Electrical & Computer Engineering Theses and Dissertations
Permanent URI for this collectionhttp://hdl.handle.net/1903/2765
Browse
10 results
Search Results
Item An Optimal Control Model for Human Postural Regulation(2010) Li, Yao; Levine, William S; Electrical Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Human upright stance is inherently unstable without a balance control scheme. Many biological behaviors are likely to be optimal with respect to some performance measure that involves energy. It is reasonable to believe that the human is (unconsciously) optimizing some performance measure as he regulates his balance posture. In experimental studies, a notable feature of postural control is a small constant sway. Specifically, there is greater sway than would occur with a linear feedback control without delay. A second notable feature of the human postural control is that the response to perturbations varies with their amplitude. Small disturbances produce motion only at the ankles with the hip and knee angles unchanging. Large perturbation evoke ankle and hip angular movement only. Still larger perturbation result in movement of all three joint angles. Inspired by these features, a biomechanical model resembling human balance control is proposed. The proposed model consists of three main components which are the body dynamics, a sensory estimator for delay and disturbance, and an optimal nonlinear control scheme providing minimum required corrective response. The human body is modeled as a multiple segment inverted pendulum in the sagittal plane and controlled by ankle and hip joint torques. A series of nonlinear optimal control problems are devised as mathematical models of human postural control during quiet standing. Several performance criteria that are high even orders in the body state or functions of these states (such as joint angle, Center of Pressure COP or Center of Mass COM) and quadratic in the joint control are utilized. This objective function provides a trade-off between the allowed deviations of the position from its nominal value and the neuromuscular energy required to correct for these deviations. Note that this performance measure reduces the actuator energy used by penalizing small postural errors very lightly. By using the Model Predictive Control (MPC) technique, the discrete-time approximation to each of these problems can be converted into a nonlinear programming problem and then solved by optimization methods. The solution gives a control scheme that agrees with the main features of the joint kinematics and its coordination process. The derived model is simulated for different scenarios to validate and test the performance of the proposed postural control architecture.Item Parallelization of Non-Rigid Image Registration(2008) Philip, Mathew; Shekhar, Raj; Electrical Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Non-rigid image registration finds use in a wide range of medical applications ranging from diagnostics to minimally invasive image-guided interventions. Automatic non-rigid image registration algorithms are computationally intensive in that they can take hours to register two images. Although hierarchical volume subdivision-based algorithms are inherently faster than other non-rigid registration algorithms, they can still take a long time to register two images. We show a parallel implementation of one such previously reported and well tested algorithm on a cluster of thirty two processors which reduces the registration time from hours to a few minutes. Mutual information (MI) is one of the most commonly used image similarity measures used in medical image registration and also in the mentioned algorithm. In addition to parallel implementation, we propose a new concept based on bit-slicing to accelerate computation of MI on the cluster and, more generally, on any parallel computing platform such as the Graphics processor units (GPUs). GPUs are becoming increasingly common for general purpose computing in the area of medical imaging as they can execute algorithms faster by leveraging the parallel processing power they offer. However, the standard implementation of MI does not map well to the GPU architecture, leading earlier investigators to compute only an inexact version of MI on the GPU to achieve speedup. The bit-slicing technique we have proposed enables us to demonstrate an exact implementation of MI on the GPU without adversely affecting the speedup.Item Design, Fabrication, and Testing of a Microsystem for Monitoring Bacterial Quorum Sensing(2009) Koev, Stephan Todorov; Ghodssi, Reza; Electrical Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Most pathogenic bacteria communicate with each other using signaling molecules. Their coordinated behavior, known as quorum sensing (QS), enables them to infect host organisms collectively and form drug-resistant biofilms. The study of bacterial signaling pathways may lead to discovery of new antimicrobials. Lab-on-a-chip technology can significantly accelerate the screening of candidate drugs that inhibit QS. This dissertation develops for the first time miniaturized sensors embedded in microfluidic channels to monitor the activity of an enzymatic pathway that produces signaling molecules. These devices can be used as building blocks of future high-throughput systems for drug discovery. The sensors presented here are gold-coated microcantilevers, and they detect the aminoacid homocysteine, a byproduct of the bacterial signaling pathway. It binds to the gold surface, causing stress and cantilever displacement that is measured optically. Samples are synthesized using bacterial enzymes and tested with the sensors. The minimal detected concentration of homocysteine is 1uM. It is demonstrated that deactivation of the enzymes causes a change in the sensor response; this effect can be used for finding drugs that inhibit the enzyme. The traditional method for measuring cantilever displacement requires an elaborate optical setup, and it can only test one device at a time. Two new methods are developed here to overcome these limitations. The first one uses a transparent cantilever which is also an optical waveguide. Light is coupled from the cantilever to a fixed output waveguide and measured with a