BỘ GIÁO DỤC VÀ ĐÀO TẠO TRƯỜNG ĐẠI HỌC SƯ PHẠM KỸ THUẬT THÀNH PHỐ HỒ CHÍ MINH ĐỒ ÁN TỐT NGHIỆP DEPARTMENT OF MECHATRONICS TECHNOLOGY DESIGN AND FABRICATE EMG ANALYSIS SYSTEM GVHD: Dr. BUI HA DUC SVTH : TRAN HUU TRONG MSSV: 14146292 SVTH : NGUYEN MANH CUONG MSSV: 14146026 SVTH : NGUYEN HONG PHUC MSSV: 14146271 SKL 0 0 5 2 4 7 Tp. Hồ Chí Minh, tháng 07/2018 do an HCMC UNIVERSITY OF TECHNOLOGY AND EDUCATION FACULTY OF QUALITY TRAINING DEPARTMENT OF MECHATRONICS TECHNOLOGY GRADUATION THESIS DESIGN AND FABRICATE EMG ANALYSIS SYSTEM INSTRUCTOR: Dr. BUI HA DUC STUDENT’S NAME: TRAN HUU TRONG 14146292 NGUYEN MANH CUONG 14146026 NGUYEN HONG PHUC 14146271 MAJOR: MECHATRONICS Ho Chi Minh City, July 2018 1 do an ABSTRACT EMG is a bioelectrical signal produced by muscle activity.
EMG has a lot of advantages for applying in control field. Nowadays, there are many research using the EMG signal to control the robot arm and peripheral devices. However, acquisition of EMG signal have some challenges… In this research, our group focus on designing and fabricating the EMG acquisition system in real time. EMG acquisition system circuit including ADS1293 communicate with Raspberry 3 via SPI protocol.
This circuit using electrodes for collecting signal from surface skin. The collected signal is processed, analyzed on acquisition system circuit and displayed on the computer. During the research, our group successfully fabricate the EMG acquisition system with low cost and compact size. The signal is collected accurately which represent the full characteristic of muscle activity, the results of this research show the EMG acquisition system could be response particularly the acquisition, processing and analysis.
This project opens opportunity to the development and application of EMG signals in the technical control field. It can become one of the most widely used scientific fields in the future. 2 do an ACKNOWLEDGEMENT We sincerely thank the lecturers of the department of mechatronics of high quality training who have continuously encouraged us and also provided precious knowledge. Without it, we could not complete our project.
We would like to express my special thanks to Dr.Bui Ha Duc who supported us a lot for doing this study and we explored so many new things. The group is looking forward to receiving feedback, instructions of lecturers to revise and improve the project better. My group sincerely thanks! Ho Chi Minh City, July 2018 Group of students: Tran Huu Trong Nguyen Manh Cuong Nguyen Hong Phuc 3 do an Contents List of figures. 6 List of Tables.
9 List of Abbreviations. Scope of research. 12 Chapter II: LITARETURE REVIEW. Theory of EMG.
Surface EMG electrodes placement. Muscle map frontal. Forearm electrode placement. EMG signal analysis.
Amplitude and frequency of EMG Signal. Theory of Neural Networks. Overview of previous research. 31 Chapter III: DESIGN AND FABRICATE EMG ACQUISITION SYSTEM 35 3.
Method 1: Using OPAMP. Amplifiers and filters. Amplification and filtering circuitry. Collect EMG signal experiment in practical.
Method 2: Using ADS1293 and Raspberry Pi 3. Data acquisition with ADS1293. 55 The EMG signal is collected using surface EMG electrodes. These signals are processed by a circuit including ADS1293 and Raspberry Pi 3, data from processing via SPI protocol is transferred to computer via wireless network for program and control.
Design acquisition system circuit. The EMG acquisition system circuit. 69 Chapter IV: DESIGN PROTOCOL EXPIREMENT. Interface between ADS1293 with Raspberry Pi 3.
