Biomedical Commons^™

Explainable Machine Learning For Biomedical Diagnostics: Optical Imaging And Eeg Signal Analysis, Fozia Rajbdad

LSU Doctoral Dissertations

The growing convenience of complex biomedical data begins new roads for better disease detection and functional identification via artificial intelligence (AI). Nevertheless, conventional analysis methods often rely on basic metrics that drop sensitive biotic differences, and various AI systems are difficult to infer, limiting their clinical reliability and practical use. There is a growing need for explainable, physiologically relevant computational models that can extract key biomarkers from diverse biomedical data sources. This dissertation addresses this problem by obtaining explainable machine learning and deep learning procedures for studying biomedical signals and optical imaging data.

This dissertation is divided into two parts; …

Practical Multimodal Wearable Sensing For Functional Upper Extremity Primitive Classification With Application To Stroke Rehabilitation, Nicholas Weiss

Master's Theses

Stroke often causes long-term weakness and impaired motor control in the upper extremity (UE), making everyday tasks such as reaching, grasping, and moving objects more difficult. Restoring functional arm use is therefore a central goal of post-stroke rehabilitation. Measuring affected arm use continuously and objectively is important because isolated clinical assessments may not fully capture how the affected arm is used during therapy or daily life. Wearable sensors offer a promising approach for monitoring, but raw sensor signals are difficult to interpret directly. Functional movement primitives address this issue by describing UE behavior as smaller, task-agnostic movement units.

This thesis …

Visualization And Marker-Less Tracking Of User-Defined Pre-Processed Mri Articulator Data Using Deep Learning, Michael De George

Student Theses

This thesis presents a comprehensive framework for the automated tracking and visualization of articulatory movements based on magnetic resonance imaging (MRI) data. A well-known data analysis tool for markerless pose estimation, known as DeepLabCut, is investigated for this purpose. The performance of this tool is enhanced through the design and implementation of a pre-processor. DeepLabCut is a markerless pose estimation toolbox based on deep learning, which overcomes the issue of making manual annotations frame-by-frame. Limitations from manually marking the MRI images are addressed by implementing transfer learning with convolutional neural networks to achieve accurate, user-defined articulator tracking without markers. Current …

Exploring Longitudinal Stability Of Spatiotemporal Sequential Patterns By Means Of Autoencoder Schemes For Eeg Based Personal Identification, Muhammed E. Oztemel

LSU Doctoral Dissertations

Robust personal identification remains a critical and challenging task in the digital era. Electroencephalography (EEG) offers a unique biometric modality that captures individual brain dynamics through complex neural signals. This dissertation proposes autoencoder (AE) based feature extraction and subject identification through these features. EEG recordings are first transformed into topographic maps to represent spatial brain activity. Consecutive topomaps are then concatenated to capture temporal transitions across frames. Convolutional autoencoders (CAEs) are used to learn spatial and temporal patterns, while domain-adaptive AEs are designed to model evoked potential based responses. Additionally, self-attention mechanism is incorporated to enhance feature representation. To analyze …

A Deep Learning Approach For Semantic Segmentation And Its Application On Ctc., Samir Farag Harb

Electronic Theses and Dissertations

This dissertation explores the modeling and analysis of medical images, focusing on the intricate task of colon segmentation and subsequent 3D reconstruction, which are critical steps in Computed Tomography Colonography (CTC) systems. The primary objective of this research is to develop precise segmentation approaches to enhance the accuracy of colon identification and reconstruction from abdominal CT scans. Three distinct segmentation approaches are proposed and evaluated: a Markov Random Field (MRF)-based approach, a convolutional neural network (CNN)-based deep learning (DL) approach, and a sequential episodic training with dual contrastive learning Approach (G-SET-DCL) that has a flavor of few-shot learning (FSL). To …

Optimizing And Training An Svm-Based Breast Cancer Tumor Classifier, Kevin Lopatka

Master's Theses

With advancements in technology, turning to machine learning has become a popular choice for aiding clinicians in the diagnoses of breast cancer malignancies. While the neural networking approach has been vetted thoroughly, this work aims to take advantage of traditional machine learning techniques; mainly support vector machine learning and the optimizing of feature extraction. The discrete-wavelet transform is used in the feature extraction stage of machine learning. Previous works that use this feature extraction technique are analyzed and expanded upon by utilizing a variety of different wavelets as well as other color-spaces with the goal of achieving higher result metrics …

