Theses and Dissertations
This collection includes both ASU Theses and Dissertations, submitted by graduate students, and the Barrett, Honors College theses submitted by undergraduate students.
Displaying 1 - 10 of 11
Filtering by
- Creators: Barrett, The Honors College
- Creators: Turaga, Pavan
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
Recent advances in camera architectures and associated mathematical representations now enable compressive acquisition of images and videos at low data-rates. While most computer vision applications of today are composed of conventional cameras, which collect a large amount redundant data and power hungry embedded systems, which compress the collected data for further processing, compressive cameras offer the advantage of direct acquisition of data in compressed domain and hence readily promise to find applicability in computer vision, particularly in environments hampered by limited communication bandwidths. However, despite the significant progress in theory and methods of compressive sensing, little headway has been made in developing systems for such applications by exploiting the merits of compressive sensing. In such a setting, we consider the problem of activity recognition, which is an important inference problem in many security and surveillance applications. Since all successful activity recognition systems involve detection of human, followed by recognition, a potential fully functioning system motivated by compressive camera would involve the tracking of human, which requires the reconstruction of atleast the initial few frames to detect the human. Once the human is tracked, the recognition part of the system requires only the features to be extracted from the tracked sequences, which can be the reconstructed images or the compressed measurements of such sequences. However, it is desirable in resource constrained environments that these features be extracted from the compressive measurements without reconstruction. Motivated by this, in this thesis, we propose a framework for understanding activities as a non-linear dynamical system, and propose a robust, generalizable feature that can be extracted directly from the compressed measurements without reconstructing the original video frames. The proposed feature is termed recurrence texture and is motivated from recurrence analysis of non-linear dynamical systems. We show that it is possible to obtain discriminative features directly from the compressed stream and show its utility in recognition of activities at very low data rates.
ContributorsKulkarni, Kuldeep Sharad (Author) / Turaga, Pavan (Thesis advisor) / Spanias, Andreas (Committee member) / Frakes, David (Committee member) / Arizona State University (Publisher)
Created2012
Description
Diabetic retinopathy (DR) is a common cause of blindness occurring due to prolonged presence of diabetes. The risk of developing DR or having the disease progress is increasing over time. Despite advances in diabetes care over the years, DR remains a vision-threatening complication and one of the leading causes of blindness among American adults. Recent studies have shown that diagnosis based on digital retinal imaging has potential benefits over traditional face-to-face evaluation. Yet there is a dearth of computer-based systems that can match the level of performance achieved by ophthalmologists. This thesis takes a fresh perspective in developing a computer-based system aimed at improving diagnosis of DR images. These images are categorized into three classes according to their severity level. The proposed approach explores effective methods to classify new images and retrieve clinically-relevant images from a database with prior diagnosis information associated with them. Retrieval provides a novel way to utilize the vast knowledge in the archives of previously-diagnosed DR images and thereby improve a clinician's performance while classification can safely reduce the burden on DR screening programs and possibly achieve higher detection accuracy than human experts. To solve the three-class retrieval and classification problem, the approach uses a multi-class multiple-instance medical image retrieval framework that makes use of spectrally tuned color correlogram and steerable Gaussian filter response features. The results show better retrieval and classification performances than prior-art methods and are also observed to be of clinical and visual relevance.
ContributorsChandakkar, Parag Shridhar (Author) / Li, Baoxin (Thesis advisor) / Turaga, Pavan (Committee member) / Frakes, David (Committee member) / Arizona State University (Publisher)
Created2012
Description
The ability of cochlear implants (CI) to restore auditory function has advanced significantly in the past decade. Approximately 96,000 people in the United States benefit from these devices, which by the generation and transmission of electrical impulses, enable the brain to perceive sound. But due to the predominantly Western cochlear implant market, current CI characterization primarily focuses on improving the quality of American English. Only recently has research begun to evaluate CI performance using other languages such as Mandarin Chinese, which rely on distinct spectral characteristics not present in English. Mandarin, a tonal language utilizes four, distinct pitch patterns, which when voiced a syllable, conveys different meanings for the same word. This presents a challenge to hearing research as spectral, or frequency based information like pitch is readily acknowledged to be significantly reduced by CI processing algorithms. Thus the present study sought to identify the intelligibility differences for English and Mandarin when processed using current CI strategies. The objective of the study was to pinpoint any notable discrepancies in speech recognition, using voice-coded (vocoded) audio that simulates a CI generated stimuli. This approach allowed 12 normal hearing English speakers, and 9 normal hearing Mandarin listeners to participate in the experiment. The number of frequency channels available and the carrier type of excitation were varied in order to compare their effects on two cases of Mandarin intelligibility: Case 1) word recognition and Case 2) combined word and tone recognition. The results indicated a statistically significant difference between English and Mandarin intelligibility for Condition 1 (8Ch-Sinewave Carrier, p=0.022) given Case 1 and Condition 1 (8Ch-Sinewave Carrier, p=0.001) and Condition 3 (16Ch-Sinewave Carrier, p=0.001) given Case 2. The data suggests that the nature of the carrier type does have an effect on tonal language intelligibility and warrants further research as a design consideration for future cochlear implants.