photodetector. The cantilever displacement is determined from the change in output power. The change is approximately 0.7% per nanometer displacement. The minimal detectable displacement and surface stress are 6nm and 1.3 mN/m respectively. The second measurement method uses a transparent cantilever that is close to a reflective substrate. When the device is imaged with an optical microscope, an interference pattern forms. The cantilever displacement is calculated from the lateral shift of the interference fringes. This shift is determined from images of the device using custom software. The response of multiple cantilevers is captured by translating the microscope stage. The minimal detectable displacement and surface stress are 1nm and 340 uN/m respectively.Item Analog VLSI Circuits for Biosensors, Neural Signal Processing and Prosthetics(2009) Haas, Alfred M.; Peckerar, Martin C.; Electrical Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Stroke, spinal cord injury and neurodegenerative diseases such as ALS and Parkinson's debilitate their victims by suffocating, cleaving communication between, and/or poisoning entire populations of geographically correlated neurons. Although the damage associated with such injury or disease is typically irreversible, recent advances in implantable neural prosthetic devices offer hope for the restoration of lost sensory, cognitive and motor functions by remapping those functions onto healthy cortical regions. The research presented in this thesis is directed toward developing enabling technology for totally implantable neural prosthetics that could one day restore lost sensory, cognitive and motor function to the victims of debilitating neural injury or disease. There are three principal components to this work. First, novel integrated biosensors have been designed and implemented to transduce weak extra-cellular electrical potentials and optical signals from cells cultured directly on the surface of the sensor chips, as well as to manipulate cells on the surface of these chips. Second, a method of detecting and identifying stereotyped neural signals, or action potentials, has been mapped into silicon circuits which operate at very low power levels suitable for implantation. Third, as one small step towards the development of cognitive neural implants, a learning silicon synapse has been implemented and a neural network application demonstrated. The original contributions of this dissertation include: * A contact image sensor that adapts to background light intensity and can asynchronously detect statistically significant optical events in real-time; * Programmable electrode arrays for enhanced electrophysiological recording, for directing cellular growth, for site-specific in situ bio-functionalization, and for analyte and particulate collection; * Ultra-low power, programmable floating gate template matching circuits for the detection and classification of neural action potentials; * A two transistor synapse that exhibits spike timing dependent plasticity and can implement adaptive pattern classification and silicon learning.Item An Optical MEMS Sensor for On-chip Catechol Detection(2008-12-08) Dykstra, Peter Hume; Ghodssi, Reza; Electrical Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)This thesis reports the successful design, fabrication and testing of an optical MEMS sensor for the detection of the toxic phenol, catechol. Catechol's presence in food and drinking water posses a health concern due to its harmful effects on cell respiration. By-products of catechol oxidation have demonstrated increased absorbance changes in a chitosan film in the UV and near UV range. Our reported sensor utilizes patterned SU-8 waveguides and a microfluidic channel to deliver catechol samples to an electrodeposited chitosan film for absorbance measurements at 472 nm. Concentrations as low as 1 mM catechol are detected while control experiments including ascorbic acid display no measurable response. By using optical detection methods, our device does not suffer from many of the problems which plague conventional electrochemical based sensors.Item Integrated CMOS Capacitance Sensor And Microactuator Control Circuits For On-Chip Cell Monitoring(2008-10-07) Bangalore Prakash, Somashekar; Abshire, Pamela; Electrical Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)"Cell Clinics," CMOS/MEMS hybrid microsystems for on-chip investigation of biological cells, are currently being engineered for a broad spectrum of applications including olfactory sensing, pathogen detection, cytotoxicity screening and biocompatibility characterization. In support of this effort, this research makes two primary contributions towards designing the cell-based lab-on-a-chip systems. Firstly it develops CMOS capacitance sensors for characterizing cell-related properties including cell-surface attachment, cell health and growth. Assessing these properties is crucial to all kinds of cell applications. The CMOS sensors measure substrate coupling capacitances of anchorage-dependent cells cultured on-chip in a standard in vitro environment. The biophysical phenomenon underlying the capacitive behavior of cells is the counterionic polarization around the insulating cell bodies when exposed to weak, low frequency electric fields. The measured capacitance depends on a variety of factors related to the cell, its growth environment and the supporting substrate. These include membrane integrity, morphology, adhesion strength and substrate proximity. The demonstrated integrated cell sensing technique is non-invasive, easy-to-use and offers the unique advantage of automated real time cell monitoring without the need for disruptive external forces or biochemical labeling. On top of the silicon-based cell sensing platform, the cell clinics microsystem comprises MEMS structures forming an array of lidded microvials for confining single cells or small cell groups within controllable microenvironments in close proximity to the sensor sites. The opening and closing of the microvial lids are controlled by actuator hinges employing an electroactive polymer material that can electrochemically actuate. In macro-scale setups such electrochemical actuation reactions are controlled by an electronic instrument called