Neurovascular Coupling Impairments In Acute Traumatic Brain Injury: An Eeg-Nirs Analysis, Zachary Armstrong

Bioengineering Theses - Archive

Traumatic brain injury (TBI) is a major cause of neurological impairment, often leading to variable recovery and uncertain prognosis in the neurocritical care setting. There is a pressing clinical need for robust, physiologically grounded biomarkers to inform prognosis and therapeutic decision-making in acute TBI. This thesis investigates neurovascular coupling (NVC), the physiological coordination between neuronal activity and cerebral blood flow, as a candidate biomarker for brain function and recovery after injury.

A prospective cohort study was performed using simultaneous electroencephalography (EEG) and near-infrared spectroscopy (NIRS) recordings in patients with moderate-to-severe TBI and healthy controls. Wavelet transform coherence (WTC) analysis was …

Cochlear Electrode Insertion Training Model, Sarah Powell, Kaelyn E. Kraley, Nathan J. Smith

Williams Honors College, Honors Research Projects

Cochlear implant surgery is a delicate procedure performed by Otolaryngologists (ENTs) to implant an electronic device into the inner ear to provide a sense of sound for people who are profoundly deaf or hard of hearing. The current practices of training involve cadavers and 3D-printed models. Cadavers are commonly used but are expensive, single-use, and do not provide visual and haptic feedback, which are essential for medical students. 3D printed models are less commonly used and are hard to fabricate and not as realistic. If medical students are not properly trained for this delicate procedure, then risks are significantly increased …

Identification Of Fiducial Points In Seismocardiographic Cycles Using Manual And Automated Annotation Methods, Jasmine-Vy T. Truong

Honors Undergraduate Theses

There is currently a need for complementary methods for non-invasive cardiac monitoring. Seismocardiography (SCG), the measurement of cardiac-induced vibrations at the chest surface, has shown potential clinical utility. Improving the reliability of detecting fiducial points of electrocardiography (ECG) and SCG, which collectively capture the electro-mechanical cardiac activities, could expand ECG/SCG utility as a low-cost, accessible tool for clinical assessment. This study identifies commonly accepted criteria for fiducial point detection in SCG and ECG through an extensive literature review and signal processing techniques. The previous criteria were evaluated to identify their strengths and weaknesses. Based on the findings, an improved set …

Design Of A Portable Fast Scan Cyclic Voltammetry Device Utilizing Pulse Width Modulation For Waveform Generation, Nora Szymkowski

Masters Theses

Fast Scan Cyclic Voltammetry (FSCV) is a widely used electrochemical technique for real-time measurement of the brain’s chemical messengers, including the molecule dopamine, with high temporal resolution. Currently the financial burden of performing FSCV is quite high, ranging from $8,000 to $20,000+ making the barrier to entry nearly insurmountable for laboratories and classrooms at small institutions. The purpose of this project was to develop a Do-It-Yourself (DIY), portable, and cost-effective FSCV system for use in laboratory and classroom settings. The project aimed to create a compact and cost-effective system that could be used by researchers and educators to study dopamine …

Novel Sensors, Algorithms And Metrics For Human-Robot Interaction., Henry Lee Reynolds

Electronic Theses and Dissertations

The increased presence and deployment of robotics in sectors such as the medical field results in the demand for robots to, directly and indirectly, interface with people and their environment, making human-robot interaction (HRI) a vital thrust of robotics research. Assistive robots, for example, aid humans in accomplishing tasks or by providing support in the workforce. As the demand for nurses and the aging population increases, the assistive robots deployed will be deeply rooted in environments that require constant interaction with humans. This work contributed to improving aspects of HRI through 1) Expanding accessibility of the methods used for interfacing …

Authenticated Diagnosing Of Covid-19 Using Deep Learning-Based Ct Image Encryption Approach, Mohamed Attia Abdelgwad, Amira Hassan Abed, Mahmoud Bahloul

Future Computing and Informatics Journal

Researchers are motivated to use artificial intelligence in biometrics, medical imaging encryption, as well as cybersecurity due to its rapid progress. An encryption method for CT scans—which are used to diagnose COVID-19 disease—is proposed in this study. The suggested encryption method creates a connection among an individual's face picture and CT image to increase confidentiality. The simple CT picture is first enhanced with a host image. An encryption key is multiplied by the final result. This key is produced by applying a Convolutional Neural Network (CNN) to recognize characteristics from people's face photographs. Additionally, a straightforward CNN with three convolutional …

