Matching Items (11)
Filtering by
- Creators: Arizona State University
- Creators: Harrington Bioengineering Program
- Resource Type: Text
Description
In recent years, machine learning and data mining technologies have received growing attention in several areas such as recommendation systems, natural language processing, speech and handwriting recognition, image processing and biomedical domain. Many of these applications which deal with physiological and biomedical data require person specific or person adaptive systems. The greatest challenge in developing such systems is the subject-dependent data variations or subject-based variability in physiological and biomedical data, which leads to difference in data distributions making the task of modeling these data, using traditional machine learning algorithms, complex and challenging. As a result, despite the wide application of machine learning, efficient deployment of its principles to model real-world data is still a challenge. This dissertation addresses the problem of subject based variability in physiological and biomedical data and proposes person adaptive prediction models based on novel transfer and active learning algorithms, an emerging field in machine learning. One of the significant contributions of this dissertation is a person adaptive method, for early detection of muscle fatigue using Surface Electromyogram signals, based on a new multi-source transfer learning algorithm. This dissertation also proposes a subject-independent algorithm for grading the progression of muscle fatigue from 0 to 1 level in a test subject, during isometric or dynamic contractions, at real-time. Besides subject based variability, biomedical image data also varies due to variations in their imaging techniques, leading to distribution differences between the image databases. Hence a classifier learned on one database may perform poorly on the other database. Another significant contribution of this dissertation has been the design and development of an efficient biomedical image data annotation framework, based on a novel combination of transfer learning and a new batch-mode active learning method, capable of addressing the distribution differences across databases. The methodologies developed in this dissertation are relevant and applicable to a large set of computing problems where there is a high variation of data between subjects or sources, such as face detection, pose detection and speech recognition. From a broader perspective, these frameworks can be viewed as a first step towards design of automated adaptive systems for real world data.
ContributorsChattopadhyay, Rita (Author) / Panchanathan, Sethuraman (Thesis advisor) / Ye, Jieping (Thesis advisor) / Li, Baoxin (Committee member) / Santello, Marco (Committee member) / Arizona State University (Publisher)
Created2013
Description
Electromyogram (EMG)-based control interfaces are increasingly used in robot teleoperation, prosthetic devices control and also in controlling robotic exoskeletons. Over the last two decades researchers have come up with a plethora of decoding functions to map myoelectric signals to robot motions. However, this requires a lot of training and validation data sets, while the parameters of the decoding function are specific for each subject. In this thesis we propose a new methodology that doesn't require training and is not user-specific. The main idea is to supplement the decoding functional error with the human ability to learn inverse model of an arbitrary mapping function. We have shown that the subjects gradually learned the control strategy and their learning rates improved. We also worked on identifying an optimized control scheme that would be even more effective and easy to learn for the subjects. Optimization was done by taking into account that muscles act in synergies while performing a motion task. The low-dimensional representation of the neural activity was used to control a two-dimensional task. Results showed that in the case of reduced dimensionality mapping, the subjects were able to learn to control the device in a slower pace, however they were able to reach and retain the same level of controllability. To summarize, we were able to build an EMG-based controller for robot devices that would work for any subject, without any training or decoding function, suggesting human-embedded controllers for robotic devices.
ContributorsAntuvan, Chris Wilson (Author) / Artemiadis, Panagiotis (Thesis advisor) / Muthuswamy, Jitendran (Committee member) / Santos, Veronica J (Committee member) / Arizona State University (Publisher)
Created2013
Description
The use of electromyography (EMG) signals to characterize muscle fatigue has been widely accepted. Initial work on characterizing muscle fatigue during isometric contractions demonstrated that its frequency decreases while its amplitude increases with the onset of fatigue. More recent work concentrated on developing techniques to characterize dynamic contractions for use in clinical and training applications. Studies demonstrated that as fatigue progresses, the EMG signal undergoes a shift in frequency, and different physiological mechanisms on the possible cause of the shift were considered. Time-frequency processing, using the Wigner distribution or spectrogram, is one of the techniques used to estimate the instantaneous mean frequency and instantaneous median frequency of the EMG signal using a variety of techniques. However, these time-frequency methods suffer either from cross-term interference when processing signals with multiple components or time-frequency resolution due to the use of windowing. This study proposes the use of the matching pursuit decomposition (MPD) with a Gaussian dictionary to process EMG signals produced during both isometric and dynamic contractions. In particular, the MPD obtains unique time-frequency features that represent the EMG signal time-frequency dependence without suffering from cross-terms or loss in time-frequency resolution. As the MPD does not depend on an analysis window like the spectrogram, it is more robust in applying the timefrequency features to identify the spectral time-variation of the EGM signal.
