Category: Research

Faculty Candidate Muhammad Fahad to Present Lecture April 9

The College of Computing’s Department of Applied Computing invites the campus community to a lecture by MERET faculty candidate Muhammad Fahad on Thursday, April 9, 2020, at 3:30 p.m., via an online Zoom meeting. His talk is titled, “Motion Planning and Control of Autonomous Mobile using Model Free Method.”

Link to the Zoom meeting here.

Dr. Fahad currently works as a robotics engineer at National Oil Well Varco. He received his M.S. and Ph.D. in electrical engineering from Stevens Institute of Technology, Hoboken, NJ, and his B.S in EE at University of Engineering and Technology, Lahore, Pakistan.

Fahad has extensive experience designing control and automation systems for the process industry using traditional control methods and robots. His research interests include cooperative distributed localization, human robot interaction (HRI), deep reinforcement learning (DRL), deep inverse reinforcement learning (DIRL) and generative adversarial imitation learning (GAIL), simulation tools design, parallel simulation frameworks and multi-agent learning.

Lecture Abstract. Robots are playing an increasingly important part in our daily lives. This increasing involvement of robots in our everyday lives has highlighted the importance of human-robot interaction, specifically, robot navigation of environments occupied by humans, such as offices, malls and airports. Navigation in complex environments is an important research topic in robotics.

The human motion model consists of several complex behaviors that are difficult to capture using analytical models. Existing analytical models, such as the social force model, although commonly used, are unable to generate realistic human motion and do not fully capture behaviors exhibited by humans. These models are also dependent on various parameters that are required to be identified and customized for each new simulation environment. 

Artificial intelligence has received booming research interest in recent years. Solving problems that are easy for people to perform but difficult to describe formally is one of the main challenges for artificial intelligence. The human navigation problem falls directly in this category, where it is hard to define a universal set of rules to navigate in an environment with other humans and static obstacles.

Reinforcement learning has been used to learn model-free navigation, but it requires a reward function that captures the behaviors intended to be inculcated in the learned navigation policy. Designing such a reward function for human like navigation is not possible due to complex nature of human navigation behaviors. The speaker proposes to use measured human trajectories to learn both the reward function and navigation policy that drives the human behavior.

Using a database of real-world human trajectories–collected over a period of 90 days inside a mall–we have developed a deep inverse reinforcement learning approach that learns the reward function capturing human motion behaviors. Further, this dataset was visualized in a robot simulator to generate 3D sensor measurement using a simulated LIDAR sensor onboard the robot. A generative adversarial imitation learning based method is developed to learn the human navigation policy using these human trajectories as expert demonstration. The learned navigation policy is shown to be able to replicate human trajectories both quantitatively, for similarity in traversed trajectories, and qualitatively, in the ability to capture complex human navigation behaviors. These navigation behaviors include leader follower behavior, collision avoidance behavior, and group behavior. 


Faculty / Researcher Profile: Weihua Zhou

Faculty/Researcher Profile: Weihua Zhou, Multi-Disciplinary Digital Healthcare Solutions

By Karen Johnson, Communications Director, College of Computing and Institute of Computing and Cybersystems

How can the cost-effectiveness of healthcare be improved, especially for complicated chronic diseases? This is the overarching question Dr. Weihua Zhou is seeking to answer with his research. The multi-disciplinary solutions he is investigating merge the fields of medical imaging and informatics, computer vision, and machine learning. 

An assistant professor in Michigan Tech’s Health Informatics program, and an affiliated associate professor in the Biomedical Engineering department, Zhou is working with students on a number of research projects in Michigan Tech’s Medical Imaging and Informatics Lab, which he directs. He is a member of the Institute of Computing and Cybersystems’s Center for Data Science.

Zhou says his research is driven by clinical significance, and he is especially interested in developing practical solutions to improve the cost-effectiveness of treating complicated chronic diseases, such as coronary artery disease, heart failure and senile dementia. 

He is excited about his career, his international research, and his work at Michigan Tech. “We have a very productive team, including dedicated Ph.D. students, self-motivated graduate and undergraduate students, and a lot of experienced clinical and technical collaborators,” he says of his colleagues and collaborators at Michigan Tech and around the world.

Zhou feels that he can be dedicated to both his research and teaching at Michigan Tech. “I joined the Health Informatics program at Michigan Tech, both because health informatics is my research focus, and because Michigan Tech’s leading reputation among engineering schools opens opportunities to find new and respected technical collaborators. 

Zhou often calls himself a salesman. “I sell techniques to our clinical collaborators and ask them to design the projects with me, provide the patient data, and test our tools,” he explains. “I also sell my ideas about clinical problems to technical collaborators and ask them to work with us to solve the important clinical problems.”

And when he communicates with his Ph.D. students, “sometimes I also consider them as my buyers and let them appreciate my ideas so that they can be really inspired.”

Primary Research

Zhou identifies two of his research projects of as primary. 

“This first is exploring image-guided approaches to improving the treatment of heart failure, which has been supported by AHA grants, and is now being supported by a new faculty startup grant,” Zhou says. “The second main project is seeking to employ machine learning to improve the risk stratification for osteoporosis, which is supported by a National Institutes of Health (NIH) subcontract award from Tulane University.”

