Neural Engineering Student Projects

Permanent URI for this collection

These projects were created as part of the graduate level neural engineering course which focuses on current topics in neural engineering. Examples include brain-machine interfaces, neurophysiological methods, instrumentation for interfacing electronics to the nervous system (in vivo and in vitro), optogenetics, ultrasound interfaces (for stimulation and recording in the nervous system), neural modeling, and prosthesis design.


Recent Submissions

Now showing 1 - 20 of 42
  • Item
    Evaluating the Plausibility of using Brain-to-Brain Communication to Create Neuromuscular Connection for Prosthetic Usage
    (2022) Asuagbor, Alesazem
    When prosthetics are developed, they are created for traumatic amputees to restore quality of life so they can live as they did before their limb loss. However, for an individual with congenital amputation — an underdeveloped or nonexistent limb from birth — there are not as many options due to many prosthetics relying on the user’s prior experience with the missing limb[1][2]. The purpose of this project is to investigate the possibility of using electroencephalogram (EEG) signals to control a prosthetic hand. Based on this knowledge, a prosthetic will be designed for an individual with a congenital amputation. Use of this device may be difficult for these individuals compared to those with trauma-based amputations because they lack experience with the limb in question. This can result in more difficulty during the rehabilitation process due to the neurological disconnect. This can be counteracted by utilizing brain-to-brain communication. By sending the neurophysiological signals of an able-bodied person to the individual with the amputation they can teach the amputee’s brain these new motor skills. First, the biosignals responsible for the motor functions of the specific limb must be identified and analyzed[1][3]. By determining if the signals for moving one limb are identical to the signals for the other limb, they can be sent via brain-to-brain communication to the patient with the missing limb. Sharing these signals will help them learn how to use the prosthetic, thus making rehabilitation easier. The hypothesis is that using the EEG signals communicated to the congenital amputee via brain-to-brain communication will make the neurological connections needed to make moving the prosthetic easier. The end device would be equivalent to a myoelectric prosthetic controlled by Phantom Limb Syndrome (PLS), except it will utilize learned EEG signals instead of depending solely on PLS[1][2].
  • Item
    EEG-based Emotion Recognition with Music: A Model and Application
    (2022-11) Scavotto, Zakariyya
    With the growth of music streaming, both for pleasure and other applications, such as music therapy, being able to understand how music makes someone feel has increased in importance. The goal of this study was twofold: first, create a machine learning model to predict a subject’s emotional response to music; then integrate this trained model into an application that can predict someone’s emotional response based on live data. Using support vector machines (SVMs) as the basis of the machine learning model, a model was trained to recognize the correct emotional response with 64% accuracy, and the model was successfully implemented into a demonstration web application.
  • Item
    Predicting Finger Movement Using an Ensemble Machine Learning Approach
    (2022-05) Handjinicolaou, Peter; Peixoto, Nathalia
    Debilitating brain trauma caused by injury or stroke, and other neurological disorders, can hinder a person’s ability to use their hands. Brain-Computer Interfaces (BCIs) are a subject of great interest regarding augmenting or restoring functionality to victims of this type of trauma. Motor imagery is a process in which a subject under test imagines performing an action without physically doing so [3]. Using brainwave sensors such as EEGs, the state of neural communication for that action can be recorded without the interference caused by the movement associated with it [8]. Using Machine learning classification techniques such as Support Vector Machines (SVMs), Multilayer Perception models (MLPs), and Fischer Linear Discrimination Analysis (LDA), it is possible to select for features and accurately predict upcoming movement by using motor imagery training and EEG data collection.
