Dejan B. Popovic

Sensory nerve recording for closed-loop control to restore motor functions

A method is developed for using neural recordings to control functional electrical stimulation (FES) to nerves and muscles. Experiments were done in chronic cats with a goal of designing a rule-based controller to generate rhythmic movements of the ankle joint during treadmill locomotion. Neural signals from the tibial and superficial peroneal nerves were recorded with cuff electrodes and processed simultaneously with muscular signals from ankle flexors and extensors in the cats hind limb. Cuff electrodes are an effective method for long-term chronic recording in peripheral nerves without causing discomfort or damage to the nerve. For real-time operation the authors designed a low-noise amplifier with a blanking circuit to minimize stimulation artifacts. They used threshold detection to design a simple rule-based control and compared its output to the pattern determined using adaptive neural networks. Both the threshold detection and adaptive networks are robust enough to accommodate the variability in neural recordings. The adaptive logic network used for this study is effective in mapping transfer functions and therefore applicable for determination of gait invariants to be used for closed loop control in an FES system. Simple rule-bases will probably be chosen for initial applications to human patients. However, more complex FES applications require more complex rule-bases and better mapping of continuous neural recordings and muscular activity. Adaptive neural networks have promise for these more complex applications.<<ETX>>

Machine learning in control of functional electrical stimulation systems for locomotion

Two machine learning techniques were evaluated for automatic design of a rule-based control of functional electrical stimulation (FES) for locomotion of spinal cord injured humans. The task was to learn the invariant characteristics of the relationship between sensory information and the FES-control signal by using off-line supervised training. Sensory signals were recorded using pressure sensors installed in the insoles of a subjects shoes and goniometers attached across the joints of the affected leg. The FES-control consisted of pulses corresponding to time intervals when the subject pressed on the manual push-button to deliver the stimulation during FES-assisted ambulation. The machine learning techniques used were the adaptive logic network (ALN) and the inductive learning algorithm (IL). Results to date suggest that, given the same training data, the IL learned faster than the ALN while both performed the test rapidly. The generalization was estimated by measuring the test errors and it was better with an ALN, especially if past points were used to reflect the time dimension. Both techniques were able to predict future stimulation events. An advantage of the ALN over the IL was that ALNs can be retrained with new data without losing previously collected knowledge. The advantages of the IL over the ALN were that the IL produces small, explicit, comprehensible trees and that the relative importance of each sensory contribution can be quantified.<<ETX>>

Nonanalytical methods for motor control

This text describes non-conventional methods of control of human extremities, emphasizing the fact that conventional approaches used in robotics are limited when used in humans for restoration of reaching and grasping (goal-oriented movements), standing and locomotion (cyclic movements). The use of artificial neural networks, inductive learning, skill-based expert systems and finite-state representation of movements is the base of this non-conventional control theory. A specific number of realized applications are included in the book to illustrate how these computer techniques can improve the function of assistive systems in physically challenged humans. The theory presented is applicable to the control of robots and industrial manipulators.

Properties of implanted electrodes for functional electrical stimulation

Implanted wire electrodes are increasingly being used for the functional electrical stimulation of muscles in partially paralysed patients, yet many of their basic characteristics are poorly understood. In this study we investigated the selectivity, recruitment characteristics and range of control of several types of electrode in triceps surae and plantaris muscles of anaesthetized cats. We found that nerve cuffs are more efficient and selective (i.e., cause less stimulus spread to surrounding muscles) than intramuscular electrodes. Bipolar intramuscular stimulation was more efficient and selective than monopolar stimulation, but only if the nerve entry point was between the electrodes. Monopolar electrodes are efficient and selective if located close to the nerve entry point, but their performance declines with distance from it. Nonetheless, for a variety of reasons monopolar stimulation provides the best compromise in many current applications. Short duration pulses offer the best efficiency (least charge per pulse to elicit force) but high peak currents, increasing the risk of electrode corrosion and tissue damage. Electrode size has little effect on recruitment and should therefore be maximised because this minimises current density.

Optimal control for the active above-knee prosthesis

Control of an active above-knee prosthesis has been simulated for a selected gait activity using a hierarchical closed-loop method. An extension of finite-state control, referred to as artificial reflex control, was adopted at the strategic level of control. At the actuator level of control an optimal tracking method, based on dynamic programming, is applied. This deals mainly with the actuator level of control, but considers the interaction of the leg dynamics and the switching effects of artificial reflex control. Optimal tracking at the actuator level of the above-knee prosthesis reduces the on-off effects of finite-state methods, such as artificial reflex control. The proposed method can also be used for the design of prosthetic elements. Specific attention is paid to the limited torque and power in the prosthetic joint actuator, which are imposed by the principle of self-containment in the artificial leg. The hierarchical structure, integrating artificial reflex control and optimal tracking, can be used in real time, as estimated from the number of computer operations required for the suggested method.

