Matching Items (43)
Filtering by
- All Subjects: Machine Learning
- All Subjects: Biomaterials
- Creators: Harrington Bioengineering Program
- Creators: Spanias, Andreas
- Member of: Theses and Dissertations
Description
Colorimetric assays are an important tool in point-of-care testing that offers several advantages to traditional testing methods such as rapid response times and inexpensive costs. A factor that currently limits the portability and accessibility of these assays are methods that can objectively determine the results of these assays. Current solutions consist of creating a test reader that standardizes the conditions the strip is under before being measured in some way. However, this increases the cost and decreases the portability of these assays. The focus of this study is to create a machine learning algorithm that can objectively determine results of colorimetric assays under varying conditions. To ensure the flexibility of a model to several types of colorimetric assays, three models were trained on the same convolutional neural network with different datasets. The images these models are trained on consist of positive and negative images of ETG, fentanyl, and HPV Antibodies test strips taken under different lighting and background conditions. A fourth model is trained on an image set composed of all three strip types. The results from these models show it is able to predict positive and negative results to a high level of accuracy.
ContributorsFisher, Rachel (Author) / Blain Christen, Jennifer (Thesis director) / Anderson, Karen (Committee member) / School of Life Sciences (Contributor) / Harrington Bioengineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2021-05
Description
The goal of this research project is to create a Mathcad template file capable of statistically modelling the effects of mean and standard deviation on a microparticle batch characterized by the log normal distribution model. Such a file can be applied during manufacturing to explore tolerances and increase cost and time effectiveness. Theoretical data for the time to 60% drug release and the slope and intercept of the log-log plot were collected and subjected to statistical analysis in JMP. Since the scope of this project focuses on microparticle surface degradation drug release with no drug diffusion, the characteristic variables relating to the slope (n = diffusional release exponent) and the intercept (k = kinetic constant) do not directly apply to the distribution model within the scope of the research. However, these variables are useful for analysis when the Mathcad template is applied to other types of drug release models.
ContributorsHan, Priscilla (Author) / Vernon, Brent (Thesis director) / Nickle, Jacob (Committee member) / Harrington Bioengineering Program (Contributor, Contributor) / Barrett, The Honors College (Contributor)
Created2021-05
Description
Effective modeling of high dimensional data is crucial in information processing and machine learning. Classical subspace methods have been very effective in such applications. However, over the past few decades, there has been considerable research towards the development of new modeling paradigms that go beyond subspace methods. This dissertation focuses on the study of sparse models and their interplay with modern machine learning techniques such as manifold, ensemble and graph-based methods, along with their applications in image analysis and recovery. By considering graph relations between data samples while learning sparse models, graph-embedded codes can be obtained for use in unsupervised, supervised and semi-supervised problems. Using experiments on standard datasets, it is demonstrated that the codes obtained from the proposed methods outperform several baseline algorithms. In order to facilitate sparse learning with large scale data, the paradigm of ensemble sparse coding is proposed, and different strategies for constructing weak base models are developed. Experiments with image recovery and clustering demonstrate that these ensemble models perform better when compared to conventional sparse coding frameworks. When examples from the data manifold are available, manifold constraints can be incorporated with sparse models and two approaches are proposed to combine sparse coding with manifold projection. The improved performance of the proposed techniques in comparison to sparse coding approaches is demonstrated using several image recovery experiments. In addition to these approaches, it might be required in some applications to combine multiple sparse models with different regularizations. In particular, combining an unconstrained sparse model with non-negative sparse coding is important in image analysis, and it poses several algorithmic and theoretical challenges. A convex and an efficient greedy algorithm for recovering combined representations are proposed. Theoretical guarantees on sparsity thresholds for exact recovery using these algorithms are derived and recovery performance is also demonstrated using simulations on synthetic data. Finally, the problem of non-linear compressive sensing, where the measurement process is carried out in feature space obtained using non-linear transformations, is considered. An optimized non-linear measurement system is proposed, and improvements in recovery performance are demonstrated in comparison to using random measurements as well as optimized linear measurements.