  • Item
    EEG Analysis to detect Attention
    (2022-05) Nesseem, Caren; Peixoto, Nathalia
  • Item
    A Review of the Effects of Microwave Radiation on Spatial Memory and Learning
    (2022-05) Dockum, Allison; Peixoto, Nathalia
    Microwave radiation refers to electromagnetic waves between 300 MHz and 300 GHz. Microwaves are used in communication systems, manufacturing, medical treatments, and military operations. The ability to modify waveform parameters, such as frequency and duty cycle, contribute towards the versatility of microwave radiation. In the same way, the ability to adjust each parameter also contributes to the complexity of understanding the biological effects of microwaves. For some time, researchers have studied the microwave effects on learning and spatial memory in rodents. Rodents provide scientists with a neurologically similar model to humans, which is easy to study both on the cellular levels and to assess behaviorally due to the development of maze performance tests. However, the mechanisms disrupting spatial memory remain largely unknown because of the nearly infinite number of ways microwave can be modified, combined with the multitude of neurological effects, which could impact behavior. It is important for scientists to continue to study rodent models under various microwave exposure conditions to prevent harmful exposure conditions with humans. In this review microwave exposure conditions will be introduced, followed by an introduction to the cruciality of synaptic plasticity to spatial memory and learning. Synaptic plasticity can be impacted through various neurological mechanisms; of which the NMDAR receptors, neurotransmitter release, and activation of intracellular signaling cascades, and cell apoptosis will be reviewed.
  • Item
    Mixed Reality Utilization in Anatomical Surgical Practice with HoloLens
    (2022-05) Montenegro, Fabian; Peixoto, Nathalia
    Surgeons have demonstrated common issues when in procedures. MR and AR utilization have recently been implemented into the medical field. Headsets like the HoloLens give advantages that are able to benefit both the patient and surgeon. Studies have shown this and provided data to which makes this a desirable change. An early build of this sort of software is being developed which include holograms of organs and characteristics to aid in medical procedures.
  • Item
    Gesture Classification from sEMG Signals using Machine Learning Approaches
    (2022-05-12) Rabbat, Nada; Peixoto, Nathalia
    The open-access sEMG dataset [2] is utilized to classify hand motor movements to the respective gesture they represent, using machine learning approaches. This is done as a prospective benchmark for fast and efficient communication in cases of disability, where the gesture class can be replaced with a letter in the alphabet, to form a sentence from a combination of gestures. The Machine Learning approaches tested in this paper are Logistic Regression, Random Forest, and Bagging Classifier algorithms. All approaches will be tested for their accuracy in classifying the sEMG data. The Bagging classification algorithm had the highest accuracy score, followed by Random Forest and Logistic Regression.
  • Item
    Attempted Prediction of Emotional Valence from EEG Using Multidimensional Directed Information
    (2022-05) Clayton A Baker; Peixoto, Nathalia
    Quantitative measurement of a person’s emotional state can aid performance in a number of areas, such as human-machine interactions, and psychological research. Electroencephalogram (EEG) data has shown potential as a predictor of emotional valence based on asymmetric activation patterns between the left and right hemispheres of the prefrontal cortex. Multidimensional directed information (MDI) is a computational tool that allows the measurement of information content transferred between different signals in a connected system, and has previously seen applications in EEG-based affective measurement in order to detect the presence of an emotional response. This study aimed to use MDI with EEG data from published datasets in order to derive a directional bias metric as a predictor for emotional valence based on frontal hemisphere asymmetry. Two methods of MDI computation were attempted; significant differences were observed in results between the two, suggesting possible errors in implementation. Neither method yielded output correlating with valence.
  • Item
    Review of different modes of TES used in treatment of neurodegenerative disorders
    (2022-05) Kaur, Antarjot; Peixoto, Nathalia
    Literature Review
  • Item
    The Study of Retinal Organoids: Development, Modeling, and Transplantation
    (2022-05) Farag, Sally; Peixoto, Nathalia
    There are several different retinal diseases prominent within the United States including age-related macular degeneration (AMD), diabetic retinopathy, and glaucoma. These three particular retinal diseases affect a total of 12.5 million people in the US alone, showing that retinal diseases are a very prevalent issue that require an improvement in the methodology of treatment. Retinal diseases can cause blindness and drastically decrease quality of life. To better treat these diseases and provide patients with proper care, it is important to maintain a proper understanding of the retina. It has been found that retinal organoids (ROs) can be produced through the use of stem cells. When perfected, these ROs can provide advanced modeling of the retina, as well as modeling of novel treatments for retinal diseases and allow for better testing. Additionally, ROs can be modified in order to better depict retinal diseases and their microenvironments. They may also be used for retinal transplants as a method of treatment. However, ROs are still being developed and understood. Before the full use and benefits of ROs can be reaped, the differences between ROs and natural retinas must be understood and minimized. Additionally, various issues within the ROs themself must be tackled in order for their uses to become more advanced.
