16484 research outputs found
Sort by
On the Detection of Malware on Virtual Assistants Based on Behavioral Anomalies
No full textThe Internet of Things (IoT) refers to the growing network of ``smart objects." The increase in popularity of IoT devices, due to their efficiency and convenience, has given rise to new security concerns. The variety and novelty of IoT devices provide a corpus of malware that is of insufficient size to employ classic machine learning algorithms. This makes anomaly detection methods for IoT device security more attractive, especially in the short term, until there are enough behavioral signatures for malware to train more sophisticated machine learning detection models for these devices. This thesis explores some of the security concerns pertaining to running software similar to Amazon Alexa home assistant on IoT-like platforms. We implement a behavioral-based malware detector and compare the effectiveness of different system attributes that are used in detecting malware, i.e., system calls, network traffic, and the integration of system call and network traffic features. Given the small number of malware samples for IoT devices, we create a parameterizable malware sample that mimics Alexa behavior in varying degrees, while exfiltrating data from the device to a remote host. The performance of our anomaly detector is evaluated based on how well it determines the presence of our parameterized malware on an Alexa-enabled IoT device.M.S., Electrical Engineering -- Drexel University, 201
A Study of the Relationships between Early Onset Scoliosis, Thoracic Deformity, and Pulmonary Function
No full textEarly Onset Scoliosis (EOS) is an idiopathic disorder diagnosed in children under the age of ten years that is characterized by a three- dimensional deformity of the spine and rib cage. Thoracic insufficiency syndrome (TIS), commonly diagnosed in patients with EOS, describes how the physical deformity causes lung dysfunction and impaired lung growth. It has previously been hypothesized that surgical correction of spinal deformities can reverse pulmonary dysfunction, although measuring lung function via pulmonary function testing (PFT) in infants is quite difficult and not routinely done. In this study, we sought to further describe the relationships between EOS, thoracic deformity, and pulmonary function in three ways. First, we evaluated the pre- and postoperative differences in thoracic deformity of EOS patients. Second, we described the relationships between lung volumes calculated from CT scans (which are routinely collected for scoliosis patients) with pulmonary function measures collected clinically. Third, we developed multivariable models from thoracic deformity parameters that could be used to predict pulmonary function. Overall, these goals were all met. Pre- and postoperative differences were seen in four thoracic deformity parameters (Haller, kyphosis-lordosis, pectus, and frontosagittal index), with the decrease in frontosagittal index being similar to the trend seen in normal patients over time. Additionally, computational lung volumes correlated strongly with five clinical measures of pulmonary function (residual volume, functional residual capacity, total lung capacity, forced vital capacity, and forced expiratory volume). Lastly, clinically relevant multivariable models were created for predicting five pulmonary function measures (forced expiratory volume, forced vital capacity, forced residual capacity, total lung capacity, and residual volume) from four predictor thoracic deformity parameters (Haller index, posterior hemithoracic symmetry ratio, rib hump index, and sternovertebral distance). These methods were limited by sample size and subjectivity of the calculations, so future work should look to increase the sample size and automate the calculation process. Overall, these results have the ability to guide surgical planning and provide clinicians with a comprehensive understanding of an EOS patient's specific lung function.M.S., Biomedical Engineering -- Drexel University, 201
A Dynamic Data-Driven Framework for Damage Predictions in Materials & Structures
No full textMonitoring real-time failure in fiber-reinforced composites using nondestructive evaluation (NDE) is a fairly difficult task due to the associated evolving material state uncertainties, as well as due to data management mining and signal processing challenges. In this context, a number of statistical, probabilistic and physics-based models have been proposed to make predictions of remaining useful life (RUL) in the built environment. In parallel to modeling efforts, the data quality from sensing, NDE and testing methods is compromised by a variety of intrinsic (hardware) and extrinsic (noise) factors that make the damage assessment process both challenging to visualize and computationally expensive to analyze. Most importantly, the uncertainties in the recorded data, also hinder efforts to create data-driven methods in the framework of what is currently called digital twin modeling. To address such issues more efficient data workflows and processing procedures are currently sought to assist with both damage monitoring assessment, as well as modeling and predictions. The objective, therefore, of this manuscript is to present a novel data-driven probabilistic modeling methodology that is capable of producing RUL estimates for composites. The framework consists of two parts: the hardware/software integration that allows the establishment of data streaming and handling procedures that ultimately feed in real time the probabilistic