|
 |
|
 |
|
-
"ENGAGING IN STEM EDUCATION WITH BIG DATA ANLYTICS AND TECHNOLOGIES: A ROWAN-COVE INITIATIVEG,"
National Science Foundation (NSF), 09/2016 to 09/2019.
Abstract:
Large amounts of data have become available across fields in science, industry, government, healthcare and pharmaceuticals. Big data analytics and technologies hold tremendous promise to boost economic productivity, enhance national security and improve the quality of life. Recognizing the acute need for big data technology, the aim of this proposal is to create a national model by developing multi-year curricular material that allows course content in separate classes to be naturally inter-connected. In collaboration with industry (Lockheed Martin Inc. and Hewlett Packard Enterprise), this proposal designs a series of laboratory experiments in big data analytics and technologies that become more complex from the freshmen to the senior year. The proposed approach will cut across artificial course boundaries and introduce fundamental, contemporary and multidisciplinary big data concepts through a series of problem-oriented laboratory experiments. Students will also gain a better knowledge of policy, ethical, and societal impact issues of big data.
The collaboration with Palmyra Cove Education Foundation will bring big data activities to K-12 and further evaluate the educational impact and increase related dissemination efforts. The activities run at Cove will excite K-12 students about engineering and enhance teacher expertise in mathematics, science and technology. Palmyra Cove will reach out to local underrepresented K-12 school districts targeting women and minorities. The Foundation will also organize a workshop for undergraduate students, university faculty and K-12 students and teachers on big data as applied to environmental engineering and the geosciences.
-
"THEORETICAL AND ALGORITHMIC FOUNDATIONS OF CONSTRAINED PARTICLE FILTERING,"
National Science Foundation (NSF), 09/2015 to 09/2018.
Abstract:
Dynamical systems are often required to satisfy certain constraints arising from basic physical laws, mathematical
properties or geometric considerations of the underlying system, e.g., maximum power, energy
conservation laws and bounded parameters. In fact, constrained systems are already omnipresent in realworld
control applications. Incorporating constraints improves the performance of state estimation and
increases the accuracy compared to unconstrained estimation.
Yet, while particle filters (PF) have gained popularity, within the statistical signal processing community,
thanks to their asymptotically optimal estimation for nonlinear and non-Gaussian state-space models; their
constrained formulation has emerged only very recently, and has never been analytically investigated. Despite
the growing prevalence of potential industrial and real-world applications, the recent developments to
incorporate state constraints in particle filters have been heuristic in nature, mainly relying on constraining
all particles of the PF. This approach, however, constrains the posterior density rather than its mean, which
leads to more stringent conditions and possibly completely different or even irrelevant conditions than the
original constraints. We will, therefore, refer to the approach of constraining all particles as the pointwise
density truncation (PDT) method.
Applications in diverse fields, including robotics, navigation, target tracking, vision-based systems, fault
diagnosis, chemical processes and biomedical systems depend upon the development of a comprehensive
framework for characterizing constrained dynamic systems. Developing this framework is the primary goal
of this grant.
-
"ACQUISITION OF A HIGH PERFORMANCE COMPUTER TO INTEGRATE DATA INTENSIVE RESEARCH AND EDUCATION: BRINGING BIG DATA TO SOUTH JERSEY,"
National Science Foundation (NSF), 09/2014 to 09/2017.
Abstract:
This grant supports for the acquisition of a High Performance Computing (HPC) system for highly parallel and computing-intensive applications across the disciplines within the College of Engineering and the College of Science and Mamthematics at Rowan University. This acquisition will not only enhance research in existing areas of excellence, such as machine learning, biomedical research and geotechnical engineering, but will also expand support for the emerging areas of bioinformatics, dynamic systems optimization and environmental sustainability research. The project team is highly multi-disciplinary, representing two Colleges and more than 8 departments. The awarded HPC facility will be a university-wide shared resource that will be a focal point for strengthening and extending existing research and collaborations and will serve as a platform for developing new research and educational initiatives.
-
"MINIMAL-PERTURBATION DYNAMIC CONTROL OF THE MELANOMA GENE REGULATORY NETWORK," National Institutes of Health (NIH), 08/2010 to 07/2015.
Abstract: The aim of this project is to predict and control by minimal perturbation the dynamics of melanoma gene
regulatory networks. Gene regulatory networks are composed of individual genes that interact with each other in a
complex fashion to regulate the production of proteins viable for cell function. This project uses the regulatory
network structure to introduce an optimal control, in the sense of minimal perturbation, necessary to ensure that
the network will converge to a desired steady-state distribution of gene regulation. The control is formulated as the
inverse problem of perturbation analysis in a general-topology network. The mathematical findings will be confirmed
by computer simulations on a known melanoma gene regulatory network and validated biologically on melanoma
cells.
-
"A SCALABLE ENCRYPTED HUMAN BODY AREA NETWORK SYSTEM," Kathleen Thomsen-Hall Charitable Trust Grant, 06/2011 to 06/2013.