  • Item
    Development of a low cost Electroencephalogram patient simulator
    (2022-05) Lancaster, Brandon; Peixoto, Nathalia
    Conducting human subjects research has some challenges for bioengineers such as lack of repeatability of bio-signals, IRB approval and recruitment. The COVID-19 pandemic has made it even more difficult for researchers to access research subjects. One biosignal that is economical, portable and quick to collect data from is Electroencephalography (EEG). Here we present a solution to researchers’ reduction of access to in vivo EEG signals. The solution is in the form of a patient EEG simulator that is controlled by an ATmega328P, connected to an analog to digital converter and other analog circuitry. Other commercial devices exist but this device was also constructed for educational purposes. The output signal can be adjusted to better match a real bio-potential from a human. Neurophysiology recordings can be loaded onto the device for use. Open source data is readily available for download to add to the database on the device. This simulator has other potential applications such as unit testing for devices under development and more convenient and rapid prototype iteration. Optimization of the device is also discussed.
  • Item
    The Use of Neuroengineering to Model Brain Disease: Alzheimer’s Disease, Parkinson’s Disease, and Schizophrenia
    (2022-05) Wright, Lacey; Peixoto, Nathalia
    Predicting, diagnosing, and treating neurological disease is a significant challenge in current medical research. This review will consider computational neuroengineering models of the brain for three specific neurological diseases: Alzheimers disease (AD), Parkinson’s disease (PD), and schizophrenia. These diseases were specifically chosen because they affect similar areas of the brain and involve similar symptoms. The physiology and pathology of each disease will be reviewed, as well as the most promising computational models of these diseases. Computational simulation is expected to help improve the understanding of the relationship between cellular pathology and cognitive performance. Future approaches for diagnosing and treating AD, PD, and schizophrenia will be suggested based on conclusions made from the models’ results.
  • Item
    A Literature Review of Network Models in Neuroscience
    (2022-05) Kazemi Abharian, Sanaz; Parsa, Maryam; Peixoto, Nathalia
    Neuroscience is developing in size, scope, and complexity of neural data obtained from large portions of nervous systems. The primary aim of network models in neuroscience is to map, record, and model the components and interactions of the human brain and nervous systems. In this paper, our goal is to survey graph-based network models in neuroscience and study their applications to detect diseases and disorders related to the human brain.
  • Item
    Investigation into the Link between Object Perception and Neural Activity in the Human Brain Using ECoG Data
    (2022-05) Stenberg, Sabrina; Peixoto, Nathalia
    One area under investigation in the field of neuroscience is the link between object perception and neural activity in visual cortical areas of the human brain. By investigating the electrical potentials from the ventral temporal cortical surface in humans, the Stanford University study selected for this paper sought to collect sufficient information for spontaneous and near-instantaneous identification of a subject’s perceptual state. The brain signal data collection technique used by the researchers was electrocorticography (ECoG), using ECoG arrays placed on the subtemporal cortical surface of seven epilepsy patients. ECoG is an invasive electrogram method, requiring access to the surface of the brain, which can be applied to measure brain signals in response to specific stimuli. Using publicly available human ECoG recording data previously collected and made publicly available, this paper investigates visual object processing in the human brain. The data are taken from a study where seven epilepsy patients were shown house and face images in quick succession. We use those data and filter, process, and plot selected data to investigate the correct identification of the stimuli. We discovered that the incorrect stimuli matches are driven by variance in the human brain activity corresponding to the same set of stimuli. Better understanding of the visual processing capabilities of the human brain could lead to developments in machine learning, as well as generate recommendations for future data collection in human visual object processing.