On the NIH grant, awarded in December 2019, Zhou is working with internationally renowned researcher and educator Dr. Hong-Wen Deng, an endowed chair and professor in the School of Public Health and Tropical Diseases at Tulane University, New Orleans, La. Zhou and Deng are studying trans-omics integration of multi-omics studies for male osteoporosis.

Zhou is also co-PI with Jinshan Tang, professor of Applied Computing (eff. 7/1/20) at Michigan Tech, on a Portage Health Foundation Infrastructure Enhancement Grants titled, “High Performance Graphics Processing Units.” The project is focused on building big data computing capabilities toward advancing research and education. Several additional proposals are under review and revision. Zhou’s past research support includes an American Heart Association award, which studied a new image-guided approach for cardiac resynchronization therapy.

Teaching and Mentoring

Zhou, who started at Michigan Tech in fall 2019, instructed Introduction to Health Informatics in the fall semester, and Applied Artificial Intelligence in Health this spring.  He says that in the Medical Informatics program, the subjects he teaches are very practical.

“I believe the following strategies are very important and I practice them in my classes every day: 1) Make the class interactive; 2) Make the assignments and projects practical; 3) Emphasize the learning process; and 4) Keep the teaching materials up to date,” Zhou says.

Zhou supervises two Ph.D. candidates in the Department of Applied Computing, and a Health Informatics master’s student.

Applied Computing Ph.D. candidate Zhuo He’s primary research project concerns information fusion between electrical signal propagation and mechanical motion to improve the treatment of heart failure. Ph.D. candidate Chen Zhao’s primary research concerns using image fusion and computer vision to improve interventional cardiology. And Zhou’s Health Informatics master’s student, Rukayat Adeosun, is studying nuclear image-guided approaches to improving cardiac resynchronization therapy.

Education and Post-Doc

Zhou was awarded his Ph.D. in computer engineering by the Department of Electrical and Computer Engineering at Southern Illinois University Carbondale in 2012; his dissertation is titled, “Image reconstruction and imaging configuration optimization with a novel nanotechnology enabled breast tomosynthesis multi-beam X-ray system.”

Following, Zhou was a post-doctoral researcher in the Department of Radiology and Imaging Sciences at Emory University, Atlanta, Georgia, then he was appointed a Nina Bell Suggs Endowed Professor at University of Southern Mississippi, where he was a tenure-track assistant professor. Zhou also completed an MSc.-Ph.D. in computer science (2007) and a B.E. in computer science and technology (2003), both at Wuhan University, China.

Achievement

Zhou received the USM College of Arts and Sciences Scholarly Research Award in March 2019, participated in the AHA Research Leaders Academy of the American Heart Association in September 2017 and August 2018, and received the USM Butch Oustalet Distinguished Professorship Research Award in April 2018.

University and Professional Service

Zhou serves on Michigan Tech’s Review Committee for Graduate Dean’s Awards Advisory Committee, and in October 2019 he served on the Review Committee for Research Excellence Fund (REF) – Research Seed Grants (RS).

He was an invited speaker at the Machine Learning in SPECT MPI Applications session at the Annual Scientific Session of the American Society of Nuclear Cardiology in Washington, D.C., in 2009.

Zhou is a member of the American Heart Association (AHA) and the American Society of Nuclear Cardiology (ASNC).

Peer-Review

Since Zhou joined Michigan Tech in August 2019, he has published five scholarly papers, in Journal of Nuclear Cardiology and the IEEE Journal of Translational Engineering in Health and Medicine. Two additional articles are under revision with Journal of Nuclear Cardiology and the journal Medical Physics, and one is under review by the Medical Image Computing and Computer Assisted Intervention (MICCAI) Conference 2020.

Since 2007, he has published more than 80 peer-reviewed journal and conference papers and book chapters in publications including JACC: Journal of The American College of Cardiology: Cardiovascular Imaging, Journal of Nuclear Cardiology, and IEEE Journal of Translational Engineering in Health and Medicine.

Zhou is a translator of featured papers and abstracts for the Journal of Nuclear Cardiology, and a paper reviewer for the Journal of Nuclear Cardiology, JACC: Journal of The American College of Cardiology, and JACC: Cardiovascular Imaging. He is a reviewer for American Heart Association data science grants. 

Commercial Success

Zhou holds a number of patents and invention disclosures, including new methods to 1) diagnose apical hypertrophic cardiomyopathy from gated single-photon emission computed tomography (SPECT), and 2) measure right-ventricular and interventricular mechanical dyssynchrony from gated single-photon emission computed tomography (SPECT) myocardial perfusion imaging (MPI); and 3) the integration of fluoroscopy venogram and myocardial perfusion SPECT image with left-ventricular contraction sequence and scar distribution to guide the real-time surgery of cardiac resynchronization therapy. 

He and his colleagues have developed a number of software tools, some of which are being used in hospitals for research purposes, and he believes that the tools can be successfully validated and become commercially available. For example, Zhou’s nuclear image-guided software toolkit to improve cardiac resynchronization therapy is being validated by a large clinical trial. 

