Department of Computer Science and Electrical Engineering

CSEE professor Jian Chen recently received an award from Department of Defense to to develop new techniques to visualize health informatics data. The award will support two UMBC research students for two years and be done in collaboration with Jesus Caban, Gerard Reidy and Joseph Bleiberg from the Walter Reed National Military Medical Center's National Intrepid Center of Excellence.

The research will the support temporal exploration and analysis of traumatic brain injury and post traumatic stress disorder from patient cohorts. The group will design interactive visualization to move beyond using a small subset of data from the wealth and breadth of clinical information to improve diagnostic accuracy.

The project will help clinicians obtain new insights about the underlying conditions of patients, analyze complex hidden clinical patterns, and visually explore the correlations between many assessment techniques and imaging modalities including neuroimaging, neuropsychiatric measures, patient history, demographic information, and clinical tests).

CSEE Professors Nilanjan Banerjee, Ryan Robucci and Chintan Patel with Dr. Richard Allen at Johns Hopkins University were awarded a TEDCO Maryland Innovation Initiative grant to develop a wearable system that can non-intrusively monitor sleep quality in a home setting. The grant will fund students in the eclipse cluster. The researchers will work closely with JHU to develop and help commercialize this novel sensor system.

TEDCO is an independent organization that provides entrepreneurial business assistance and seed funding to foster startup companies in Maryland’s innovation economy and support the commercialization of qualified university technologies.

CSEE Professor Tülay Adali and Professor Kelly Westlake from the University of Maryland School of Medicine received an award from the joint UMBC-UMB Research and Innovation Partnership Seed Grant Program for a project  that ultimately will improve the recovery of stroke victims. The new joint UMBC-UMB Seed Grant Program pairs primary investigators from UMBC and the University of Maryland, Baltimore to conduct research as a team. Successful partners are offered research funding of up to $75,000 over twelve months to pursue their collaboration.

Nearly 800,000 U.S. citizens have a stroke each year, making it the leading cause of long-term disability. Treatment for stroke victims is mainly targeted towards residual functional deficits, especially regaining hand functions. Their project (“Independent Vector Analysis to Investigate Cognitive Neural Networks after Stroke: A Comparison between Two Rehabilitation Interventions”) will have a direct impact on stroke rehabilitation through objective evaluation of the two main treatment paradigms currently in use: unimanual (involving one hand) and bimanual (involving both hands) training. The evaluation will use the new class of medical image analysis techniques, independent vector analysis (IVA) algorithms, developed by Dr. Adali and her research group. The new class of IVA algorithms successfully captures subject variability and perform significantly better than the approaches traditionally used for the problem.

The initial results of the project demonstrate the advantages using IVA for the problem and will be presented this month at the 48th Annual Conference on Information Sciences and Systems in Princeton, NJ. The PIs are preparing two journal submissions, one based on the methods developed for the task and a second one emphasizing clinical significance of the results. These results will also provide the preliminary data for the proposal that Professors Adali and Westlake plan to submit to the NIH later this year.

Professor Marie desJardins received a research award from the National Science Foundation focused on increasing the expertise of Maryland high school teachers for teaching computer science, with the ultimate goal of increasing the number of computer science classes offered in Maryland and the number and diversity of students taking them.

The project, CS10K: Collaborative Research: A Structured CS Principles Approach to Professional Development for Maryland High School Teachers, is led by Dr. desJardins and will involve collaborators from the Computer Science department at the University of Maryland, College Park, and high school teachers from Charles County and Baltimore County. The NSF award will provide $845,000 in funding over a three year period and is part of the Computing Education for the 21st Century (CE21) program, which aims to build a robust computing research community, a computationally competent 21st century workforce, and a computationally empowered citizenry.

The project is also part of the CS 10K community, which works to make computer science accessible in high schools. The nationwide effort has a goal of getting rigorous academic computer science courses into 10,000 schools taught by 10,000 well-prepared teachers. The project pays careful attention to the inclusion of women and minorities–groups that have been traditionally underrepresented in computing.

The project will develop and evaluate professional development activities focused on increasing the expertise of Maryland high school teachers for teaching computer science. Experienced higher education faculty and highly effective high school teachers will collaborate to train other high school teachers to develop and offer a college preparatory CS curriculum. In the first year, an initial cohort of 12 master teachers will collaborate with the project leadership team to develop standardized classroom materials based on the CS Principles curriculum and then offer the CS Principles course in their home schools. In the second year, the project staff will train less experienced teachers in the CS Principles curriculum, and in the third year, the leadership team will demonstrate scalability by supporting the delivery of multiple independently funded regional workshops. The project’s ultimate goal is to achieve significantly increased interest of Maryland students in studying and pursuing careers in computer science.

CSEE Professors Tinoosh Mohsenin and Gymama Slaughter each received recent grants from NSF to apply their computing engineering expertise to develop new medical technology. What follows is an excerpt from a recent article on their work written by Joel N.Shurkin.

The Body Electric: UMBC researchers forge links between tech and medicine

Monitoring significant developments in a patient’s health outside a hospital setting can be challenging, but two UMBC researchers – Tinoosh Mohsenin and Gymama Slaughter – have won separate grants from the National Science Foundation (NSF) to help meet those challenges.

Mohsenin received a $100,000 grant from the NSF to develop signal processing architecture to detect seizures. The award of $150,000 to Slaughter from the foundation was to pursue work on nanoelectric probe arrays.

Not only does the work done by these two UMBC professors have important implications for basic medical science, but it is also research that may also provide more insight into how we think and feel, or improve how people with disabilities navigate in the world.