SPI communication between the ADS1293 with RPi3. Raspberry Pi 3 interface to WEB .js to transmit data to Pusher .js min to draw plot of dataPoint on browser. 82 Chapter VI: RESULT AND ANALYSIS. 90 Chapter VII: CONCLUSION.
94 5 do an List of figures Figure 1. Medical benefits of EMG. Schematic illustration of depolarization/repolarization .The action potential. The "Raw" EMG Signal.
Electrode leads with cable built-in preamplifiers. The effect of A/D sampling frequency. Diagram of detecting EMG signal into MUAPTs. Muscle Map Frontal.
Flexor muscle of anterior forearm. Extensor muscle of posterior forearm. Electrode orientation for EMG recording. Electrode Placement for wrist movement.
Block Diagram of EMG analysis system. a) Biological neural networks. Layers of Neural networks. Myo gesture control armband.
e-Health Sensor Platform. Schematic and block diagram of an ideal amplifier with gain of G. An amplifier with different input configuration. A differential amplifier configured as an inverting and non- inverting amplifier.
Idealized Filter Responses. Key Filter Parameters. Block diagram of amplification. Structure of INA114.
42 6 do an Figure 33. Structure of OP07CP. The 1st Amplifier circuit. Low-pass filter circuit.
High-pass filter circuit. 45 Figure 37: The Second Amplifier circuit. The 2nd Amplifier circuit. Table of components on Altium Designer.
Printed circuit board. Image 3D of PCB. Top face of the EMG acquisition system circuit. Simulate the low-pass filter circuit.
Low-pass filter signal at 400Hz. Low-pass filter signal at 1000Hz. Simulate the high-pass filter circuit. High-pass filter signal at 400Hz.
High-pass filter signal at 4Hz. Simulate the 2nd amplification circuit. Signal at 400Hz. Signal at 1000Hz.
Block diagram of EMG Acquisition System. EMG Acquisition system. Medical cable 3 snap electrode lead wire and electrode of 3M RedDot. Electrode placement on the forearm.
Image of ADS1293. Function Block Diagram. The Flexible Routing Switch for single channel. Top View 28-Pin of ADS1293.
Functional Block of Raspberry Pi. ADS 1293 communicate with RPI3 via SPI protocol. 69 7 do an Figure 70. The acquisition system using ADS1293 and RPi3.
SPI four-wire serial bus. Synchronize clock line and data line. SPI bus: Single master and single slave. An 8 bit shift register is contain for each SPI device.
SCK is defined by CPOL. Data on the MOSI and MISO is defined by CPHA. SPI Protocol on ADS1293. Define of Transfer function.
Functions Read Register and Write Register. Function to stream data from ADS1293 to Rapberry Pi 3. Syntax to open file result.csv to write data out. Syntax to write adc_data received into file opened below.
Function of Notch filter 50Hz in file. Image of baseline in Oscilloscope. Image of baseline with ODR = 833Hz was plotted in MATLAB. Image of Sinwave in Oscilloscope.
Image of Sinwave was plotted in MATLAB with ODR = 833Hz 84 Figure 93. Image of Sinwave was plotted in MATLAB with ODR = 1580Hz. Image of Sinwave in Oscilloscope. Image of Sinwave was plotted in MATLAB with ODR = 1580Hz.
Image of Sinwave was plotted in MATLAB with ODR = 833Hz86 Figure 97. Image of Squarewave in Oscilloscope. Image of Squarewave was plotted in MATLAB with 833Hz. Image of Squarewave was plotted in MATLAB with 1580Hz.
Image of Squarewave in Oscilloscope .Image of Squarewave was plotted in MATLAB with 1580Hz. Image of Squarewave was plotted in MATLAB with 833Hz. Image of ECG was plotted in MATLAB. 90 8 do an Figure 105.
Image of EMG signal was plotted in MATLAB at 1st and 2nd sec. Image of EMG signal was plotted in MATLAB at 7th sec. 91 List of Tables TABLE 1. OF REVELANT MUSCLE, ACTION AND NERVE SUPPLY.