potentiostat. In order to enable system miniaturization and enhance portability of cell clinics, this research makes its second contribution by implementing and demonstrating a CMOS potentiostat module for in situ control of the MEMS actuators.Item Model-Based Genomic/Proteomic Signal Processing in Cancer Diagnosis and Prediction(2007-07-20) Qiu, Peng; Liu, K. J. Ray; Electrical Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)In recent years, high throughput measurement technologies (gene microarray, protein mass spectrum) have made it possible to simultaneously monitor the expression of thousands of genes or proteins. A topic of great interest is to study the difference of gene/protein expressions between normal and cancer subjects. In the literature, various data-driven methods have been proposed, i.e. clustering and machine learning methods. In this thesis, an alternative model-driven approach is proposed. The proposed dependence model focuses on the interactions among genes or proteins. We have shown that the dependence model is highly effective in the classification of normal and cancer data. Moreover, different from data-driven methods, the dependence model carries specific biological meanings, and it has the potential for the early prediction of cancer. The concept of dependence network is proposed based on the dependence model. The interactions and co-regulation relationships among genes or proteins are modeled by the dependence network, from which we are able to reliably identify biomarkers, important genes or proteins for cancer prediction and drug development. The analysis extends to cell cycle time-series, where one subject is measured at multiple time points during the cell cycle. Understanding the cell cycle will greatly improve our understanding of the mechanism of cancer development. In the cell cycle time-series, measurements are based on a population of cells which are supposed to be synchronized. However, continuous synchronization loss is observed due to the diversity of individual cell growth rates. Therefore, the time-series measurement is a distorted version of the single-cell expression. In this thesis, we propose a polynomial-model-based resynchronization scheme, which successfully removes the distortion. The time-series data is further analyzed to identify gene regulatory relationships. For the identification of regulatory relationships, existing literatures mainly study the relationship between several regulators and one regulated gene. In this thesis, we use the eigenvalue pattern of the dependence model to characterize several regulated genes, and propose a novel method that examines the relationship between several regulator and several regulated genes simultaneously.Item Radiation dose reduction strategies for intraoperative guidance and navigation using CT(2007-02-20) Shetye, Avanti Satish; Shekhar, Raj; Electrical Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)The advent of 64-slice computed tomography (CT) with high speed scanning makes CT a highly attractive and powerful tool for navigating image guided procedures. Interactive navigation needs scanning to be performed over extended time periods or even continuously. However, continuous CT is likely to expose the patient and the physician to potentially unsafe radiation levels. Before CT can be used appropriately for navigational purposes, the dose problem must be solved. Simple dose reduction is not adequate, because it degrades image quality. This study proposes two strategies for dose reduction; the first is the use of a statistical approach representing the stochastic nature of noisy projection data at low doses to lessen image degradation and the second, the modeling of local image deformations in a continuous scan. Taking advantage of modern CT scanners and specialized hardware, it may be possible to perform continuous CT scanning at acceptable radiation doses for intraoperative navigation.Item Design, Fabrication, and Testing of a Chitosan Based Optical Biosensor(2006-07-31) Powers, Michael; Ghodssi, Reza; Electrical Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)This work presents the design, fabrication, and testing of an original concept for an optical biosensor device intended for use in a microfluidic network. The device uses planar waveguides intersecting a microfludic channel with biofunctionalized patterned sidewalls to detect biomolecules via fluorescent labeling. The optical-biological interface is provided through chitosan, a natural biopolymer. Chitosan is electrodepositable, and this material platform was developed to enable spatially selective and temporally selective assembly of biospecies in the sensor using electrical signals. The unique fabrication process flow integrates waveguides and microfluidic channels which are fabricated in a single step with a thick polymer layer on a Pyrex substrate. Key to the success of the device was the development of a process to pattern indium tin oxide on the sidewalls of deep (130 um) fluid channels. The device was tested in several modes of operation and the proof of concept was shown.Item Low Power Amplifiers for Recording Activity of Electrically Active Cells(2003-12-16) Loganathan, Makeswaran; Abshire, Pamela; Etienne-Cummings, Ralph; Horiuchi, Timothy; Smela, Elisabeth; Electrical EngineeringBiological applications that require sensing individual cells have led to developments in the synthesis of large multielectrode arrays and single cell isolating microstructures. This in turn drives the need for the integration of low power electronic circuitry at or very close to the site of activity. We describe low voltage low power CMOS amplifiers that address this need by rejecting DC offsets, and have tunable bandwidths. They operate at 1.35V, with a power consumption of 37.8μW and have an input referred noise of 23μV. We also describe the design of a wireless transmission system capable of transmitting the electrical signals sensed from cells. Integration of the amplifier array with the wireless link brings continuous monitoring of neurophysiologic activity of unanesthized and freely moving animals closer to realization. The transmitter is capable of generating an ASK modulated signal at a power level of -36 dBm at a frequency of 820 MHz.