Segmentation And Classification Of Left Ventricular Abnormalities In Cardiac Mri Using Initial Point Prediction Based Deformable Model, Md. Asadur Rahman, Md. Al Noman, A. B. M. Aowlad Hossain

Future Computing and Informatics Journal

The shape of the left ventricle (LV) of a cardiac magnetic resonance image (CMRI) helps physicians to diagnose different cardiac abnormalities. The similarity of pixel intensity and shape of LV with neighbor tissues, the imprecision of boundaries, and the presence of noise are the challenges to accurate segmentation of LV. This paper contributes to the successful implementation of an automatic edge contouring method to segment LV area from CMRI and detect whether the ventricle belongs to abnormalities. This method proposes the regression-based artificial neural network to predict the possible initial position of the deformable edge-based active contour model for precise …

Shape Memory Alloy Capsule Micropump For Drug Delivery Applications, Youssef Mohamed Kotb

Theses and Dissertations

Implantable drug delivery devices have many benefits over traditional drug administration techniques and have attracted a lot of attention in recent years. By delivering the medication directly to the tissue, they enable the use of larger localized concentrations, enhancing the efficacy of the treatment. Passive-release drug delivery systems, one of the various ways to provide medication, are great inventions. However, they cannot dispense the medication on demand since they are nonprogrammable. Therefore, active actuators are more advantageous in delivery applications. Smart material actuators, however, have greatly increased in popularity for manufacturing wearable and implantable micropumps due to their high energy …

Technologies For Wearable Seizure Detection: A Systematic Review, Rhema Losli

University Honors Theses

Knowing when a seizure occurred is helpful because this information can be used to evaluate the effectiveness of seizure interventions and possibly alert caregivers to emergency situations. The current practice for recording seizures outside of a hospital and without sensors is through keeping a self-reported seizure diary. This practice may be unreliable if the diary is not updated or the person having the seizure does not realize it is happening. Wearable seizure detectors aim to solve this problem by reliably recording when a seizure happened and either sending out an alert or storing the data for later analysis. In this …

2-Channel Eeg Neurofeedback System, Tim Erwin, Donna Nikjou, Sebastian Turkewitz

Electrical Engineering

This work describes the design of an EEG-based neurofeedback system which provides users with real-time feedback on their level of focus or relaxation. By analyzing the spectral content of the brain activity measured via scalp electrodes, focus and relaxation levels can be quantified. Based on these measurements, live feedback in the form of moving bar graphs is provided to users, allowing them to gain awareness of their mental-state and more efficiently learn how to consciously relax or focus. This project covers the design of the system, including amplification and filtering stages, digitization, and signal processing. The system interfaces with a …

Affordable Bioprint Head-Adapter For 3d Printers, Andrew Ceralde, Dominic Drake, Noah Engles, Mohammad Alwan, Nevada Perry

Honors Capstones

Introduction: Addressing the need for affordable and accessible bioprinting technology, the Northern Illinois University bioprinting research and design project aims to democratize the field by developing an open source bioprint head. The initiative targets the prohibitive cost of commercial bioprinters by proposing an optimal design that can be integrated with widely available 3D printers, like the Creality Ender-3. This approach seeks to enable the widespread adoption of bioprinting technology, particularly for applications in tissue engineering and regenerative medicine.

Methods: Leveraging SolidWorks for design, the project employs a methodology that combines a precision extrusion system with thermal and UV crosslinking capabilities. …

Towards A Wearable Device For Measuring Impedance Plethysmography Of The Radial Artery, Pritom Chowdhury

Dartmouth College Master’s Theses

Recent advancements in bioimpedance technology have demonstrated significant promise in the application of cardiac health monitoring. This research explores the design and development of a forearm-based wearable bioimpedance device for non-invasive measurement of heart rate and respiratory rate at an accuracy level comparable to medical-grade monitors. It utilizes a tetrapolar electrode configuration to analyze bioimpedance changes in the radial artery due to blood flow.