Chapter 35 Finite state model of locomotion for functional electrical stimulation systems

A finite state model of locomotion was developed to simplify a controller design for motor activities of handicapped humans. This paper presents a model developed for real time control of locomotion with functional electrical stimulation (FES) assistive systems. Hierarchical control of locomotion was adopted with three levels: voluntary, coordination and actuator level. This paper deals only with coordination level of control. In our previous studies we demonstrated that a skill-based expert system can be used for coordination level of control in multi-joint FES systems. Basic elements in this skill-based expert system are production rules. Production rules have the form of If-Then conditional expressions. A technique of automatic determination of these conditional expressions is presented in this paper. This technique for automatic synthesis of production rules uses fuzzy logic and artificial neural networks (ANN). The special class of fuzzy logic elements used in this research is called preferential neurons. The preferential neurons were used to estimate the relevance of each of the sensory inputs to the recognition of patterns defined as finite states. The combination of preferential neurons forms a preferential neural network. The preferential neural network belongs to a class of ANNs. The preferential neural network determined the set of finite states convenient for a skill-based expert system for different modalities of locomotion.

Improved Methods for Control of FES for Locomotion

Abstract Two methods of generating the rules for control systems for control of functional electrical stimulation (FES) in locomotion of subjects with incomplete spinal cord injury were studied and qualitatively compared. The aim of the rule-based control system was to synthesize and replace the manual FES-switching function operated by the subject or a skilled physiotherapist. The first method, “hand-crafting” of rules, requires very detailed signal analysis by an experienced person, which may result in systems that are limited either by their functionality or safety for the subject. The second method, i.e. automatic generation of rules through learning from examples by a machine learning program, does not have such limitations because it is capable of learning everything that is presented to it during the training phase. It is obvious that for simple control systems, “hand-crafted” rules are a fast and simple solution, but for complex multichannel FES applications automatic generation of the rules can save many hours of trial-and-error experiments with hand-crafted rules.

Learning of EMG-patterns by adaptive logic networks

An adaptive logic network (ALN) has been used for nonparametric identification of the system consisting of two peripheral nerves and two muscles in the freely moving cat. The aim of this identification was to design a rule-based control for a functional electrical stimulation (FES) of the cats hind limb. We recorded from peripheral nerves and muscles while a chronic cat walked an a powered treadmill. We noticed a very reproducible firing in tibial and superficial peroneal nerves related to patterns of EMG activity in medial gastrocnemius (MC) and anterior tibialis (AT) muscles in the cats hind limb with respect to a phase of the gait cycle (e.g. beginning stance, beginning swing). We succeeded to restore and predict EMG activity of MG and AT muscles from neural recordings after applying Adaptive Logic Networks (ALNs). ALN is a type of artificial neural network having a tree configuration and using a Boolean operations in its nodes. We supplied a training set consisting of coded recordings from nerves and muscles (20 cats steps were sampled at 50 Hz) for supervised training. Applied encoding and training parameters resulted in a reasonably short training time and high correlation between an originai and predicted EMG signals. The variance accounted for (VAF) by the ALN prediction of the test data was 80% qualify- ing this technique as a good candidate for implementation in real-time FES control systems.

Integrated control system for FES-assisted locomotion after spinal cord injury

The objective of this study was to develop an integrated control system (ICS) for FES-assisted locomotion after incomplete spinal cord injury (SCI). The ICS incorporates a method for automatic generation of control rules for rule-based control. The rules are extracted from a set of sensory feedback signals and stimulation control signals recorded during FES-assisted walking controlled by a skilled therapist or the subject. The rule-generation method uses Adaptive Logic Networks (ALNs), a type of artificial neural network. The ICS provides a very efficient tool to acquire sensory and control signals, to process these signals, to train the ALNs in mapping the control function, to test the trained ALNs, and to use them for control signal generation in real-time control of the FES-assisted walking. Through experimental work its been demonstrated that ALNs are able to generate control rules quickly and to generalize not only over daily subsequent walking sessions but also over the sessions occurring several days after the training, which provides a good basis for design of robust control systems for FES-assisted walking. Evaluation of new subjects and automatic generation of control rules using ICS is possible within minutes compared to classic hand-crafting methods which usually require weeks.

Machine learning in control of functional electrical stimulation for locomotion

Two machine learning techniques were evaluated for automatic design of a rule-based control of functional electrical stimulation (FES) for locomotion of spinal cord injured humans. The task was to map the relationship between sensory information and the FES-control signal by using off-line supervised training. Signals were recorded using pressure sensors installed in insoles of a patients shoes and goniometers attached across the joints of the affected leg. The FES-control signal consisted of pulses corresponding to time intervals when the patient pressed on the manual push-button to deliver the stimulation during FES-assisted ambulation. The machine learning techniques evaluated were the adaptive logic network (ALN) and inductive learning algorithm (IL). Results to date suggest that, given the same training data, the IL learned faster than ALN while both performed the test rapidly. The generalization was better with an ALN, especially if past points were used to reflect the time dimension. Both techniques were able to predict future stimulation events. An advantage of ALN was that it can be retrained with new data without losing previously collected knowledge. The advantages of IL were that IL produces explicit and comprehensible trees and that the relative importance of each sensory contribution can be quantified.<<ETX>>