ContributorsNatesan Ramamurthy, Karthikeyan (Author) / Spanias, Andreas (Thesis advisor) / Tsakalis, Konstantinos (Committee member) / Karam, Lina (Committee member) / Turaga, Pavan (Committee member) / Arizona State University (Publisher)
Created2013
Description
Image understanding has been playing an increasingly crucial role in vision applications. Sparse models form an important component in image understanding, since the statistics of natural images reveal the presence of sparse structure. Sparse methods lead to parsimonious models, in addition to being efficient for large scale learning. In sparse modeling, data is represented as a sparse linear combination of atoms from a "dictionary" matrix. This dissertation focuses on understanding different aspects of sparse learning, thereby enhancing the use of sparse methods by incorporating tools from machine learning. With the growing need to adapt models for large scale data, it is important to design dictionaries that can model the entire data space and not just the samples considered. By exploiting the relation of dictionary learning to 1-D subspace clustering, a multilevel dictionary learning algorithm is developed, and it is shown to outperform conventional sparse models in compressed recovery, and image denoising. Theoretical aspects of learning such as algorithmic stability and generalization are considered, and ensemble learning is incorporated for effective large scale learning. In addition to building strategies for efficiently implementing 1-D subspace clustering, a discriminative clustering approach is designed to estimate the unknown mixing process in blind source separation. By exploiting the non-linear relation between the image descriptors, and allowing the use of multiple features, sparse methods can be made more effective in recognition problems. The idea of multiple kernel sparse representations is developed, and algorithms for learning dictionaries in the feature space are presented. Using object recognition experiments on standard datasets it is shown that the proposed approaches outperform other sparse coding-based recognition frameworks. Furthermore, a segmentation technique based on multiple kernel sparse representations is developed, and successfully applied for automated brain tumor identification. Using sparse codes to define the relation between data samples can lead to a more robust graph embedding for unsupervised clustering. By performing discriminative embedding using sparse coding-based graphs, an algorithm for measuring the glomerular number in kidney MRI images is developed. Finally, approaches to build dictionaries for local sparse coding of image descriptors are presented, and applied to object recognition and image retrieval.
ContributorsJayaraman Thiagarajan, Jayaraman (Author) / Spanias, Andreas (Thesis advisor) / Frakes, David (Committee member) / Tepedelenlioğlu, Cihan (Committee member) / Turaga, Pavan (Committee member) / Arizona State University (Publisher)
Created2013
Description
Cardiac tissue engineering is an emerging field that has the potential to regenerate and repair damaged cardiac tissues after myocardial infarction. Numerous studies have introduced hydrogel-based cardiac tissue constructs featuring suitable microenvironments for cell growth along with precise surface topographies for directed cell organization. Despite significant progress, previously developed cardiac tissue constructs have suffered from electrically insulated matrices and low cell retention. To address these drawbacks, we fabricated micropatterned hybrid hydrogel constructs (uniaxial microgrooves with 50 µm with) using a photocrosslinkable gelatin methacrylate (GelMA) hydrogel incorporated with gold nanorods (GNRs). The electrical impedance results revealed a lower impedance in the GelMA-GNR constructs versus the pure GelMA constructs. Superior electrical conductivity of GelMA-GNR hydrogels (due to incorporation of GNRs) enabled the hybrid tissue constructs to be externally stimulated using a pulse generator. Furthermore, GelMA-GNR tissue hydrogels were tested to investigate the biological characteristics of cultured cardiomyocytes. The F-actin fiber analysis results (area coverage and alignment indices) revealed higher directed (uniaxial) cytoskeleton organization of cardiac cells cultured on the GelMA-GNR hydrogel constructs in comparison to pure GelMA. Considerable increase in the coverage area of cardiac-specific markers (sarcomeric α-actinin and connexin 43) were observed on the GelMA-GNR hybrid constructs compared to pure GelMA hydrogels. Despite substantial dissimilarities in cell organization, both pure GelMA and hybrid GelMA-GNR hydrogel constructs provided a suitable microenvironment for synchronous beating of cardiomyocytes.