ContributorsAustin, Hiroko (Author) / Papandreou-Suppappola, Antonia (Thesis advisor) / Kovvali, Narayan (Committee member) / Muthuswamy, Jitendran (Committee member) / Arizona State University (Publisher)
Created2012
Description
Myoelectric control is lled with potential to signicantly change human-robot interaction.
Humans desire compliant robots to safely interact in dynamic environments
associated with daily activities. As surface electromyography non-invasively measures
limb motion intent and correlates with joint stiness during co-contractions,
it has been identied as a candidate for naturally controlling such robots. However,
state-of-the-art myoelectric interfaces have struggled to achieve both enhanced
functionality and long-term reliability. As demands in myoelectric interfaces trend
toward simultaneous and proportional control of compliant robots, robust processing
of multi-muscle coordinations, or synergies, plays a larger role in the success of the
control scheme. This dissertation presents a framework enhancing the utility of myoelectric
interfaces by exploiting motor skill learning and
exible muscle synergies for
reliable long-term simultaneous and proportional control of multifunctional compliant
robots. The interface is learned as a new motor skill specic to the controller,
providing long-term performance enhancements without requiring any retraining or
recalibration of the system. Moreover, the framework oers control of both motion
and stiness simultaneously for intuitive and compliant human-robot interaction. The
framework is validated through a series of experiments characterizing motor learning
properties and demonstrating control capabilities not seen previously in the literature.
The results validate the approach as a viable option to remove the trade-o
between functionality and reliability that have hindered state-of-the-art myoelectric
interfaces. Thus, this research contributes to the expansion and enhancement of myoelectric
controlled applications beyond commonly perceived anthropomorphic and
\intuitive control" constraints and into more advanced robotic systems designed for
everyday tasks.
Humans desire compliant robots to safely interact in dynamic environments
associated with daily activities. As surface electromyography non-invasively measures
limb motion intent and correlates with joint stiness during co-contractions,
it has been identied as a candidate for naturally controlling such robots. However,
state-of-the-art myoelectric interfaces have struggled to achieve both enhanced
functionality and long-term reliability. As demands in myoelectric interfaces trend
toward simultaneous and proportional control of compliant robots, robust processing
of multi-muscle coordinations, or synergies, plays a larger role in the success of the
control scheme. This dissertation presents a framework enhancing the utility of myoelectric
interfaces by exploiting motor skill learning and
exible muscle synergies for
reliable long-term simultaneous and proportional control of multifunctional compliant
robots. The interface is learned as a new motor skill specic to the controller,
providing long-term performance enhancements without requiring any retraining or
recalibration of the system. Moreover, the framework oers control of both motion
and stiness simultaneously for intuitive and compliant human-robot interaction. The
framework is validated through a series of experiments characterizing motor learning
properties and demonstrating control capabilities not seen previously in the literature.
The results validate the approach as a viable option to remove the trade-o
between functionality and reliability that have hindered state-of-the-art myoelectric
interfaces. Thus, this research contributes to the expansion and enhancement of myoelectric
controlled applications beyond commonly perceived anthropomorphic and
\intuitive control" constraints and into more advanced robotic systems designed for
everyday tasks.