SPI FOUR DIFFERENT MODES. 73 9 do an List of Abbreviations EMG - Electromyogram SEMG – Surface Electromyogram MUAP - Motor Unit Action Potential MUAPT - Motor Unit Action Potential Trains CMRR - Common Mode Rejection Ratio ADC - Analog to Digital Converter ECG - Electrocardiogram EMI - Electromagnetic Interference INA - Instrumentation Amplifier SDM - Sigma-Delta Modulator PCB - Printed circuit board RPi3 - Raspberry Pi 3 SPI - Serial peripheral interface SCLK, SCK - Serial Clock MOSI - Master Output Slave Input MISO - Master Input Slave Output SS - Slave Select CS - Chip Select CSB - Chip Select Bar DRDYB - Data Ready Bar CPOL - Clock polarity CPHA - Clock phase HTTP - Hypertext Transfer Protocol FTP - File Transfer Protocol SMTP - Simple Mail Transfer Protocol POP - Post Office Protocol URL - Uniform Resource Locator JS - Java Script CSV - Comma Separated Values DPS - Data Points 10 do an Chapter I: INTRODUCTION 1. Motivation Electromyogram (EMG) is bioelectrical signal produced by motor control neuron that transmit signal for active muscle cells, this electric affected by level of muscle activities. This signal appears before moving of muscle in about 150ms.
That is a big advantage. If the EMG signal is utilized effectively, it could significantly enhance the robustness of control system. There were several groups have researched and applied EMG signal into control some devices. For example, Myo Gesture Control Armband is researched and manufactured by Thalmic Labs.
The EMG signal can be applied to some devices such as a computer game, digital pointer controller for zoom in on your slides and touch-free. However, acquisition of EMG signal have some challenges. First, cost of parts and devices is very expensive. Second, it is not easily to find in Vietnam.
Usually, we must order this from abroad. Third, some devices are big or oversized. So, it is unsuitable for portable system. Four, the collected signal is easily noised by external factors.
This require the acquisition system have to perfect and work correctly. And finally, some parts is unsuitable for technical control. To overcome these challenge, our team aims to design a compact, reliable device for acquiring and analyzing EMG signal. This devices will be affordable, compactly, easy to use and has capability to work in the real-time.
In this project, our team will use embedded circuit Raspberry Pi 3 to control an Analog Front End for Biopotential Measurements ADS1293. Interface SPI is used for transmit and receive data between Raspberry and ADS1293. Additions, we use wireless connection to push the collected data onto Cloud and draw plot real-time on Web. Project objectives - Determine technical requirements of EMG signal acquisition system.
- Design and fabricate EMG signal acquisition system circuit. - Collect and analyze the EMG signal. - Transmit and store the signal from Raspberry to the computer via wireless. 11 do an - Apply the signal to control software on the computer.
Scope of research The scope of this project is focus on designing and fabricating the hardware. Then, we apply the EMG signal which is collected to process and analyze in EMG acquisition system. We use the filters which integrated in system to eliminate the unwanted noise. Next, we process data and display on Web by plot.
We have already demoed on the presenter application. Thesis organization This thesis will have 6 chapters. In chapter 1, we will present the EMG, the generation of EMG signal, processing the signal and factors affecting the EMG signal. In chapter 2, we will learn about the EMG, the generation of EMG and EMG signal.
EMG has many benefits and applications and we can apply it for technical control, … In chapter 3, we use the ADS1293 and the Raspberry Pi 3 to fabricate the EMG acquisition system circuit. This circuit collected signal by ADS1293 and RPI3 via SPI protocol and send it to computer via wireless for program. In chapter 4, we will do some experiment for collecting the EMG signal from the forearm. This chapter provide us the characteristic of the EMG signal In chapter 5, will have the experiment results which have done in chapter 4.
The results on the circuit show quality of collected EMG signal. We will finger out how to collect EMG signal without ambient noise In chapter 6, will show research results and research directions. 12 do an Chapter II: LITARETURE REVIEW 1. Theory of EMG 1.
Definition Electromyography (EMG) [1] is an experimental technique concerned with the development, recording and analysis of myoelectric signals.