| 1. |
- Khong, L.M.D., et al.
(författare)
-
Multi-layer perceptron training algorithms for pattern recognition of myoelectric signals
- 2013
-
Ingår i: BMEiCON 2013 - 6th Biomedical Engineering International Conference. - 9781479914678
-
Konferensbidrag (refereegranskat)abstract
- A challenge in using myoelectric signals in control of motorised prostheses is achieving effective signal pattern recognition and robust classification of intended motions. In this paper, the performance of Matlab's Multi-layer Perceptron (MLP) backpropogation training algorithms in motion classification were assessed. The test and evaluation platform used was 'BioPatRec', a Matlab-based open-source prosthetic control development environment, together with algorithms sourced from Matlab's neural network toolbox. The algorithms were used to interpret multielectrode myoelectric signals for motion classification, with the aim of finding the best performing algorithm and network model. The results showed that Matlab's trainlm and trainrp algorithms could achieve a higher accuracy than other tested MLP training algorithms (94.13 ± 0.037% and 91.09 ± 0.047%, respectively). Discussion of these results investigates significant features to obtain the highest performance.
|
|
| 2. |
|
|
| 3. |
- Ortiz Catalan, Max Jair, 1982, et al.
(författare)
-
An osseointegrated human-machine gateway for long-term sensory feedback and motor control of artificial limbs
- 2014
-
Ingår i: Science Translational Medicine. - : American Association for the Advancement of Science (AAAS). - 1946-6234 .- 1946-6242. ; 6:257
-
Tidskriftsartikel (refereegranskat)abstract
- A major challenge since the invention of implantable devices has been a reliable and long-term stable transcutaneous communication. In the case of prosthetic limbs, existing neuromuscular interfaces have been unable to address this challenge and provide direct and intuitive neural control. Although prosthetic hardware and decoding algorithms are readily available, there is still a lack of appropriate and stable physiological signals for controlling the devices. We developed a percutaneous osseointegrated (bone-anchored) interface that allows for permanent and unlimited bidirectional communication with the human body. With this interface, an artificial limb can be chronically driven by implanted electrodes in the peripheral nerves and muscles of an amputee, outside of controlled environments and during activities of daily living, thus reducing disability and improving quality of life. We demonstrate in one subject, for more than 1 year, that implanted electrodes provide a more precise and reliable control than surface electrodes, regardless of limb position and environmental conditions, and with less effort. Furthermore, long-term stable myoelectric pattern recognition and appropriate sensory feedback elicited via neurostimulation was demonstrated. The opportunity to chronically record and stimulate the neuromuscular system allows for the implementation of intuitive control and naturally perceived sensory feedback, as well as opportunities for the prediction of complex limb motions and better understanding of sensory perception. The permanent bidirectional interface presented here is a critical step toward more natural limb replacement, by combining stable attachment with permanent and reliable human-machine communication.
|
|
| 4. |
- Ortiz Catalan, Max Jair, 1982, et al.
(författare)
-
Biologically inspired algorithms applied to prosthetic control
- 2012
-
Ingår i: BioMed 2012 , February 15 – 17, 2012, Innsbruck, Austria. - 9780889869097 ; :track 764
-
Konferensbidrag (refereegranskat)abstract
- Biologically inspired algorithms were used in this work to approach different components of pattern recognition applied to the control of robotic prosthetics. In order to contribute with a different training paradigm, Evolutionary (EA) and Particle Swarm Optimization (PSO) algorithms were used to train an Artificial Neural Network (ANN). Since the optimal input set of signal features is yet unknown, a Genetic Algorithm (GA) was used to approach this problem. The training length and rate of convergence were considered in the search of an optimal set of signal features, as well as for the optimal time window length. The ANN proved to be an accurate pattern recognition algorithm predicting 10 movements with over 95% accuracy. Moreover, new combinations of signal features with higher convergence rates than the commonly found in the literature were discovered by the GA. It was also found that the PSO had better performance that the EA as a training algorithm but worse than the well established Back-propagation. The latter considered accuracy, training length and convergence. Finally, the common practice of using 200 ms time window was found to be sufficient for producing acceptable accuracies while remaining short enough for a real-time control.
|
|
| 5. |
|
|
| 6. |
- Ortiz Catalan, Max Jair, 1982, et al.