approach. The proposed probabilistic modeling approach is based on the use of an outlier analysis combined with information quality metrics to identify a given set of degradation states. Predictions of RUL are then made by extracting and processing features from NDE datasets which are then used in unsupervised clustering. Such information is subsequently leveraged to train a support vector machine (SVM) alongside a hidden Markov chain model (HMM). The signal classifications from SVM combined with the HMM are then used as inputs for an adaptive neuro-fuzzy network that produces the RUL predictions. An application of the proposed approach in mechanical testing experiments of an aerospace-grade composite material is presented. Extensions that can make this approach applicable to industrial applications as well as in digital twin implementations are discussed.M.S., Mechanical Engineering and Mechanics -- Drexel University, 201
Early prediction of pressure injury development using diffuse correlation spectroscopy
No full textPressure injuries (PIs) are a serious secondary complication for patients with prolonged immobility that are often caused by prolonged ischemia and reperfusion injury. The cost to manage a single full-thickness pressure injury is estimated to be as high as $120,000 and hospitals are not reimbursed for treatment of PIs that are not documented as "present on admission." Detection of early pressure injuries is currently performed through visual inspection. However, these methods are inadequate because they only evaluate the surface of the skin and provide no information to the underlying health of deep tissue, where PIs often manifest. For this reason, deep tissue pressure injury (DTPI) is often not evident until the injury has evolved to an advanced PI and tissue breakdown. Therefore, there is a well-defined need to develop a method to noninvasively examine tissue below the surface of the skin to identify deep tissue pressure injury earlier than it can be detected by current methods. The goals of this research were to enable early detection of PI in high risk patients and to develop a procedure capable of differentiating between patients who develop advanced pressure injuries and those who do not. Diffuse correlation spectroscopy (DCS), a noninvasive optical method, was used to measure microcirculatory blood flow in skin and subcutaneous tissue. Data was collected from the DCS device and analyzed in the form of τ_exp, which is a value calculated directly from the measured temporal correlation function (TCF) of scattered light intensity. In addition a multi-distance DCS algorithm was developed in order to calculate the optical properties from multiple TCFs, thereby providing more information on the health of underlying tissue. This method was validated in vitro through the use of optical phantoms and later applied to human data. The DCS device was first tested in a clinical study that included 16 spinal cord injury patients with erythema in the sacrum from a rehabilitation hospital. Patients were measured in a three-step protocol to monitor blood flow during baseline, applied pressure, and released pressure stages. Four of the 16 patients developed an open ulcer (Advanced PIs) while the other 12 patients did not develop an open ulcer (No PIs). The baseline results showed that Advanced PIs had τ_exp values 7-8 times lower than No PIs, suggesting Advanced PIs had significantly faster blood flow than that of No PIs. Patients from an acute care setting were also measured, including individuals in the surgical intensive care unit (SICU), individuals in the trauma and surgical step-down units, and surgical patients in the post-anesthesia care unit (PACU). During these studies, several modifications were made to the device in order to increase the efficiency of the protocol and usability for a layperson. The results from these studies showed the same trends found from the rehab study. The patient who developed a PI from the SICU had an average τ_exp value about two times lower than that of other SICU patients who did not develop a PI, suggesting faster blood flow. Also the patient who developed a PI from the surgical study had a τ_exp value ~2.5 times lower post-op compared to pre-op, the highest ratio among all patients. Furthermore, when applying the multi-distance DCS algorithm to the data collected from surgical patients and showed a difference of 6.5 times between post-op and pre-op blood flow for the patient who developed an Advanced PI. The successful implementation of DCS technology as a pressure injury assessment method would result in improved and personalized patient care. It would permit clinicians to immediately assess the severity and extent of a subcutaneous injury and change the preventive treatment, if needed. It would also decrease morbidity and minimize the hospital's liability to patients who develop PIs in their care.Ph.D., Biomedical Engineering -- Drexel University, 201
Structural Characterization and Introduction of Biomimetic Proteoglycans in Temporomandibular Joint Pain