  • Item
    Upper Limb Prosthetic Literature Review
    (2022-05) Philipson, Thomas; Peixoto, Nathalia
    The goal of this study is to determine how the needs of those who operate upper limb prosthetics can be met so that the prosthetic retention rate is higher, reflecting an improvement in their quality of life. As we use our extremities to feel and interact with our environment, it is crucial that developments are made specifically in the hands to allow individuals to complete tasks that require high dexterity and precision. Upper limb prosthetic users have a high rate of abandonment due to the lack of ownership and embodiment felt regarding the prosthetic limb [1]. Improved functionality of the prosthetic limb would reduce the load applied to the functional limb, which would lead to better prosthetic retention. Another area for improvement that is addressed in this literature review is the accuracy of the BCI component in signal acquisition and translation to prosthetic automation. Accuracy is important not only for functionality but for a sense of self in the user. A sense of self is vital for those using prosthetic limbs as the goal is for it to be a cohesive part of the user. This review will discuss the different methods for data acquisition and how to improve the accuracy of the resultant movement.
  • Item
    Methods of Seizure Detection: A Literature Review
    (2021) Ibrahim, Lamis
    epileptic seizure is a neurological disorder caused by abnormal activity of nerve cells in the brain. Epileptic seizures may happen while the patient is awake or asleep, and may cause loss of consciousness, falling, or massive muscle spasms. Frequent seizure events are dangerous as they can cause extreme injury and even death. Since the period of most seizures is less than two minutes, it is impossible to directly monitor all patients at risk of seizure. Devices have been developed such as electroencephalogram (EEG), and mattress pressure sensors to detect seizures and alert caregivers. Despite all the benefits of these systems, these systems cannot accurately detect seizures. EEG and ECG are impractical and not suitable for long-term seizure detection and mattress pressure sensors do not allow for accurate detection in all patients as a weight threshold must be reached for the sensors to detect movement. To address this, researchers have developed wearable devices that combine ECG and Photoplethysmography (PPG) to monitor and detect seizure events [1]. PPG is an optical technique that uses a light source and a photodetector at the surface of the skin to measure the volumetric changes in blood circulation which allows for the detection of heart rate [2]. PPG is also being combined with wearable devices and video to allow for an innovative seizure detection system. In this literature review, traditional methods of seizure detection such as EEG, mattress pressure sensors, video-based detection, and integrating PPG with wearable devices to detect seizures will be discussed and compared to understand and further appreciate seizure detection systems.
  • Item
    A Literature Review of the Development Trends of Visual Neural Protheses
    (2021) Givens, Jordan
    According to a systemic review of population-based data sets relevant to global vision impairment and or blindness between 1980 in 2015, there is an estimated 36 million people who suffer from blindness [4]. An additional 405.1 million people with mild to severe visual impairment [4]. Visual impairment of any degree and its growing prevalence are not a new issue. However, recent advancements in neural protheses, such as cochlear implants that aid those who are hearing impaired, leading researchers to turn to visual neural protheses. Visual neural prostheses focus on the concept of artificially inducing vision by using our current understanding of electrical stimulation, visual pathways, and visual sensations. Therefore, all visual protheses focus on creating an artificial sense of vision through the electrical activation of neurons belonging to the visual system of the body [2]. There are a variety of approaches researchers have taken to accomplish this, the approaches differ in the aspects of the visual system they attempt to replace. As of 2019 they are primarily four approaches that encompass most neural prosthetics. Those that focus on the retina, optical nerve, cortical region of the brain, and or the lateral geniculate nucleus (LGN) within the thalamus [2][3]. The most prominent of which is the retina neural prosthetic, having multiple current implants such as the Argus II electronic epiretinal device, this is due to its extracranial location and simpler organization compared to other methods [3][5]. However, the retinal approach is not without faults of its own such as unwanted electrochemical reactions and low resolution. Therefore, in this paper the four approaches of visual neural prostheses will be examined to provide a greater insight into the field.