ContributorsIson, Mark (Author) / Artemiadis, Panagiotis (Thesis advisor) / Santello, Marco (Committee member) / Greger, Bradley (Committee member) / Berman, Spring (Committee member) / Sugar, Thomas (Committee member) / Fainekos, Georgios (Committee member) / Arizona State University (Publisher)
Created2015
Description
All too often, industrial designers face seemingly intractable obstacles as they endeavor to, as Simon (1996, p. 111) describes, devise "courses of action aimed at changing existing situations into preferred ones." These problems, described by Rittel and Webber (1973) as "wicked," are insurmountable due to the contradictory and changing nature of their requirements. I argue that that industrial design (ID) is largely subject to Rittel's quandary because of its penchant for producing single solutions for large populations; such design solutions are bound, in some senses, to fail due to the contradictory and changing nature of large and, thus, inherently diverse populations. This one-size-fits-all approach is not a necessary attribute of ID, rather, it is a consequence of the time in which it came into being, specifically, the period of industrial mass production. Fortunately, new, agile manufacturing techniques, inexpensive sensors, and machine learning provide an alternative course for ID to take, but it requires a new way of thinking and it requires a new set of methods, which I will elaborate in this thesis. According to Duguay, Landry, and Pasin (1997), we are entering an age where it will be feasible to produce individualized, one-off products from large-scale industrial manufacturing facilities in a way that is not only cost effective, but in many ways as cost effective as the existing techniques of mass production. By availing ourselves of these opportunities, we can tame the problem, not by defeating Rittel's logic, rather by reducing the extent to which his theories are appropriate to the domain of ID. This thesis also describes a test study: an experiment whose design was guided by the proposed design methodologies. The goal of the experiment was to determine the feasibility of a noninvasive system for measuring the health of the forearm muscles. Such a tool would provide the basis for assessing the true impact and possible pathogeny of the manual use of products or modifications to products. Previously, it was considered impossible to use surface electromyography (as opposed to needle or wire based electromyography) to assess muscular activity and muscular health due to the complexity of the arrangement of muscles in the forearm. Attempts to overcome this problem have failed because they have tried to create a single solution for all people. My hypothesis is that, by designing for each individual, a solution may be found. Specifically, I show that, for any given individual, there is a high correlation between the EMG signal and the movements of the fingers that, ostensibly, those muscles control. In other words, by knowing, with great accuracy, the position and the motion of the hand then it would become possible to disambiguate the mixed signals coming from the complex web of muscles in the forearm and enable the assessment of the forearm's health by non-invasive means.
ContributorsBraiman, Stuart (Author) / Giard, Jacques (Thesis advisor) / Black Jr., John A (Committee member) / Herring, Donald (Committee member) / Arizona State University (Publisher)
Created2012
Description
Bioscience High School, a small magnet high school located in Downtown Phoenix and a STEAM (Science, Technology, Engineering, Arts, Math) focused school, has been pushing to establish a computer science curriculum for all of their students from freshman to senior year. The school's Mision (Mission and Vision) is to: "..provide a rigorous, collaborative, and relevant academic program emphasizing an innovative, problem-based curriculum that develops literacy in the sciences, mathematics, and the arts, thus cultivating critical thinkers, creative problem-solvers, and compassionate citizens, who are able to thrive in our increasingly complex and technological communities." Computational thinking is an important part in developing a future problem solver Bioscience High School is looking to produce. Bioscience High School is unique in the fact that every student has a computer available for him or her to use. Therefore, it makes complete sense for the school to add computer science to their curriculum because one of the school's goals is to be able to utilize their resources to their full potential. However, the school's attempt at computer science integration falls short due to the lack of expertise amongst the math and science teachers. The lack of training and support has postponed the development of the program and they are desperately in need of someone with expertise in the field to help reboot the program. As a result, I've decided to create a course that is focused on teaching students the concepts of computational thinking and its application through Scratch and Arduino programming.
ContributorsLiu, Deming (Author) / Meuth, Ryan (Thesis director) / Nakamura, Mutsumi (Committee member) / Computer Science and Engineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2016-05
Description
The goal of this project was to explore biomimetics by creating a jellyfish flying device that uses propulsion of air to levitate while utilizing electromyography signals and infrared signals as mechanisms to control the device. Completing this project would require knowledge of biological signals, electrical circuits, computer programming, and physics to accomplish. An EMG sensor was used to obtain processed electrical signals produced from the muscles in the forearm and was then utilized to control the actuation speed of the tentacles. An Arduino microprocessor was used to translate the EMG signals to infrared blinking sequences which would propagate commands through a constructed circuit shield to the infrared receiver on jellyfish. The receiver will then translate the received IR sequence into actions. Then the flying device must produce enough thrust to propel the body upwards. The application of biomimetics would best test my skills as an engineer as well as provide a method of applying what I have learned over the duration of my undergraduate career.