No full textTemporomandibular Joint disorders (TMJD) are the second-most common source of orofacial pain with 33% of adults having at least one TMJ disorder symptom. A significant subset of patients with TMJD develop osteoarthritis (OA), a progressive pathology characterized by intraarticular inflammation and fibrocartilage degeneration. Specific indications of osteoarthritic development are reorganization of collagen and loss of proteoglycans such as aggrecan. Loss of proteoglycans cause a decrease in tissue hydration, joint lubrication, and compromised compressive mechanics. Recent research suggests these structural changes are identifiable early within the pericellular matrix of the tissue. The goal of this thesis was to evaluate the use of a novel biomimetic proteoglycan (BPG) as a treatment during early stages of TMJ osteoarthritis. BPGs have similar composition and properties to naturally occurring proteoglycans, but resist enzymatic degradation associated with hostile, osteoarthritic tissue environments. We initially established there is pericellular reorganization of collagen VI and aggrecan in joint overloading rat models of tunable TMJ pain, with more robust structural outcomes in joints exposed to 3.5N, resembling chronic pain, versus those exposed to 2N, representing acute pain. This established motivation to introduce BPG to this animal model. With ex-vivo diffusion of the biomimetic into TMJs from the acute and chronic pain models, we characterized BPG distribution and localization within the tissue matrix and found modulation between non-loaded, 2N, and 3.5N loaded TMJs. Further, we identified a functional role of BPGs to influence matrix mechanics of TMJ fibrocartilage. Together, these findings bridge the inflammatory and catabolic cascades of TMJ pain to tissue structural outcomes and provide a first look into the use of biomimetic proteoglycans as a way restore proteoglycans lost during early TMJ osteoarthritis known to induce associated symptomatic pain.M.S., Biomedical Engineering -- Drexel University, 201
The Voices of Leaders: A Qualitative Examination of Urban Principals' Perspectives Regarding the Reintegrating of Students with an Emotional Disturbance Back into the School Community
No full textSince the establishment of the Individuals with Disabilities Education Act (IDEA) of 1990, providing the Least Restrictive Environment (LRE) for students with an emotional disturbance (ED) have been an area of confusion and contention. Likewise, reintegrating students with ED back into the school community from an alternative placement has also been a major challenge for school teams. As a result, students face significant challenges when they return to a traditional school setting. The purpose of this phenomenological qualitative research was to understand the perceptions and lived experiences of urban high school principals as it relates to the reintegration of students from an alternative placement such as approved private schools, residential treatment facilities, and juvenile facilities back to their neighborhood public school. The research was conducted in an effort to gain insight into the planning and support, or the lack thereof, for students with ED. In addition, the research focused on uncovering best practices, barriers and challenges, and components for successful reintegration of students with ED. Through interviews and reflective journal responses related to the research questions, the study revealed a need for additional central office support, training, formalized reintegration meetings, comprehensive transition plans, and the development of reintegration-transition programs.Ed.D., Educational Leadership and Management -- Drexel University, 201
Advanced Solver Development for Large-Scale Dynamic Building System Simulation
No full textEfficiently, robustly and accurately solving large and sparse nonlinear algebraic and differential equation system for dynamic building simulation is becoming more and more essential due to increasing demands to simulate large-scale problems for multiple buildings coupled with various levels of strength either through the smart grid or other means, such as district heating/cooling and shared distributed energy resources. This study is interested in advancing solving techniques that either improve the quality and efficiency of a dynamic building simulation model generically or improve the performance of the underlying equation solver. Nowadays, many commonly used tools for dynamic building system simulation still employ direct Newton methods. These methods are not only lack of convergence for stiff problems or cold starts, but also fail to meet the increased memory requirements associated with large-scale problems or more specific issues that arise in problems where the nonlinear equations resulted from the discretization of an underlying engineering differential equation. Therefore, a Newton-Krylov method that satisfies the computational need for large-scale dynamic building system simulation is investigated. An ideal preconditioner and an automatic update scheme are employed to ensure fast and robust simulation by way of the Newton-Krylov method. In addition to the comparison study focuses on the numerical solution methods, a generic function smoothing technique for the rare occasion that discontinuous functions are encountered is also investigated. Four testbeds, namely, 4Z5B, 4Z1B, 12Z5B, and 40Z5B, are developed in an HVACSIM+ environment to evaluate the advancement techniques. All testbeds simulate the airflow and thermal behaviour of building zones (from four zones, 4Z, to forty zones, 40Z) that are served by air handling unit (AHU) and variable air volume (VAV) systems. 