(författare)
-
BioPatRec: A modular research platform for the control of artificial limbs based on pattern recognition algorithms
- 2013
-
Ingår i: Source Code for Biology and Medicine. - : Springer Science and Business Media LLC. - 1751-0473. ; 8
-
Tidskriftsartikel (refereegranskat)abstract
- Background Processing and pattern recognition of myoelectric signals have been at the core of prosthetic control research in the last decade. Although most studies agree on reporting the accuracy of predicting predefined movements, there is a significant amount of study-dependent variables that make high-resolution inter-study comparison practically impossible. As an effort to provide a common research platform for the development and evaluation of algorithms in prosthetic control, we introduce BioPatRec as open source software. BioPatRec allows a seamless implementation of a variety of algorithms in the fields of 1) Signal processing; 2) Feature selection and extraction; 3) Pattern recognition; and, 4) Real-time control. Furthermore, since the platform is highly modular and customizable, researchers from different fields can seamlessly benchmark their algorithms by applying them in prosthetic control, without necessarily knowing how to obtain and process bioelectric signals, or how to produce and evaluate physically meaningful outputs. Results BioPatRec is demonstrated in this study by the implementation of a relatively new pattern recognition algorithm, namely Regulatory Feedback Networks (RFN). RFN produced comparable results to those of more sophisticated classifiers such as Linear Discriminant Analysis and Multi-Layer Perceptron. BioPatRec is released with these 3 fundamentally different classifiers, as well as all the necessary routines for the myoelectric control of a virtual hand; from data acquisition to real-time evaluations. All the required instructions for use and development are provided in the online project hosting platform, which includes issue tracking and an extensive "wiki". This transparent implementation aims to facilitate collaboration and speed up utilization. Moreover, BioPatRec provides a publicly available repository of myoelectric signals that allow algorithms benchmarking on common data sets. This is particularly useful for researchers lacking of data acquisition hardware, or with limited access to patients. Conclusions BioPatRec has been made openly and freely available with the hope to accelerate, through the community contributions, the development of better algorithms that can potentially improve the patient's quality of life. It is currently used in 3 different continents and by researchers of different disciplines, thus proving to be a useful tool for development and collaboration.
|
|
| 7. |
- Ortiz Catalan, Max Jair, 1982, et al.
(författare)
-
Effect on signal-to-noise ratio of splitting the continuous contacts of cuff electrodes into smaller recording areas
- 2013
-
Ingår i: Journal of NeuroEngineering and Rehabilitation. - : Springer Science and Business Media LLC. - 1743-0003. ; 10:1, s. 22-
-
Tidskriftsartikel (refereegranskat)abstract
- ABSTRACT: BACKGROUND: Cuff electrodes have been widely used chronically in different clinical applications. This neural interface has been dominantly used for nerve stimulation while interfering noise is the major issue when employed for recording purposes. Advancements have been made in rejecting extra-neural interference by using continuous ring contacts in tripolar topologies. Ring contacts provide an average of the neural activity, and thus reduce the information retrieved. Splitting these contacts into smaller recording areas could potentially increase the information content. In this study, we investigate the impact of such discretization on the Signal-to-Noise Ratio (SNR). The effect of contacts positioning and an additional short circuited pair of electrodes were also addressed. METHODS: Different recording configurations using ring, dot, and a mixed of both contacts were studied in vitro in a frog model. An interfering signal was induced in the medium to simulate myoelectric noise. The experimental setup was design in such a way that the only difference between recordings was the configuration used. The inter-session experimental differences were taking care of by a common configuration that allowed normalization between electrode designs. RESULTS: It was found that splitting all contacts into small recording areas had negative effects on noise rejection. However, if this is only applied to the central contact creating a mixed tripole configuration, a considerable and statistically significant improvement was observed. Moreover, the signal to noise ratio was equal or larger than what can be achieved with the best known configuration, namely the short circuited tripole. This suggests that for recording purposes, any tripole topology would benefit from splitting the central contact into one or more discrete contacts. CONCLUSIONS: Our results showed that a mixed tripole configuration performs better than the configuration including only ring contacts. Therefore, splitting the central ring contact of a cuff electrode into a number of dot contacts not only provides additional information but also an improved SNR. In addition, the effect of an additional pair of short circuited electrodes and the "end effect" observed with the presented method are in line with previous findings by other authors.
|
|
| 8. |
- Ortiz Catalan, Max Jair, 1982, et al.