ContributorsMoore, Nathan Allen (Author) / Nikkhah, Mehdi (Thesis director) / Smith, Barbara (Committee member) / Harrington Bioengineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2016-05
Description
Morphine is a commonly used analgesic in pain management. Opioid administration to a patient after surgery, such as spinal decompression surgery, can lead to adverse side effects. To demonstrate these adverse side effects could be decreased we created a model of how morphine and its metabolites are transported and excreted from the body. Using the of morphine and a standard compartment approach this thesis aimed at projecting pharmacokinetics trends of morphine overtime. A Matlab compartment model predicting the transport of morphine through the body can contribute to a better understanding of the concentrations at the systemic level, specifically with respect to a CSF, and what happens when you compare an intravenous injection to a local delivery. Other studies and models commonly utilized patient data over small periods of time2,3,5. An extended period of time will provide information into morphine’s time course after surgery. This model focuses on a compartmentalization of the major organs and the use of a simple Mechalis-Menten enzyme kinetics for the metabolites in the liver. Our results show a CSF concentration of about 1.086×〖10〗^(-12) nmol/L in 6 weeks and 1.0097×〖10〗^(-12) nmol/L in 12 weeks. The concentration profiles in this model are similar to what was expected. The implications of this suggest that patients who reported effects of morphine paste, a locally administered opioid, weeks after the surgery were due to other reasons. In creating a model we can determine important variables and dosage information. This information allows for a greater understanding of what is happening in the body and how to improve surgical outcomes. We propose this study has implications in general research in the pharmacokinetics and dynamics of pharmacology through the body.
ContributorsJacobs, Danielle Renee (Author) / Caplan, Michael (Thesis director) / Giers, Morgan (Committee member) / Barrett, The Honors College (Contributor) / Harrington Bioengineering Program (Contributor)
Created2014-05
Description
There is an increasing interest in developing thermo-responsive polymers for treating aneurysms. In this thesis project, the potential for poly(NIPAAm-co-JAAm-co-HEMA-Acrylate) (PNJHAc) as a treatment method for brain aneurysms was investigated. Five different batches of polymer were synthesized, purified, lyophilized, and characterized using nuclear magnetic resonance and cloud point techniques over the course of several months. Two were tested in aneurysm models. Of these five batches, there were two that showed promise as liquid embolic agents for endovascular embolization.
ContributorsLoui, Michelle (Author) / Vernon, Brent (Thesis director) / Pal, Amrita (Committee member) / Harrington Bioengineering Program (Contributor, Contributor) / Barrett, The Honors College (Contributor)
Created2020-05
Description
Previous studies have found that the detection of near-threshold stimuli is decreased immediately before movement and throughout movement production. This has been suggested to occur through the use of the internal forward model processing an efferent copy of the motor command and creating a prediction that is used to cancel out the resulting sensory feedback. Currently, there are no published accounts of the perception of tactile signals for motor tasks and contexts related to the lips during both speech planning and production. In this study, we measured the responsiveness of the somatosensory system during speech planning using light electrical stimulation below the lower lip by comparing perception during mixed speaking and silent reading conditions. Participants were asked to judge whether a constant near-threshold electrical stimulation (subject-specific intensity, 85% detected at rest) was present during different time points relative to an initial visual cue. In the speaking condition, participants overtly produced target words shown on a computer monitor. In the reading condition, participants read the same target words silently to themselves without any movement or sound. We found that detection of the stimulus was attenuated during speaking conditions while remaining at a constant level close to the perceptual threshold throughout the silent reading condition. Perceptual modulation was most intense during speech production and showed some attenuation just prior to speech production during the planning period of speech. This demonstrates that there is a significant decrease in the responsiveness of the somatosensory system during speech production as well as milliseconds before speech is even produced which has implications for speech disorders such as stuttering and schizophrenia with pronounced deficits in the somatosensory system.
