Automatic Search Algorithm using Signal Enhancement, Flex Sensors and exploration of Machine learning Features to improve Multiple Channel Neuromuscular Electrical Stimulation (NMES) systems
(2024) BMEM01 20241Department of Biomedical Engineering
- Abstract
- Neuromuscular Electrical Stimulation (NMES) is a method to stimulate the body's muscles using bursts of electricity on the skin that can strengthen the muscles, increase coordination, and simulate human muscle movement to prevent the formation of Deep Vein Thrombosis (DTV). Electrically sensitive skin areas, called Motor Points (MP), must be stimulated to achieve a robust muscular response and decrease perceived discomfort. Conventionally, locating MPs is done using a search pen, which is a slow and time-consuming process. This study investigates using a baseline automatic search algorithm to locate the motor points by utilizing flex sensors to measure plantar flexion (PF) response when electrically stimulating the muscles in the calf.... (More)
- Neuromuscular Electrical Stimulation (NMES) is a method to stimulate the body's muscles using bursts of electricity on the skin that can strengthen the muscles, increase coordination, and simulate human muscle movement to prevent the formation of Deep Vein Thrombosis (DTV). Electrically sensitive skin areas, called Motor Points (MP), must be stimulated to achieve a robust muscular response and decrease perceived discomfort. Conventionally, locating MPs is done using a search pen, which is a slow and time-consuming process. This study investigates using a baseline automatic search algorithm to locate the motor points by utilizing flex sensors to measure plantar flexion (PF) response when electrically stimulating the muscles in the calf. Furthermore, signal processing is utilized to enhance accuracy and attempt to reduce search time. A total of 120 searches using the baseline algorithm and signal processing were conducted on three individuals to evaluate the accuracy of locating the MPs. This study demonstrates that performing a Fast Fourier Transform (FFT) on the sensor data enhances the detection of PF and, thus, the location of MP's at low current amplitude. Further research is warranted to enhance accuracy and search time by utilizing multi-electrode stimulation and machine learning in decision-making. (Less)
- Popular Abstract
- Reducing the risk of blood clot formation using electrical stimulation, a comfortable substitute for movement-impaired people
In today's society, where one out of four people succumbs to blood clot-related conditions, with 60% of these cases occurring during or after hospitalization, the need for a solution is pressing. The primary cause of these clots is a lack of muscle movement in the legs. Prolonged blood flow interruption can cause blood to pool and form clots. This thesis delves into the potential of Neuromuscular Electrical Stimulation (NMES) to emulate muscular movement in a comfortable manner, thereby significantly reducing the risk of blood clot formation and potentially saving lives.
The use of electricity to stimulate... (More) - Reducing the risk of blood clot formation using electrical stimulation, a comfortable substitute for movement-impaired people
In today's society, where one out of four people succumbs to blood clot-related conditions, with 60% of these cases occurring during or after hospitalization, the need for a solution is pressing. The primary cause of these clots is a lack of muscle movement in the legs. Prolonged blood flow interruption can cause blood to pool and form clots. This thesis delves into the potential of Neuromuscular Electrical Stimulation (NMES) to emulate muscular movement in a comfortable manner, thereby significantly reducing the risk of blood clot formation and potentially saving lives.
The use of electricity to stimulate muscles has its roots in the late 18th century when Luigi Galvani demonstrated that electrical impulses could create muscle movement in frog legs. Since these discoveries, electrical stimulation has been widely used, from strengthening the muscles through post-surgical procedures to relieving muscle pain. However, as the method uses electrical pulses directly on the skin, it is often perceived as rather uncomfortable when the strength of the electrical pulse is high.
Through discoveries made from previous research, it is known that areas on the skin are extra sensitive to electrical stimulations, resulting in muscle movement at low intensity, making it more comfortable; these areas are called muscular Motor Points (MPs). This study evaluates different methods of automatically locating these motor points by utilizing several electrodes and measuring the muscle response upon stimulation.
The results demonstrate promising results but highlight the importance of further research to enhance the performance. Using signal processing on the data gathered from the muscle response, the number of times the correct motor points were found increased by 45\% compared to simply analyzing the raw response. It is essential to locate and isolate the actual muscle movement from the signal, as it was shown to be contaminated with noise. The key takeaway from the study is that signal processing seemed to assist in enhancing the muscle response. However, more sensitive sensors are essential for measuring movement and improving accuracy.
As the study concludes, it opens the door to exciting possibilities in the field of NMES research. By venturing into more advanced methods, such as Adaptive Filters and Machine Learning, we can potentially increase the accuracy and efficiency of locating the MPs. The prospect of gathering a vast amount of data from numerous tests and uncovering hidden statistical information is not only promising but also inspiring for the future of NMES. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/9173430
- author
- Berthold, Elliot LU
- supervisor
- organization
- alternative title
- Självständig sökalgoritm med signalförstärkning, böjsensorer och utforskning av maskininlärningsfunktioner för att förbättra system för neuromuskulär elektrisk stimulering (NMES) med flera kanaler
- course
- BMEM01 20241
- year
- 2024
- type
- H2 - Master's Degree (Two Years)
- subject
- keywords
- Neuromuscular Electrical Stimulation, Deep Vein Thrombosis, Motor Point
- language
- English
- additional info
- 2024-22
- id
- 9173430
- date added to LUP
- 2024-09-09 12:47:36
- date last changed
- 2024-09-09 12:47:36
@misc{9173430,
abstract = {{Neuromuscular Electrical Stimulation (NMES) is a method to stimulate the body's muscles using bursts of electricity on the skin that can strengthen the muscles, increase coordination, and simulate human muscle movement to prevent the formation of Deep Vein Thrombosis (DTV). Electrically sensitive skin areas, called Motor Points (MP), must be stimulated to achieve a robust muscular response and decrease perceived discomfort. Conventionally, locating MPs is done using a search pen, which is a slow and time-consuming process. This study investigates using a baseline automatic search algorithm to locate the motor points by utilizing flex sensors to measure plantar flexion (PF) response when electrically stimulating the muscles in the calf. Furthermore, signal processing is utilized to enhance accuracy and attempt to reduce search time. A total of 120 searches using the baseline algorithm and signal processing were conducted on three individuals to evaluate the accuracy of locating the MPs. This study demonstrates that performing a Fast Fourier Transform (FFT) on the sensor data enhances the detection of PF and, thus, the location of MP's at low current amplitude. Further research is warranted to enhance accuracy and search time by utilizing multi-electrode stimulation and machine learning in decision-making.}},
author = {{Berthold, Elliot}},
language = {{eng}},
note = {{Student Paper}},
title = {{Automatic Search Algorithm using Signal Enhancement, Flex Sensors and exploration of Machine learning Features to improve Multiple Channel Neuromuscular Electrical Stimulation (NMES) systems}},
year = {{2024}},
}