4Z5B and 4Z1B testbeds simulate the same building system with the same number of equations but with different equation groupings while 4Z5B. 12Z5B and 40Z5B testbeds have the same equation grouping but are corresponding to very different building system sizes (four, twelve, and forty zones, respectively) and therefore different numbers of equations to be solved. The following tasks are completed and summarized in this report: (1) Develop numerical testbeds to evaluate solution methods and techniques. (2) Investigate potential numerical issues in a typical dynamic building system simulation model and seek generic techniques to improve the quality of the model. (3) Examine the performance of a Newton-Krylov method on solving dynamic building system simulation equations. (4) Improve the performance of the Newton-Krylov method by developing and employing proper preconditioning techniques. (5) Investigate potential strategies to construct physics-based preconditioners. (6) Investigate the impact of finite difference step size in Jacobian approximation on the performance of dynamic building system simulation. The major numerical issue found in the testbeds mentioned above is the discontinuity of the simple coil component model. A generic smoothing technique is employed to improve the performance of the discontinuous simple coil component model, and the smoothed model results in a more stable and more accurate solution. A Newton-Krylov method is employed to increase the computational speed of a large-scale simulation. However, the direct implementation of the Newton-Krylov method results in stability issues. Therefore, a preconditioned Newton-Krylov method that employs the ideal preconditioner and an automatic update scheme is developed in this study, referred to as INB-PSGMRES(m). This method performs as robust as the default Powell's Hybrid (PH) method in HVACSIM+ while saving a significant amount of computational time. Its computational time saving against the PH method is at least 49.7%, 91.8%, 88.7%, and 97.1% for 4Z5B, 4Z1B, 12Z5B, and 40Z5B testbeds, respectively. It is found that because of the employment of preconditioning, two important parameters of the INB-PSGMRES(m) method, i.e., the forcing term and the restarting parameter, have little impact on the simulation performance. A few potential partitioning strategies for developing a physics-based preconditioner are investigated. Due to the strong coupling of mass flow rates and pressures between each nodal point of the airflow network system, it is difficult to construct an effective physics-based preconditioner for the airflow network of an AHU-VAV system. On the other hand, the thermal network can be effectively exploited. A preconditioner that targets the coil related equations is found effective at reducing the condition number of the Jacobian (which typically leads to fast linear convergence in a Krylov method) due to the high nonlinearity of the coil component model and its strong impact on the temperature and humidity in the HVAC system. Four finite difference step sizes for the Jacobian approximation and four finite difference step sizes for the Jacobian-vector approximation are investigated. For the Jacobian approximation, the current finite difference step size employed by HVACSIM+ is effective for the operating period. Its performance can be improved for the nonoperating period by adding a lower bound to the finite difference step size.Ph.D., Architectural Engineering -- Drexel University, 201
Investigation of light propagation and detection in human head under healthy and clinical settings
No full textNear infrared spectroscopy (NIRS) is a neuroimaging modality that allows investigation of brain tissue oxygenation non-invasively. It is widely used to measure changes in the concentration of oxy-hemoglobin and deoxy-hemoglobin in tissue. Infrared light emitted from a source placed over scalp propagates through the tissue and eventually part of it is back-scattered and can be collected by a photodetector. The attenuated light received at the detector encodes the information about brain activity as a consequence of absorption and scattering dominated light tissue interaction. Understanding and modeling light tissue interaction is critical for developing next generation NIRS systems. Several photon migration models have been proposed to investigate light tissue interaction through computerized Monte Carlo (MC) simulations. Using these, a set of NIRS system parameters have already been explored, such as wavelength selection, source-detector separation (SDS), depth of penetration, and effect of layers' thickness. Among those simulation studies, most have not declared the detector or fiber size clearly, also the selection of core system parameters remains controversial, like SD separation. More importantly, all these studies were performed only under healthy settings, no clinical conditions were taken into consideration. With numerous applications of NIRS technology in the assessment of brain function under various clinical conditions caused by traumatic brain injury (TBI) or stroke indicate the importance of study and evaluation of light tissue interaction under such conditions. In this thesis, we developed a reconfigurable and adaptive digital head model for healthy and clinical conditions that can be used to study diverse NIRS parameters for optimization. The thesis provided several novel contributions to the knowledge base that can further optical neuroimaging research applications, technology and algorithm development. First, it investigated new sensor parameters within digital head phantom, such as detector surface area and SDS, which are potential sources of systematic error in calculating hemoglobin concentrations. Secondly, several clinical conditions such as cerebral hematoma and edema development were modeled in silico, their effect on optical parameters and NIRS measurements were demonstrated with modeling for the first time. Such modeling and evaluation of neurological conditions and their effect on optical parameters and measurements can further help in the development of advanced algorithms for NIRS to provide more accurate hematoma and edema detection. Furthermore, virtual measurements from MC simulation on head models for different age groups were extracted and compared to actual measurements on equivalent physical models. The findings of this research can be used to optimize NIRS sensors and provide guidance for the design of next generation optical brain imaging systems for the monitoring of brain activity under healthy and clinical conditions.Ph.D., Biomedical Engineering -- Drexel University, 201