ContributorsMcguffin, Brianna Jean (Author) / Daliri, Ayoub (Thesis director) / Liss, Julie (Committee member) / Department of Psychology (Contributor) / School of Life Sciences (Contributor) / Barrett, The Honors College (Contributor)
Created2019-05
Description
Human potential is characterized by our ability to think flexibly and develop novel solutions to problems. In cognitive neuroscience, problem solving is studied using various tasks. For example, IQ can be tested using the RAVEN, which measures abstract reasoning. Analytical problem solving can be tested using algebra, and insight can be tested using a nine-dot test. Our class of problem-solving tasks blends analytical and insight processes. This can be done by measuring multiply-constrained problem solving (MCPS). MCPS occurs when an individual problem has several solutions, but when grouped with simultaneous problems only one correct solution presents itself. The most common test for MCPS is known at the CRAT, or compound remote associate task. For example, when given the three target words “water, skate, and cream” there are many compound associates that can be assigned each of the target words individually (i.e. salt-water, roller-skate, whipped-cream), but only one that works with all three (ice-water, ice-skate, ice-cream).
This thesis is a tutorial for a MATLAB user-interface, known as EEGLAB. Cognitive and neural correlates of analytical and insight processes were evaluated and analyzed in the CRAT using EEG. It was hypothesized that different EEG signals will be measured for analytical versus insight problem solving, primarily observed in the gamma wave production. The data was interpreted using EEGLAB, which allows psychological processes to be quantified based on physiological response. I have written a tutorial showing how to process the EEG signal through filtering, extracting epochs, artifact detection, independent component analysis, and the production of a time – frequency plot. This project has combined my interest in psychology with my knowledge of engineering and expand my knowledge of bioinstrumentation.
This thesis is a tutorial for a MATLAB user-interface, known as EEGLAB. Cognitive and neural correlates of analytical and insight processes were evaluated and analyzed in the CRAT using EEG. It was hypothesized that different EEG signals will be measured for analytical versus insight problem solving, primarily observed in the gamma wave production. The data was interpreted using EEGLAB, which allows psychological processes to be quantified based on physiological response. I have written a tutorial showing how to process the EEG signal through filtering, extracting epochs, artifact detection, independent component analysis, and the production of a time – frequency plot. This project has combined my interest in psychology with my knowledge of engineering and expand my knowledge of bioinstrumentation.
ContributorsCobban, Morgan Elizabeth (Author) / Brewer, Gene (Thesis director) / Ellis, Derek (Committee member) / Harrington Bioengineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2019-05
Description
Traumatic brain injury (TBI) is a major concern in public health due to its prevalence and effect. Every year, about 1.7 million TBIs are reported [7]. According to the According to the Centers for Disease Control and Prevention (CDC), 5.5% of all emergency department visits, hospitalizations, and deaths from 2002 to 2006 are due to TBI [8]. The brain's natural defense, the Blood Brain Barrier (BBB), prevents the entry of most substances into the brain through the blood stream, including medicines administered to treat TBI [11]. TBI may cause the breakdown of the BBB, and may result in increased permeability, providing an opportunity for NPs to enter the brain [3,4]. Dr. Stabenfeldt's lab has previously established that intravenously injected nanoparticles (NP) will accumulate near the injury site after focal brain injury [4]. The current project focuses on confirmation of the accumulation or extravasation of NPs after brain injury using 2-photon microscopy. Specifically, the project used controlled cortical impact injury induced mice models that were intravenously injected with 40nm NPs post-injury. The MATLAB code seeks to analyze the brain images through registration, segmentation, and intensity measurement and evaluate if fluorescent NPs will accumulate in the extravascular tissue of injured mice models. The code was developed with 2D bicubic interpolation, subpixel image registration, drawn dimension segmentation and fixed dimension segmentation, and dynamic image analysis. A statistical difference was found between the extravascular tissue of injured and uninjured mouse models. This statistical difference proves that the NPs do extravasate through the permeable cranial blood vessels in injured cranial tissue.
ContributorsIrwin, Jacob Aleksandr (Author) / Stabenfeldt, Sarah (Thesis director) / Bharadwaj, Vimala (Committee member) / Harrington Bioengineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2018-05