Advanced Rehabilitative Technology

Neural Interfaces and Devices

Qingsong Ai, Quan Liu, Wei Meng, Sheng Quan Xie

• Publisher: Academic Press
• Year: 2018
• ISBN:
  9780128145982
• Topic: Biological Sciences
• Subtopic: Biotechnology
• Index: Biological Sciences

Advanced Rehabilitative Technology: Neural Interfaces and Devices teaches readers how to acquire and process bio-signals using biosensors and acquisition devices, how to identify the human movement intention and decode the brain signal, how to design physiological and musculoskeletal models and establish the neural interfaces, and how to develop neural devices and control them efficiently using biological signals. The book takes a multidisciplinary theme between the engineering and medical field, including sections on neuromuscular/brain signal processing, human motion and intention recognition, biomechanics modelling and interfaces, and neural devices and control for rehabilitation.

Each chapter goes through a detailed description of the bio-mechatronic systems used and then presents implementation and testing tactics. In addition, it details new neural interfaces and devices, some of which have never been published before in any journals or conferences. With this book, readers will quickly get up-to-speed on the most recent and future advancements in bio-mechatronics engineering for applications in rehabilitation.

• Presents insights into emerging technologies and developments that are currently used or on the horizon in biological systems and mechatronics for rehabilitative purposes
• Gives a comprehensive background of biological interfaces and details of new advances in the field
• Addresses the challenges of rehabilitative applications in areas of bio-signal processing, bio-modelling, neural and muscular interface, and neural devices.
• Provides substantial background materials and relevant case studies for each subject

Chapter 1

Introduction

Abstract

People have speculated that humans could directly control machines by using neural interfaces and devices, for example, by biological signals from human limbs, muscles, or the brain. The neural interfaces would help the aged or disabled control the rehabilitation and assistive devices themselves allowing them to physically engage with the world. Recent developments in neural and muscular interfacing technologies could open a new window that allows the patient's biological system to directly communicate with the world. These developments can potentially help millions of individuals who need rehabilitation or assistance from robotic devices.

Keywords

Rehabilitation training; Rehabilitation robot; Biosignal; Human biological systems; Brain computer interface; Neural Interfaces

Contents

• 1.1. Background
• 1.2. Human Biological Systems
• 1.3. Neural Interfaces and Devices
• 1.4. Critical Issues
• 1.5. Chapter Summary
• References
• Further Reading

1.1 Background

According to data from the United Nations, the proportion of the world's population over 60 will double from 11% to 22% between 2000 and 2050. During the same period, the number of older adults aged over 60 will increase from 605 million to 2 billion. Currently, more than half of elderly people worldwide live in Asia (54%), followed by Europe (22%) [1]. Many countries have gradually entered the aged society. Meanwhile, there are about 650 million people with disabilities worldwide, accounting for about 10% of the world's total population, with 80% of disabled people living in developing countries [2,3]. With the tendency toward an aging society and an increasing disabled population, there will be an obvious recession in patients’ physiological functions, severely affecting their daily lives [4,5].

Rehabilitation training of the elderly, the disabled, and other movement disorders have become a major social problem that needs to be resolved. However, conventional manual therapy mainly relies on the therapist's experience, making it difficult to meet the requirements of high-intensity and repetitive training [6]. There is a considerable increase in the needs of advanced medical and rehabilitation devices, which are expected to precisely, quantitatively, and scientifically assist patients to perform rehabilitation training [7]. Rehabilitation robotics has become a research field that has attracted more and more attention during the last decade [8]. Applying robots to rehabilitation not only can release physicians from the heavy burden of the training mission but also provide a platform to evaluate convalescence results by analyzing the data recorded in training process [9]. Due to the various advantages in terms of accuracy and reliability, rehabilitation robots can provide an efficient approach to improve recovery outcomes after stroke or surgery.

In recent years, novel control strategies, especially bioinspired control, are widely used in rehabilitation robots, which can provide users with more intuitive feelings, that is, they can manipulate the robot with their intentions. With the emerging biosignal acquisition and processing techniques, biofeedback, and hybrid interfaces, devices as well as control strategies have become increasing important in newly developed rehabilitation robots. Some books have mentioned that biosignal-based control strategies had been regarded as effective strategies and a popular research area, however none of them have investigated or summarized recent studies on these neural interfaces and devices [10]. This book will provide a review and analysis of neural signals, interface, and control strategies of bioinspired rehabilitation robots or devices, especially control methods utilizing neuromuscular and brain signals, and their modeling techniques. In this section, we will provide a basic concept of the most recent development of biological signal acquisition and processing technologies in robot-assisted medical rehabilitation, and also summarize research gaps and potential future directions.

