Research on reading muscle signals from the EMG sensor during knee flexion — Extension using the Arduino Uno controller

Dao, Duc Mịnh; Phuoc, Pham Dang; Tuy, Tran Xuan; Le, Tram Thuy

doi:10.1109/atc.2017.8167632

Cited by 2 publications

(1 citation statement)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this study Arduino that was interfaced with the output of a muscle sensor v3 to transformed analog-to-digital signal [21]. The signal was then transferred to excel and been uploaded to MATLAB for extraction and classification [22]. [23].…”

Section: Fig 2 Muscle Sensor V3mentioning

confidence: 99%

Machine Learning Technique to Classify EMG Signal for Diabetes Person

Zulkifli,

Nasir

2023

ijmst

View full text Add to dashboard Cite

Diabetes can cause a disease known as diabetic peripheral neuropathy (DPN), which affects the blood vessels and nerves in the legs and feet. This condition can result in plantar foot ulcers and muscular weakness. Detecting DPN early stage is crucial so patients can receive early treatment before their disease worsens. Most technology that detects this disease is usually expensive, like an Electromyography machine (EMG). But, with the increasing popularity of machine learning classification in the health sciences, DPN can be identified early by producing a low-cost equipment. This study aimed to develop a low-cost surface EMG (sEMG) system to detect electrical activity in the lower limb muscles and classify healthy and diabetic subjects during muscle fatigue using K-Nearest Neighbors (KNN) and Artificial Neural Networks (ANN) as two methods of machine learning technique. This study used Muscle Sensor V3 as sEMG to record the signal and extract using time domain feature extraction before classification. In KNN, 1–10 values (K) are used, while in ANN, 1–10 values of the number of hidden neurons are used to compare the classification performance. The result shows that ANN is suitable compared to KNN with the method used in this study. ANN algorithm performs better using four hidden neurons with an accuracy of 100% for the training and testing process. The study can conclude that this low-cost sEMG system, with the help of ANN, can effectively classify two subjects (healthy and diabetic) according to the EMG data obtained. This system can help to identify any diabetes neuropathy at an early stage and able to prevent lower limb complications related to the disease.

show abstract

Section: Fig 2 Muscle Sensor V3mentioning

confidence: 99%

Machine Learning Technique to Classify EMG Signal for Diabetes Person

Zulkifli,

Nasir

2023

ijmst

View full text Add to dashboard Cite

show abstract

Improved Multichannel Electromyograph Using Off-the-Shelf Components for Education and Research

Staderini¹,

Mugnaini

Kambampati

et al. 2022

Sensors

View full text Add to dashboard Cite

Most students and researchers with limited funding are often looking for simple and low-cost devices for the acquisition of the electromyogram signal (EMG) in an educational or research setting. Thus, off-the-shelf devices are used and they have already been described in the literature, but they are used without considering their real performances, which are, in general, quite poor from the electronic and signal processing points of view. It is the purpose of this communication to present the evidence of these issues, and to describe an improved version of the “classical” duo, composed of the common ECG/EMG Olimex board and the Arduino microprocessor board. In this case, the Arduino-DUE is used. Three main points are highlighted in this paper: (a) the bandpass characteristics of the ECG/EMG Olimex board and how they can be improved to cope with EMG bandwidth requirements; (b) the increase in sampling frequency of the signal; and, finally, (c) the possibility of automatic detection of more ECG/EMG Olimex boards installed at the same time as the shields on the Arduino-DUE board. Very simple and low-cost modifications on the ECG/EMG Olimex board could deliver a much better performing multichannel EMG acquisition system, suitable for educational classroom experiments and early proof-of-concept research.

show abstract

Research on reading muscle signals from the EMG sensor during knee flexion — Extension using the Arduino Uno controller

Cited by 2 publications

References 6 publications

Machine Learning Technique to Classify EMG Signal for Diabetes Person

Machine Learning Technique to Classify EMG Signal for Diabetes Person

Improved Multichannel Electromyograph Using Off-the-Shelf Components for Education and Research

Contact Info

Product

Resources

About