject in a virtual reality environment.

Synergistic control of forearm based on accelerometer data and artificial neural networks

In the present study, we modeled a reaching task as a two-link mechanism. The upper arm and forearm motion trajectories during vertical arm movements were estimated from the measured angular accelerations with dual-axis accelerometers. A data set of reaching synergies from able-bodied individuals was used to train a radial basis function artificial neural network with upper arm/forearm tangential angular accelerations. The trained radial basis function artificial neural network for the specific movements predicted forearm motion from new upper arm trajectories with high correlation (mean, 0.9149-0.941). For all other movements, prediction was low (range, 0.0316-0.8302). Results suggest that the proposed algorithm is successful in generalization over similar motions and subjects. Such networks may be used as a high-level controller that could predict forearm kinematics from voluntary movements of the upper arm. This methodology is suitable for restoring the upper limb functions of individuals with motor disabilities of the forearm, but not of the upper arm. The developed control paradigm is applicable to upper-limb orthotic systems employing functional electrical stimulation. The proposed approach is of great significance particularly for humans with spinal cord injuries in a free-living environment. The implication of a measurement system with dual-axis accelerometers, developed for this study, is further seen in the evaluation of movement during the course of rehabilitation. For this purpose, training-related changes in synergies apparent from movement kinematics during rehabilitation would characterize the extent and the course of recovery. As such, a simple system using this methodology is of particular importance for stroke patients. The results underlie the important issue of upper-limb coordination.

A radial basis classifier for the automatic detection of aspiration in children with dysphagia

Background

Silent aspiration or the inhalation of foodstuffs without overt physiological signs presents a serious health issue for children with dysphagia. To date, there are no reliable means of detecting aspiration in the home or community. An assistive technology that performs in these environments could inform caregivers of adverse events and potentially reduce the morbidity and anxiety of the feeding experience for the child and caregiver, respectively. This paper proposes a classifier for automatic classification of aspiration and swallow vibration signals non-invasively recorded on the neck of children with dysphagia.

Methods

Vibration signals associated with safe swallows and aspirations, both identified via videofluoroscopy, were collected from over 100 children with neurologically-based dysphagia using a single-axis accelerometer. Five potentially discriminatory mathematical features were extracted from the accelerometry signals. All possible combinations of the five features were investigated in the design of radial basis function classifiers. Performance of different classifiers was compared and the best feature sets were identified.

Results

Optimal feature combinations for two, three and four features resulted in statistically comparable adjusted accuracies with a radial basis classifier. In particular, the feature pairing of dispersion ratio and normality achieved an adjusted accuracy of 79.8 ± 7.3%, a sensitivity of 79.4 ± 11.7% and specificity of 80.3 ± 12.8% for aspiration detection. Addition of a third feature, namely energy, increased adjusted accuracy to 81.3 ± 8.5% but the change was not statistically significant. A closer look at normality and dispersion ratio features suggest leptokurticity and the frequency and magnitude of atypical values as distinguishing characteristics between swallows and aspirations. The achieved accuracies are 30% higher than those reported for bedside cervical auscultation.

Conclusion

The proposed aspiration classification algorithm provides promising accuracy for aspiration detection in children. The classifier is conducive to hardware implementation as a non-invasive, portable "aspirometer". Future research should focus on further enhancement of accuracy rates by considering other signal features, classifier methods, or an augmented variety of training samples. The present study is an important first step towards the eventual development of wearable intelligent intervention systems for the diagnosis and management of aspiration.

The effects of training set on prediction of elbow trajectory from shoulder trajectory during reaching to targets

Patients with transhumeral amputations and C5/C6 quadriplegia may be able to use voluntary shoulder motion as command signals for powered prostheses and functional electrical stimulation, respectively. Spatio-temporal synergies exist for goal oriented reaching movements between the shoulder and elbow joints in able bodied subjects. We are using a multi-layer perceptron neural network to discover and embody these synergies. Such a network could be used as a high level functional electrical stimulation (FES) controller that could predict elbow joint kinematics from the voluntary movements of the shoulder joint. Counter-intuitively, a well-chosen reduced data set for training the network resulted in better performance than use of the whole data set against which the predictions of the network were evaluated.