An ongoing aspect of this work involves the preliminary development of an embedded framework intended to integrate signal generation, acquisition, and processing within the device to achieve compact and efficient system design, anticipated to contribute to …

Non-Invasive Monitoring Device For Early Detection Of Breast Cancer Related Lymphedema, Amy Prendergast

Honors Theses and Capstones

Breast Cancer Related Lymphedema (BCRL) is a common co-morbidity in cancer survivors following neoadjuvant therapies such as chemotherapy, radiation, and/or surgery. It is brought about by the disruption in the lymphatic system (think lymph node biopsy) that leads to a buildup of lymphatic fluid in the arm. Current diagnostic strategies for this condition are merely retroactive, and fairly limited in the parameters that are examined to ensure patient well-being long term. We hypothesize that with an approach that mimics bioimpedance spectroscopy analysis, we will be able to provide a clinical support tool that would better determine early stages of lymphedema …

Implementing Associative Learning Using Neuromorphic Robot, Vinay Kumar Pillalamarri

Dissertations, Master's Theses and Master's Reports

Associative learning, a key cognitive process seen across the animal kingdom, enables organisms to form connections between stimuli and adapt their behaviors based on past experiences. A particularly powerful example is fear conditioning, where animals learn to associate a neutral stimulus with an aversive one, allowing them to predict and avoid potential threats. Inspired by this mechanism, this project implements associative learning on an unmanned ground vehicle (UGV) to develop adaptive behavior through neuromorphic principles. Utilizing Nengo for neural modeling, the UGV learns to associate visual (red color) and tactile (vibration) stimuli through Hebbian learning, a biologically inspired synaptic adaptation …

Investigation Of Delta-Focused Ictal Electrical Source Imaging In Refractory Focal Epilepsy, Jared A. Rybarczyk

Theses and Dissertations--Electrical and Computer Engineering

Refractory focal epilepsy is characterized by the presence of seizures that cannot be controlled via anti-seizure medications. For patients suffering from this form of epilepsy, accurate identification of the seizure onset zone is a crucial step for many modalities of treatment. Electrical source imaging (ESI) allows for estimation of the seizure onset zone from electroencephalography. EEG feature extraction is an important step that can impact the final accuracy of source estimates. This work provides a review of 23 ictal ESI studies and proposes a delta-focused ictal ESI methodology. Our proposed delta-focused ictal ESI is implemented across 33 refractory focal epilepsy …

Improving Cellphone-Based Bio-Imaging Technique For Fluorescence Detection, Erteza T. Efaz

LSU Master's Theses

The research presented in this thesis focuses on the design, development, and evaluation of a fluorescence detection system. The system is implemented primarily as an Android application, Auto Camera, which leverages smartphone cameras to capture and analyze fluorescent images. The application provides a user-friendly interface with some configurable features like exposure time, ISO speed, and storage limit; as well as defining detection thresholds and setting acquisition intervals. This study begins with the architectural framework of the Android application, which is written in Java using Android Studio. The API compatibility is set to version 33, and users are prompted to grant …

An Enhanced Adaptive Learning System Based On Microservice Architecture, Abdelsalam Helmy Ibrahim, Mohamed Eliemy, Aliaa Abdelhalim Youssif

Future Computing and Informatics Journal

This study aims to enhance Adaptive Learning Systems (ALS) in Petroleum Sector in Egypt by using the Microservice Architecture and measure the impact of enhancing ALS by participating ALS users through a statistical study and questionnaire directed to them if they accept to apply the Cloud Computing Service “Microservices” to enhance the ALS performance, quality and cost value or not. The study also aims to confirm that there is a statistically significant relationship between ALS and Cloud Computing Service “Microservices” and prove the impact of enhancing the ALS by using Microservices in the cloud in Adaptive Learning in the Egyptian …

Visual Question Answering: A Survey, Gehad Assem El-Naggar

Future Computing and Informatics Journal

Visual Question Answering (VQA) has been an emerging field in computer vision and natural language processing that aims to enable machines to understand the content of images and answer natural language questions about them. Recently, there has been increasing interest in integrating Semantic Web technologies into VQA systems to enhance their performance and scalability. In this context, knowledge graphs, which represent structured knowledge in the form of entities and their relationships, have shown great potential in providing rich semantic information for VQA. This paper provides an abstract overview of the state-of-the-art research on VQA using Semantic Web technologies, including knowledge …

Characterization And Estimation Of Musculoskeletal Pain Using Machine Learning, Boluwatife Faremi

Master's Theses

Traditional scales utilized for recording pain are known to be highly subjective and biased due to inaccuracies in recollecting actual pain intensities. As a result, machine learning (ML) models that are trained using these scores as ground truth are reported to have low performance for objective pain classification because of the huge disparity between what was felt in moments of pain and the scores recorded afterward.