  • Item
    EEG-Based BCI for Household IoT Control
    (2021) Dunn, Bryce
    Brain-computer interface (BCI) paradigms have existed for decades to improve communication and technological control. Electroencephalography (EEG) represents one of the most common non-invasive approaches toward recording brain signals in human participants [1]. Many features within EEG signal are used to model BCI including five major types of brain wave frequency bands, autoregressive parameters, and power spectral density values [2]. Processing EEG to produce a desired output demands signal filtration, feature extraction and classification. Approach. Using the Emotiv EPOC+ headset, electrodes placed on the scalp record passive mental activity at centimeter resolution [3]. The EPOC+ represents a low-cost alternative to medical-grade hardware, which may allow the development of this platform to be more accessible to end-users. Filtration and classification methods are applied to distinguish signal frequencies of interest. The recorded EEG signal is used to demonstrate potential for passive control a simple household appliance such as a light fixture. To this end, a participant’s passive EEG is recorded during a series of tasks in varying light settings. In addition to technical challenges, there are practical considerations to overcome, such as variability and subject fatigue. Preliminary results suggest a measurable distinction in mental state between tasks. Significance. Following this line of inquiry, a platform for interfacing with the increasingly ubiquitous internet of things (IoT) may be developed in the future. The potential applications of BCI are myriad and promise to better living conditions by enhancing and supplementing central nervous system output. EEG-based signaling may provide means to greater autonomy and technological accessibility for disabled people and patients with neurological deficits.
  • Item
    Neural Control of Cardiac Function, Rhythm, and Contraction
    (2021) O'Connell, Rachel
    This literature review discusses how researchers can control cardiac cells with neural networks. The cardiac system is controlled by the autonomic nervous system and coordinates with the respiratory system. Numerous approaches evaluate the efficiency, malfunction, and abnormalities of this connection between brain and heart, through computational analysis, with pharmacological influences, and with physiological analysis. In Roger A. L. Dampney’s paper, he discusses how the connection between the nervous system and cardiovascular system is achieved through feedforward and feedback regulation to maintain homeostasis, while Y. Zhong et al. takes a more dynamic approach and manipulates the patients into supine and upright positions while administering the drug, atropine, which involuntarily blocks the nervous system from working. Lastly, X. Chen et al. derived new PNS and SNS indexes by multisignal analysis of cardiorespiratory variability using transfer functions and a combination of physiological and pharmacological experiments. He selected autonomic nervous system blockers to analyze the beta-sympathetic nervous system and its overlap with the parasympathetic nervous system. He sought to find where these systems overlap in low frequency bands and offer specific measurements of the cardiac autonomic nervous system due to inputs from heart rate. Overall, the purpose of the literature review is to study the techniques that are being used in research to evaluate the relationship between the nervous system and cardiovascular system. The literature review will range from computational analysis to physiological experiments and include the analysis of pharmacological influences on mediating the ANS. In the Introduction, I will explain some of the terminology that is necessary to understand the experiments that follow
  • Item
    The Future of Personalized Brain Stimulation for the Treatment of Neuropsychiatric Disorders
    (2021) Fredriksz, Madeline
    Over a billion individuals worldwide suffer from neuropsychiatric disorders, many of which are resistant to traditional medications and therapy. Both invasive and non-invasive electrical stimulation have emerged as promising treatment options in recent years since they have been shown to attenuate neuropsychiatric symptoms by modulating the underlying abnormal brain network activity. However, neurological disorders manifest differently across individuals and the way brain activity and patient symptoms respond to stimulation is highly individualized. In addition, brain activity is not constant over an individual’s lifetime. Changes due to development, aging, and the progression of chronic conditions need to be taken into account. All of this reveals the need for personalized treatments that can adapt to brain activity dynamics.