ContributorsTsui, Jessica W (Author) / Muthuswamy, Jitteran (Thesis director) / Blain Christen, Jennifer (Committee member) / Barrett, The Honors College (Contributor) / Harrington Bioengineering Program (Contributor)
Created2014-05
Description
The global population over the age of 60 is estimated to rise to 23% by 2050 only increase the prevalence of functional neurological disorders and stroke. Increase in cases of functional neurological disorders and strokes will place a greater burden on the healthcare industry, specifically physical therapy. Physical therapy is vital for a patient’s recovery of motor function which is time demanding and taxing on the physical therapist. Wearable robotics have been proven to improve functional outcomes in gait rehabilitation by providing controlled high dosage and high-intensity training. Accurate control strategies for assistive robotic exoskeletons are vital for repetitive high precisions assistance for cerebral plasticity to occur.
This thesis presents a preliminary determination and design of a control algorithm for an assistive ankle device developed by the ASU RISE Laboratory. The assistive ankle device functions by compressing a spring upon heel strike during gait, remaining compressed during mid-stance and then releasing upon initiation of heel-off. The relationship between surface electromyography and ground reactions forces were used for identification of user-initiated heel-off. The muscle activation of the tibialis anterior combined with the ground reaction forces of the heel pressure sensor generated potential features that will be utilized in the revised control algorithm for the assistive ankle device. Work on this project must proceed in order to test and validate the revised control algorithm to determine its accuracy and precision.
This thesis presents a preliminary determination and design of a control algorithm for an assistive ankle device developed by the ASU RISE Laboratory. The assistive ankle device functions by compressing a spring upon heel strike during gait, remaining compressed during mid-stance and then releasing upon initiation of heel-off. The relationship between surface electromyography and ground reactions forces were used for identification of user-initiated heel-off. The muscle activation of the tibialis anterior combined with the ground reaction forces of the heel pressure sensor generated potential features that will be utilized in the revised control algorithm for the assistive ankle device. Work on this project must proceed in order to test and validate the revised control algorithm to determine its accuracy and precision.
ContributorsGaytan-Jenkins, Daniel Rinaldo (Author) / Zhang, Wenlong (Thesis director) / Tyler, Jamie (Committee member) / Harrington Bioengineering Program (Contributor, Contributor) / Barrett, The Honors College (Contributor)
Created2019-05
Description
Electromyography (EMG) and Electroencephalography (EEG) are techniques used to detect electrical activity produced by the human body. EMG detects electrical activity in the skeletal muscles, while EEG detects electrical activity from the scalp. The purpose of this study is to capture different types of EMG and EEG signals and to determine if the signals can be distinguished between each other and processed into output signals to trigger events in prosthetics. Results from the study suggest that the PSD estimates can be used to compare signals that have significant differences such as the wrist, scalp, and fingers, but it cannot fully distinguish between signals that are closely related, such as two different fingers. The signals that were identified were able to be translated into the physical output simulated on the Arduino circuit.
ContributorsJanis, William Edward (Author) / LaBelle, Jeffrey (Thesis director) / Santello, Marco (Committee member) / Barrett, The Honors College (Contributor) / Computer Science and Engineering Program (Contributor)
Created2013-12
Description
Electromyography (EMG) is an extremely useful tool in extracting control signals from the human body. Needle electromyography is the current standard for obtaining superior quality muscle signals and obtaining signals corresponding to individual muscles. However, needle EMG faces many problems when converting from the laboratory to marketable devices, specifically in home devices. Many patients have issues with needles and the extra care required of needle EMG is prohibitive. Therefore, a surface EMG device that can obtain clear signals from individual muscles would be valuable to many markets in the development of next generation in home devices. Here, signals from surface EMG were analyzed using a low noise EMG evaluation system (RHD 2000; Intan Technologies). The signal to noise ratio (SNR) was calculated using MatLab. The average SNR is 4.447 for the Extensor Carpi Ulnaris, and 7.369 for the Extensor Digitorum Communis. Spectral analysis was performed using the Welch approach in MatLab. The power spectrum indicated that low frequency signals dominate the EMG of small hand muscles. Also, harmonic bands of 60Hz noise were present as part of the signal which should be accounted for with filters in future iterations of the testing method. Provided is evidence that strong, independent signals were acquired and could be used in further application of surface EMG corresponding to lifting of the fingers.
ContributorsSnyder, Joshua Scott (Author) / Muthuswamy, Jit (Thesis director) / Buneo, Christopher (Committee member) / Harrington Bioengineering Program (Contributor) / School of International Letters and Cultures (Contributor) / Barrett, The Honors College (Contributor)
Created2016-05