Effect of Photochemical and Mechanical Degradation Mechanisms on Polyethylene
No full textAs polyethylene (PE) products become more prevalent in a wide variety of engineering applications, it is critical to understand their chemical and mechanical degradation to prevent premature failure. Sunlight radiation has been recognized as the most severe environmental factor causing chemical degradation in PE. Furthermore, PE is also susceptible to mechanical failure through creep rupture and stress cracking. This dissertation focuses on the impact of these degradation mechanisms on the lifetime of PE. PE products designed for sunlight exposure are required to be stabilized with antioxidant (AO) to retard polymer degradation. In this dissertation, photochemical depletion of AO in PE samples exposed to artificial sunlight was monitored by the oxidative induction time (OIT) testing. A mathematical model was developed by implementing the photochemical reaction scheme, radiation attenuation, and diffusion of AO and oxygen. The model successfully described the AO depletion profiles and obtained a relationship between radiation intensity and AO depletion rate, based on which an equation was proposed to predict the AO depletion in PE in exposure to natural sunlight. Mechanical degradation of PE was studied by investigating creep rupture and stress cracking in thin-wall PE tubes. A test setup was designed to apply static and cyclic internal pressures to thin-wall PE tubes and to record their failure times. The pressure tests were accelerated using elevated temperatures. The failure time (and number of cycles) was predicted using two shifting methods for the dry-cooling applications. Furthermore, this dissertation investigated wave formation of PE geomembranes undergoing thermal expansion during installation. Finite element method (FEM) was used to simulate the formation of waves and quantify the tensile strains along the deformed geomembrane. The model was validated by previous experimental studies. Additionally, the effect of overburden pressure on waves was studied.Ph.D., Civil engineering -- Drexel University, 201
A Qualitative Investigation of Independent School Educators' Lived Experiences with PreK-12 Accreditation
No full textThis study investigates how independent school educators perceive their role within accreditation review processes and associated organizational reform efforts. To fulfill this purpose, research presented here answered two questions: (1) How are educators incorporated into accreditation review processes established and administered by NAIS-certified accreditation agencies? (2) How do these educators perceive their role during accreditation review and how do these perceptions inform their beliefs about (a) the accreditation review process in general, and (b) specific education reform initiatives associated with accreditation specifically? To build context, this study drew on three literature streams. First, prior literature outlined the historical and philosophical development of independent school accreditation. Second, researchers analyzed beliefs surrounding prior school evaluation processes. Finally, this study outlined accreditation principles and requirements of NAIS, the chief certifier of independent school accreditation agencies. To answer each research questions, the researcher collecting public documents relevant to NAIS, its certified agencies, and specific institutions who agreed to participate. Moreover, surveys were distributed to faculty and staff at eight independent schools. In total, the researcher collected 156 valid surveys. Finally, the researcher conducted 30 to 45 minute semi-structured interviews with eight survey participants who completed the aforementioned survey. Findings revealed five common themes present across survey and interview data. (1) Accreditation drives and is driven by school leadership, (2) Accreditation-facilitated inclusivity, (3) Accreditation as an imperfect process, (4) System of continuous improvement, and (5) Institutional honesty. Four results emerged from these findings. Specifically, participants perceived that (1) independent schools and accreditation agencies recognized imperfection and encouraged growth; (2) the accreditation review process hinged on school leadership for independent schools and accreditation agencies; (3) despite being stressful, there was value in an inclusive accreditation review process; and (4) committee deliberation was the cornerstone of accreditation-related work. This study underscores the need for research in the practices of independent school accreditation agencies. It also stresses the need for accreditation agencies to continue its embrace for growth mindset principles by encouraging continual improvement. Finally, research accentuates the imperative of agencies to reform their evaluation practices to provide its member institutions with stronger assessments.Ed.D., Educational Leadership and Management -- Drexel University, 201