Abstract: The purpose of this project is to create virtual intensive care units that will provide real-time
patient data using a secure wireless sensor network which would prove invaluable in rural
locations and disaster areas. Currently, the medical equipment employed in hospitals uses wired
sensors, which makes it nearly impossible to provide intensive care in any other part of the
hospital, which is not equipped with the requisite wiring.
A wireless Human Body Area Network system presents some challenges in both software
and hardware designs. The main challenges are: reliable communication, secure transmission of
the patients' data, and scalability of the network to multiple sensors and patients. These three
main issues will be tackled in this project.
-
"GSM: AN INTERACTIVE SOFTWARE FOR MULTIDIMENSIONAL ANALYSIS AND VISUALIZATION OF LARGE-SCALE DATA," Arkansas Department of Higher Education (ADHE), 01/2013 to 12/2013.
Abstract:
This projects aims at investigating and ranking various visualization techniques for large-scale multidimensional data and integrating the results into a user-friendly interactive software, GSM (Graphics and Scoring for Multidimensional datasets). This software will enable researchers to easily explore and make intuitive discoveries about the structure and features of the multidimensional data.
-
"DEVELOPMENT OF A CELLULAR DIGITAL TRANSMISSION SYSTEM FOR SMART GRID TECHNOLOGY," Arkansas Department of Higher Education (ADHE), 01/2012 to 12/2012.
Abstract:
Telemetry is the measurement and reporting of remote information. There are countless applications,
which may benefit from such technology. To name a few, we cite meteorology, motor racing, agriculture,
water management, defense, flight, intelligence, energy monitoring, medicine, law enforcement, and
wildlife research and management. There are already many telemetry systems which leverage the SMS
(short messaging service) capabilities of cellular telephones.
In contrast to SMS-based telemetry, we propose to transmit digital data using the voice channel.
Our main application is the smart grid, where a key aspect of this technology is the wireless transmission
of digital data from the metering infrastructure to the utility company. The proposed framework presents
several advantages in terms of range, power consumption, design simplicity, cost and delay.
-
"DESIGN AND IMPLEMENTATION OF SYNTHETIC GENETIC NETWORKS," Arkansas Department of Higher Education (ADHE), 01/2011 to 12/2011.
Abstract:
In this project, we propose to design, analyze and implement genetic networks using analog circuits. Specifically, we consider the repressilator, which is a genetic network that exhibits oscillations in protein expression. We propose two different electronic circuits: An operational amplifier circuit, which reproduces the interactions between genes and proteins in a simple and intuitive manner; and a MOSFET transistor circuit, which would allow integration of a higher number of electronic repressilators. In both implementations, the goal is to reproduce the oscillatory behavior reported in the biochemical experiments. Furthermore, we propose to study the effects of coupling in a group of these oscillators, and analyze its influence in the coherence of the global oscillations.
-
"3D BIOFILM MODELING AND ANALYSIS FOR MEDICAL PURPOSES," Arkansas Science and Technology Authority (ASTA), 01/2011 to 12/2011.
Abstract: In this proposal, we focus on a potential therapeutic target in the specific context of biofilm-associated infections, such as Infective Endocarditis (IE).
A biofilm is an aggregate of microorganisms in which cells adhere to each other and/or to a surface. These adherent cells are frequently embedded within a self-produced matrix of extracellular polymeric substance (EPS). Biofilm EPS, which is also referred to as slime, is a polymeric conglomeration generally composed of extracellular DNA, proteins, and polysaccharides in various configurations. Biofilms may form on living or non-living surfaces, and represent a prevalent mode of microbial life in natural, industrial and hospital settings. In this proposal, the biofilm consists of the Staphylococcus aureus bacteria, encapsulated in extracellular matrix. Different strains of bacteria form biofilm of different qualities. Since biofilm is a detrimental element of staphylococcal infections, it is often targeted for therapeutic purposes. Not only bacteria exist in low metabolic state within biofilm (antibiotics often target bacterial growth), biofilm also create a physical barrier for the penetration of therapeutics.
Accurate 3D model of the structure of the biofilm is crucial in studies on bacteria straits that create biofilm in their development in living organisms. High dimensional modeling adds another layer of information, introducing special correlation between cross-section scans of bacteria cultures under development.
-
"EXTENSION OF THE MUSIC ALGORITHM TO AM-FM SIGNALS," Arkansas Department of Higher Education (ADHE), 01/2010 to 12/2010.
Abstract: Recent research in cell biology shows that a number of cell types respond to external stimulation by encoding cellular information in their amplitude (AM) and frequency (FM). Furthermore, understanding this regulation is central to the study of many diseases, most notably cancer. Therefore, accurate estimation of the time-dependent (or non-stationary) amplitude and frequency modulation of the temporal expression of cellular signals is an important problem in biology.
In this project, we tackle the non-stationary estimation problem by extending the stationary Multiple Signal Classification (MUSIC) algorithm to non-stationary AM-FM signals. Our approach relies on a basis decomposition of the time-dependent amplitude and frequency signals. We show that the introduction of a basis representation reduces the non-stationary estimation problem to a stationary one. The proposed AM-FM MUSIC algorithm will be assessed using test signals, and the AM-FM calcium signaling response in lymphocyte cells.
|
|
 |
|
 |
|
|