1.2 Human Biological Systems

Biological signals contain more useful information about human limb movement than force feedback. Electromyography (EMG) is a bioelectric signal, which is the superposition of the action potentials of many motion units in the muscle in time and space, reflecting the functional state of nerves and muscles [11]. With the development of acquisition and analysis technologies, EMG signals are widely used in numerous fields. For example, EMG signals have been adopted for clinical disease diagnosis. Moreover, EMG is employed to determine muscle physiology, activity, and training; muscle function assessment; and fatigue testing for sports professionals. Also, this promising technique can be used in medical rehabilitation areas such as rehabilitation robot control [12], rehabilitation assessment, prosthetic control [13,14], and so forth.

The traditional method of rehabilitation training is for professional therapists to train patients. However, this method has great demands for manpower and training efficiency. Also, the process cannot be controlled accurately, the solution cannot be effectively improved, and the training is boring. Compared with traditional training methods, rehabilitative robots based on EMG control can use the EMG signal to reflect the characteristics of nerve and muscle function status. Through the acquisition and analysis of an EMG signal, the intention of the human body can be identified to control the movement of the robot, driving the limb for rehabilitation training. This method not only reduces the burden on the therapist, ensures the intensity of training, and facilitates the research and optimization of the rehabilitation program, but it also provides active training according to the “human intention”, which is more conducive to recovery of limb function.

It has been found that a considerable correlation exists between EMG signals, limb movement, and muscle activities. Therefore, with the development of biosignal acquisition and processing techniques in recent years, robot control based on biosignals has become popular. EMG-triggered and continuous control are two typical EMG-based control strategies developed in the last decade; an example can be found in Fig. 1.1. EMG signals are generated before limb muscle contraction, so it can be used to predict the movement intention in advance. Brain computer interface (BCI) systems could offer those with mobility impairments improved communication and independence. This would also reduce the burden of society on the payment to special facilities and disability services. Recent studies indicate that severely paralyzed people, if they have good supportive care and basic communication capacity, may yet enjoy a reasonable quality of life. Indeed, such people are usually affected by the “locked-in” syndrome when they don’t possess muscle control. Therefore, an assistive device operated independently of muscle activity appears to be their only option for communication and control. A BCI is a communication and control system in which the thoughts of the human mind are translated into real-world interactions without the use of the common neural pathways and muscles.

Fig. 1.1 Robot control based on EMG signals and another example using EEG signals. (Reprinted with permission from F. Zhang, P. Li, et al., sEMG-based continuous estimation of joint angles of human legs by using BP neural network, Neurocomputing 78 (1) (2012) 139–148.)

Human electroencephalogram (EEG) signals are spontaneous, and rhythmic electrical activity of the brain cell population is recorded by the electrodes. This phenomenon is accompanied by the end of human life. What we usually call EEG is the scalp EEG. In fact, it is the measured potential difference between time and the relationship between the chart. EEG signals are the overall activity of brain neurons, including the brain cell ion exchange, metabolism, and other comprehensive external manifestations. EEG is nonintrusive, and it is most secure in technology and relatively low cost, so it is most widely used in the research of brain-machine interface.

Motor imagery EEG [15] is evoked by subjective consciousness, which belongs to endogenous evoked response. It reflects the dynamic process of subjective thinking from form to execution. Relevant research in sports rehabilitation shows that motor imagery training can promote the rehabilitation of injured nerves and the reconstruction of motor nerve pathways. Therefore, it is of great significance to study the processing and application of motor imagery EEG signals.

There are primarily three kinds of rehabilitation methods based on BCI technology. First, the BCI system is directly used to communicate with the outside world, such as control of nerve prostheses [16], electric wheelchairs [17], and character input [18]. Second, for som...

Table of contents

• Cover image
• Title page
• Table of Contents
• Copyright
• Author Biography
• Preface
• Chapter 1: Introduction
• Chapter 2: State-of-the-Art
• Chapter 3: Neuromuscular Signal Acquisition and Processing
• Chapter 4: sEMG-Based Motion Recognition
• Chapter 5: Brain Signal Acquisition and Preprocessing
• Chapter 6: EEG-Based Brain Intention Recognition
• Chapter 7: Neuromuscular Modeling
• Chapter 8: Neural Interface
• Chapter 9: Conclusion and Future Prospects
• Nomenclatures
• Index