(författare)
-
Evaluation of Classifier Topologies for the Real-time Classification of Simultaneous Limb motions
- 2013
-
Ingår i: Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS. - 1557-170X. - 9781457702167 ; , s. 6651-6654
-
Konferensbidrag (refereegranskat)abstract
- The prediction of motion intent through the decoding of myoelectric signals has the potential to improve the functionally of limb prostheses. Considerable research on individual motion classifiers has been done to exploit this idea. A drawback with the individual prediction approach, however, is its limitation to serial control, which is slow, cumbersome, and unnatural. In this work, different classifier topologies suitable for the decoding of mixed classes, and thus capable of predicting simultaneous motions, were investigated in real-time. These topologies resulted in higher offline accuracies than previously achieved, but more importantly, positive indications of their suitability for real-time systems were found. Furthermore, in order to facilitate further development, benchmarking, and cooperation, the algorithms and data generated in this study are freely available as part of BioPatRec, an open source framework for the development of advanced prosthetic control strategies.
|
|
| 9. |
- Ortiz Catalan, Max Jair, 1982, et al.
(författare)
-
On the viability of implantable electrodes for the natural control of artificial limbs: Review and discussion
- 2012
-
Ingår i: BioMedical Engineering OnLine. - 1475-925X. ; 11:33
-
Tidskriftsartikel (refereegranskat)abstract
- The control of robotic prostheses based on pattern recognition algorithms is a widely studied subject that has shown promising results in acute experiments. The long-term implementation of this technology, however, has not yet been achieved due to practical issues that can be mainly attributed to the use of surface electrodes and their highly environmental dependency. This paper describes several implantable electrodes and discusses them as a solution for the natural control of artificial limbs. In this context "natural" is defined as producing control over limb movement analogous to that of an intact physiological system. This includes coordinated and simultaneous movements of different degrees of freedom. It also implies that the input signals must come from nerves or muscles that were originally meant to produce the intended movement and that feedback is perceived as originating in the missing limb without requiring burdensome levels of concentration. After scrutinizing different electrode designs and their clinical implementation, we concluded that the epimysial and cuff electrodes are currently promising candidates to achieving a long-term stable and natural control of robotic prosthetics, provided that communication from the electrodes to the outside of the body is guaranteed.
|
|
| 10. |
- Ortiz Catalan, Max Jair, 1982, et al.
(författare)
-
Real-Time and Simultaneous Control of Artificial Limbs Based on Pattern Recognition Algorithms
- 2014
-
Ingår i: IEEE Transactions on Neural Systems and Rehabilitation Engineering. - 1558-0210 .- 1534-4320. ; 22:4, s. 756-764
-
Tidskriftsartikel (refereegranskat)abstract
- The prediction of simultaneous limb motions is a highly desirable feature for the control of artificial limbs. In this work, we investigated different classification strategies for individual and simultaneous movements based on pattern recognition of myoelectric signals. Our results suggest that any classifier can be potentially employed in the prediction of simultaneous movements if arranged in a distributed topology. On the other hand, classifiers inherently capable of simultaneous predictions, such as the Multi-Layer Perceptron (MLP), were found to be more cost effective, as they can be successfully employed in their simplest form. In the prediction of individual movements, the One-Vs-One (OVO) topology was found to improve classification accuracy across different classifiers and it was therefore used to benchmark the benefits of simultaneous control. As opposed to previous work reporting only offline accuracy, the classification performance and the resulting controllability were evaluated in real-time using the Motion Test and Target Achievement Control (TAC) Test respectively. We propose a simultaneous classification strategy based on MLP that outperformed a top classifier for individual movements (LDA-OVO), thus improving the state-of-the-art classification approach. Furthermore, all the presented classification strategies and data collected in this study are freely available in BioPatRec, an open source platform for the development of advanced prosthetic control strategies.
|
|