ContributorsSchiltz, Jessica Hammitt (Author) / Berisha, Visar (Thesis director) / Frakes, David (Committee member) / Barrett, The Honors College (Contributor) / Harrington Bioengineering Program (Contributor)
Created2015-05
Description
Cerebral aneurysms, also known as intracranial aneurysms, are sac-like lesions in the arteries of the brain that can rupture to cause subarachnoid hemorrhaging, damaging and killing brain cells. Metal coil embolization has been traditionally used to occlude and treat cerebral aneurysms to limited success, but polymer embolization has been suggested, because it can provide a greater fraction of occlusion. One such polymer with low cytotoxicity is poly(propylene glycol)diacrylate (PPODA) crosslinked via Michael-type addition with pentaerythritol tetrakis(3-mercaptopropionate) (QT). This study was performed to examine the behavior of PPODA-QT gel in vitro under pulsatile flow emulating physiological conditions. An idealized cerebral aneurysm flow model was designed based on geometries associated with an increase in rupture risk. Pressure was monitored at the apex of the aneurysm dome for varied flow rates and polymer filling fractions of 32.4, 78.2, and 100%. The results indicate that the amount of PPODA-QT deployed into the aneurysm decreases the peak-to-peak oscillation in pressure at the aneurysm wall by an inverse proportion. The 32.4 and 78.2% treatments did not significantly decrease the mean pressure applied to the aneurysm dome, but the 100% treatment greatly reduced it by diverting flow. This study indicates that the maximum filling fraction after swelling of PPODA-QT polymer should be deployed into the aneurysmal sac for treatment.
ContributorsWorkman, Christopher David (Author) / Vernon, Brent (Thesis director) / Frakes, David (Committee member) / Barrett, The Honors College (Contributor) / Department of Chemistry and Biochemistry (Contributor) / Harrington Bioengineering Program (Contributor)
Created2014-05
Description
This paper summarizes the [1] ideas behind, [2] needs, [3] development, and [4] testing of 3D-printed sensor-stents known as Stentzors. This sensor was successfully developed entirely from scratch, tested, and was found to have an output of 3.2*10-6 volts per RMS pressure in pascals. This paper also recommends further work to render the Stentzor deployable in live subjects, including [1] further design optimization, [2] electrical isolation, [3] wireless data transmission, and [4] testing for aneurysm prevention.
ContributorsMeidinger, Aaron Michael (Author) / LaBelle, Jeffrey (Thesis director) / Frakes, David (Committee member) / Barrett, The Honors College (Contributor) / Mechanical and Aerospace Engineering Program (Contributor)
Created2014-05
Description
Glioblastoma Multiforme (GBM) is an aggressive and deadly form of brain cancer with a median survival time of about a year with treatment. Due to the aggressive nature of these tumors and the tendency of gliomas to follow white matter tracks in the brain, each tumor mass has a unique growth pattern. Consequently it is difficult for neurosurgeons to anticipate where the tumor will spread in the brain, making treatment planning difficult. Archival patient data including MRI scans depicting the progress of tumors have been helpful in developing a model to predict Glioblastoma proliferation, but limited scans per patient make the tumor growth rate difficult to determine. Furthermore, patient treatment between scan points can significantly compound the challenge of accurately predicting the tumor growth. A partnership with Barrow Neurological Institute has allowed murine studies to be conducted in order to closely observe tumor growth and potentially improve the current model to more closely resemble intermittent stages of GBM growth without treatment effects.