In the present study, two devices were designed for gathering real-time, continuous in-session subjective pain scores and the recording of the autonomic nervous system (ANS) altered endodermal (EDA) activity. 24 participants were recruited to …

Approximate And Sample Entropy Of Center Of Pressure In Unperturbed Tandem Standing: Contribution Of Embedding Dimension And Tolerance, Jayla Mashae Wesley

Masters Theses

Approximate entropy (ApEn) and sample entropy (SampEn) are statistical methods designed to quantify the regularity or predictability of a time series. Although ApEn has been a prominent choice for use, it is currently unclear as to which method and parameter selection combination is optimal for its application in biomechanics. The goal of this thesis was to examine the difference between ApEn and SampEn related to center of pressure (COP) data during standing balance tasks, while also refining tolerance r, to determine entropy optimization. Six participants completed five 30-second, feet together and tandem standing, trials under eyes-open and eyes-closed conditions. Ground …

Development Of Medical Device For Volume Status Assessment Using Peripheral Venous Pressure, Audra Beneux

Graduate Theses and Dissertations

The severe side effects of acute dehydration and blood loss may be prevented if assessed and treated quickly. As amplifier technology has improved, small peripheral venous pressure (PVP) signals collected using a transducer on an intravenous catheter can be analyzed to monitor patient volume status. This thesis details the development of an embedded system prototype that quickly and accurately assesses volume status using PVP analysis. The volume prediction algorithm classified 10 seconds of PVP data as “dehydrated” or “resuscitated”. The hardware design consisted of four primary areas: analog signal conditioning, processing, user interaction, and power. After data collection, the signal …

Portable Diffuse Reflectance Spectroscopy For Non-Invasive And Quantitative Assessment Of The Parathyroid Glands Viability During Surgery, Mark Romine, Linh Luong, Alex Moazzen, Katie Cho, Paul Lee

Symposium of Student Scholars

Portable Diffuse Reflectance Spectroscopy for Non-invasive and Quantitative Assessment of the Parathyroid Glands Viability During Surgery

Mark Romine, Linh Luong, Alex Moazzen, Katie Cho and Paul Lee

The parathyroid glands (PTGs) are responsible for the regulation of calcium levels in the blood by secreting a parathyroid hormone. This parathyroid hormone then regulates the body’s absorption, storage, and secretion of calcium, which can directly affect the way muscles and nerves operate. PTGs are often at risk of damage, or accidental removal during thyroid surgeries, because it is challenging to identify PTGs and to determine their viability. Current methods of visual inspections …

Analyzing Physiological Signals During The Menstrual Cycle: Introducing An Application Of Circular Statistics, Krystal D. Sides

UNF Graduate Theses and Dissertations

Physiological signals are used in engineering and the sciences to determine the state of functionality of certain physiological systems such as the human endometrium. Many women use the basal body temperature method to track ovulation, or their fertile window. However, temperature tracking alone is subject to environmental factors. Non-invasive wearable devices can be tools in recording cycle-related physiological features relevant to women’s health research. This research uses the Empatica E4 wristband to track changes in the physiological features during sleep across the menstrual cycle of ovulating and non-ovulating females. Due to the cyclic nature of menstrual cycles, analytic solutions to …

Small-Separation Speckle Contrast Optical Spectroscopy For Intraoperative Assessment Of Parathyroid Glands Viability During Thyroid Surgery, Connor Berger

Symposium of Student Scholars

The parathyroid glands (PTGs) are often damaged during thyroid surgeries due to a lack of methods identifying PTGs and assessing their viability. Damage to PTGs can cause hypocalcemia, a deficiency of calcium in the body. This complication can lead to detrimental consequences with economic burden. The surgeon’s current method of viability assessment is qualitative and subjective. Our technical solution is to employ an optical technique called speckle contrast optical spectroscopy (SCOS) that noninvasively quantifies the blood flow index (Db) of biological tissues at deep tissue levels (>1cm). The goal of this project is to verify SCOS at small source-detector-separation …