CSEE research professor Karuna Joshi received a $100K grant to develop an advanced prototype for a cloud service broker from the Maryland Technology Development Corporation (TEDCO). The software system will help companies and organizations negotiate for, select and procure cloud computing services based on a description of their needs and preferences.

TEDCO was created by the Maryland State Legislature to facilitate the transfer and commercialization of technology from Maryland’s research universities and federal labs into the marketplace and to assist in the creation and growth of technology-based businesses in Maryland.

Dr. Joshi’s project will build on a novel framework she developed as part of her PhD dissertation for automating the process lifecycle of cloud services. She designed technology that allowed users to compare different cloud services and determine which were best suited for their needs. Her initial prototype demonstrated a system that could negotiate for and procure storaage services from a provider like Amazon Web Services.

As part of the TEDCO funded project, Dr. Joshi will develop an commercially viable enhanced cloud broker engine that can be extended to include other services, as well as do more complex matchmaking based on functional and compliance requirements.

Professor Jian Chen received a research award from the National Institute of Standard and Technology (NIST) Measurement Science and Engineering research grant program to discover new immersive metrology, visualization, and analysis tools for interacting with and understanding multi-valued volumes of large scientific data in immersive display environments.

The research will explore new approaches to integrate scientific and information visualization, optimize information access and precise measurement to recommend good default visualization, study natural and intuitive input modalities including gestural and handheld 3D inputs, and synthesize visualizations to address complex data analysis workflow. The efficacy and insights of these scientific methods will be validated in close collaboration with engineering scientists at NIST.

The research is significant because the methods will be able to address problems in three independent scientific applications at NIST: suspension rheology, body area network and tissue engineering. It will also make possible new forms of scientific research by developing new immersive analysis capabilities, integrating new approaches into experimental research, and for the first time, creating new human-computer interaction techniques to query both scientific and information visualizations to leverage human intelligence.

The award will provide nearly $440,000 over the next five years to support Professor Chen and her students in the DaVinCI (Data Visualization, Computing, and Interaction) lab working on the project.

Professor Charles Nicholas has received research funding from the National Science Foundation to develop better ways to detect malicious software (malware) and defend computers against it. The award will provide up to $75,000 over the next year to support the research of Dr. Nicholas and his students.

The process of creating malware has become more automated in recent years, as a result of so-called exploit kits, such as the Blackhole exploit kit. The UMBC project will investigate ways of characterizing these exploit kits, as well as the malware they produce. Developing models of how current kits work will help to predict what exploit kits will look like in the future as well as suggest better techniques for detecting the malware they are used to produce.

On challenging problem the research will address is dealing with polymorphic malware, malware that makes new versions of itself as it moves from machine to machine, in the hope of avoiding detection by conventional, signature-based anti-virus software. The research will characterize malware families that exist as products of a specific exploit kit as well as those that develop through polymorphism.

Dr. Nicholas is the faculty advisor of UMBC’s Cyber Defense Team, a student group that studies information security, intrusion detection, cybersecurity, and network security and participates in competitions such as the Maryland Cyber Challenge. He is also teaching a special topics class this semester on Malware Analysis for both undergraduate and graduate students.

Dr. Tom Narock (Goddard Planetary Heliophysics Institute) and Professor Tim Finin (CSEE) are part of a team that receive a one-year, $300,000 award from the National Science Foundation to apply semantic technologies to support the data representation, discovery and integration needs in EarthCube, an NSF program that aims to transform geoscience research by developing community-guided cyberinfrastructure.  The collaborative project involves researchers and students from UMBC, Columbia University, Wright State University, and the Woods Hole Oceanographic Institution (WHOI). Narock (PI) and Finin (CO-PI) head the UMBC effort.

A wide spectrum of maturing methods and tools, collectively characterized as Semantic Web Technologies, enables machines to complete tasks automatically that previously required human direction. For the Geosciences, Semantic Web Technologies will vastly improve the integration, analysis and dissemination of research data and results. This collaborative project will conduct exploratory research applying state-of-the-art Semantic Web Technologies to support data representation, discovery, analysis, sharing and integration of datasets from the global oceans, and related resources including meeting abstracts and library holdings. A key contribution will be semantically-enabled cyberinfrastructure components capable of automated data integration across distributed repositories.

The image above shows a pyramid-shaped multicorer on the deck of the R/V Melville off Santa Barbara in October 2012. Multicorers collect seafloor sediment samples without disrupting the uppermost sediment layers and the single-celled organisms living in them. The system sends real-time images of the seafloor to scientists aboard, allowing them to guide the sampler, and collects high-resolution images that are stored in the camera for downloading on recovery. (Photo by Ellen Roosen, WHOI).

Professor Tinoosh Mohsenin received a research award from the National Science Foundation to develop multi-physiological signal processing architectures for seizure detection. The award will provide $100,000 over a two year duration to support Dr. Mohsenin and her students in the Energy Efficient High Performance Computing Lab working on this project.

The research will explore an approach to overcome problems of sensor noise, false detection, and energy/power constraints by combining the analysis of multiple physiological signals through specialized hardware implementing a multi-layer classification technique comprised of signal processing and machine learning functions. The hybrid architecture will leverage common operations and communication patterns between digital signal processing and machine learning to support these computations more efficiently than traditional approaches. The prototype system will be evaluated for use in wearable seizure detection devices by using traces of EEG and other physiological sensor data obtained from the Epilepsy Center at University of Maryland Medical Center.

The research could significantly enhance the ability to robustly and efficiently monitor multi-physiological patient data and take appropriate actions. For epilepsy, it could enable automatic seizure detection and caregiver alerts, which are critical at night, when seizures can happen without someone to help nearby. Longer term potential impacts extend to human-centered cyber-physical systems, cyber-security, and unmanned vehicles.