ContributorsSnyder, Lena Haley (Author) / Kostelich, Eric (Thesis director) / Frakes, David (Committee member) / Barrett, The Honors College (Contributor) / School of Mathematical and Statistical Sciences (Contributor) / Harrington Bioengineering Program (Contributor)
Created2014-05
Customized Endovascular Treatment Fixture Construction for Patient-Specific Cerebral Aneurysm Repair
Description
Cerebral aneurysms are pathological bulges in blood vessels of the brain that can rupture and cause brain damage or death. Treating aneurysms by isolating them from circulation can prevent aneurysm rupture. Endovascular techniques for cerebral aneurysm treatment are preferred because they are minimally invasive and have a shorter recovery time, and endovascular coiling is considered the gold standard as a result. The coils used in endovascular treatment come in standard shapes and sizes, mass-manufactured by medical device companies. Clinicians select the coils for treatment based on the aneurysm volume. However, cerebral aneurysms have unique shapes and dimensions, and vary on a patient-specific basis. Therefore, customizing the coils to fit a unique aneurysm morphology by using shape memory alloys could potentially improve endovascular treatment outcomes. In order to shape set a shape memory alloy into a customized coil configuration a fixture based on the aneurysm morphology must first be developed. Digital surface models of aneurysm patient cases were collected from an online repository and isolated from surrounding vasculature. Anchors used to assist in winding coils around these models were then added to create a computational fixture model. These fixtures were 3D printed in stainless steel, and tested on their ability to maintain their shape after being exposed to high temperatures needed in shape setting processes. The study demonstrated that customized fixtures can be created from patient-specific images or models, and manufactured with high levels of accuracy without deformation at high temperatures. The results suggest that 3D printed stainless steel fixtures could be used to develop customized endovascular coils for cerebral aneurysm treatment.
ContributorsHess, Ryan Ambrose (Author) / Kleim, Jeff (Thesis director) / Nair, Priya (Committee member) / Frakes, David (Committee member) / Barrett, The Honors College (Contributor)
Created2017-05
Description
Development of a rapid and label-free Electrochemical Impedance Spectroscopy (EIS) biosensor for Cardiovascular Disease (CVD) detection based on Inerluekin-18 (IL-18) sensitivity was proposed to fill the technology gap between rapid and portable CVD point-of-care diagnosis. IL-18 was chosen for this CVD biosensor due to its ability to detect plaque vulnerability of the heart. Custom (hand) made sensors, which utilized a three electrode configuration with a gold disk working electrode, were created to run EIS using both IL-18 and anti-IL-18 molecules in both purified and blood solutions. The EIS results for IL-18 indicated the optimal detection frequency to be 371Hz. Blood interaction on the working electrode increased the dynamic range of impedance values for the biosensor. Future work includes Developing and testing prototypes of the biosensor along with determining if a Nafion based coating on the working electrode will reduce the dynamic range of impedance values caused by blood interference.
ContributorsJha, Amit (Author) / LaBelle, Jeffrey (Thesis director) / Mossman, Kenneth (Committee member) / Frakes, David (Committee member) / Barrett, The Honors College (Contributor) / Harrington Bioengineering Program (Contributor) / Department of Management (Contributor)
Created2013-05
Description
This study investigates the application of Computational Fluid Dynamics (CFD) to the medical field. An overview of recent advances in computational simulation and modeling in medical applications is provided, with a particular emphasis on CFD. This study attempts to validate CFD and demonstrate the possibility for applying CFD to the clinical treatment and evaluation of atherosclerotic disease. Three different geometric configurations are investigated: one idealized bifurcation with a primary diameter of 8 mm, and two different patient-specific models of the bifurcation from the common femoral artery to the superficial and deep femoral arteries. CFD is compared against experimental measurements of steady state and pulsatile flow acquired with Particle Image Velocimetry (PIV). Steady state and pulsatile flow rates that are consistent with those observed in the femoral artery are used. In addition, pulsatile CFD simulations are analyzed in order to demonstrate meaningful clinical applications for studying and evaluating the treatment of atherosclerotic disease. CFD was successfully validated for steady state flow, with an average percent error of 6.991%. Potential for validation was also demonstrated for pulsatile flow, but methodological errors warrant further investigation to reformulate methods and analyze results. Quantities frequently associated with atherosclerotic disease and arterial bifurcations, such as large variations in wall shear stress and the presence of recirculation zones are demonstrated from the pulsatile CFD simulations. Further study is required in order to evaluate whether or not such phenomena are represented by CFD accurately. Further study must also be performed in order to evaluate the practicality and utility of CFD for the evaluation of atherosclerotic disease treatment.
ContributorsMortensen, Matthew James (Author) / VanAuker, Michael (Thesis director) / Frakes, David (Committee member) / Harrington Bioengineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2016-05