A personal note.

Zhou loves independent thinking, facts and exact numbers, and he values persistence, all of which express themselves in his teaching and research, and his life.

Follow Weihua Zhou on Twitter: @LabMiil

The College of Computing Department of Applied Computing will officially replace the CMH Division effective July 1, 2020.


ROTC Cybersecurity Training for Tomorrow’s Officers

The U.S. Department of Defense, Office of Naval Research, has awarded Michigan Tech faculty researchers a $249,000 grant that supports the creation of an ROTC undergraduate science and engineering research program at Michigan Tech. The primary goal of the program is to supply prepared cadets to all military branches to serve as officers in Cyber commands.

The principal investigator (PI) of the project is Andrew Barnard, Mechanical Engineering-Engineering Mechanics. Co-PIs are Timothy Havens, College of Computing; Laura Brown , Computer Science, and Yu Cai, Applied Computing. The title of the project is, “Defending the Nation’s Digital Frontier: Cybersecurity Training for Tomorrow’s Officers.”

The curriculum will be developed over the summer, and instruction associated with the award will begin in the fall 2020 semester. Cadets interested in joining the new program are urged to contact Andrew Barnard.

Initially, the program will focus on topics in cybersecurity, machine learning and artificial intelligence, data science, and remote sensing systems, all critical to the The Naval Science and Technology (S&T) Strategic Plan and the Navy’s Force of the Future, and with equal relevance in all branches of the armed forces.

The plan of work focuses on on engaging ROTC students in current and on-going Cyber research, and supports recruitment of young ROTC engineers and scientists to serve in Navy cybersecurity and cyber-systems commands. The program will compel cadets to seek positions within Cyber commands upon graduation, or pursue graduate research in Cyber fields.

“Our approach develops paid, research-based instruction for ROTC students through the existing Michigan Tech Strategic Education Naval Systems Experiences (SENSE) program,” said principal investigator Andrew Barnard, “ROTC students will receive one academic year of instruction in four Cyber domains: cybersecurity, machine learning and artificial intelligence (ML/AI), data science, and remote sensing systems.”

Barnard says the cohort-based program will enrich student learning through deep shared research experiences. He says the program will be designed with flexibility and agility in mind to quickly adapt to new and emerging Navy science and technology needs in the Cyber domain.

Placement of officers in Cyber commands is of critical long-term importance to the Navy (and other DoD branches) in maintaining technological superiority, says the award abstract, noting that technological superiority directly influences the capability and safety of the warfighter.

Also closely involved in the project are Michigan Tech Air Force and Army ROTC officers Lt. Col. John O’Kane and LTC Christian Thompson, respectively.

“Unfortunately, many ROTC cadets are either unaware of Cyber related careers, or are unprepared for problems facing Cyber officers,” said Lt. Col. O’Kane. “This proposal aims to provide a steady flow of highly motivated and trained uniformed officers to the armed-services, capable of supporting the warfighter on day-one.”

Andrew Barnard is director of Michigan Tech’s Great Lakes Research Center, an associate professor of Mechanical Engineering-Engineering Mechanics, and faculty advisor to the SENSE Enterprise.

Tim Havens is director of the Institute of Computing and Cybersystems, associate dean for research, College of Computing, and the William and Gloria Jackson Associate Professor of Computer Systems.

Laura Brown is an associate professor, Computer Science, director of the Data Science graduate program, and a member of the ICC’s Center for Data Sciences.

Yu Cai is a professor of Applied Computing, an affiliated professor of Computational Science and Engineering, a member of the ICC’s Center for Cybersecurity, and faculty advisor for the Red Team, which competes in the National Cyber League (NCL).

The Great Lakes Research Center (GLRC) provides state-of-the-art laboratories to support research on a broad array of topics. Faculty members from many departments across Michigan Technological University’s campus collaborate on interdisciplinary research, ranging from air–water interactions to biogeochemistry to food web relationships.

The Army and Air Force have active ROTC programs on Michigan Tech’s campus.

The Office of Naval Research (ONR) coordinates, executes, and promotes the science and technology programs of the United States Navy and Marine Corps.


Article by Tim Havens in IEEE Transactions on Fuzzy Systems

An article co-authored by Tim Havens, associate dean for research, College off Computing, “Soft Overlapping Community Detection in Large-Scale Networks via Fast Fuzzy Modularity Maximization,” was published in the March 2020 issue of IEEE Transactions on Fuzzy Systems.

Havens’s co-authors are Audrey Yazdanparast (ECE) and Mohsen Jamalabdollahi of Cisco Systems.

Article Abstract: Soft overlapping clustering is one of the notable problems of community detection. Extensive research has been conducted to develop efficient methods for non-overlapping and crisp-overlapping community detection in large-scale networks. In this paper, Fast Fuzzy Modularity Maximization (FFMM) for soft overlapping community detection is proposed.

FFMM exploits novel iterative equations to calculate the modularity gain associated with changing the fuzzy membership values of network vertices. The simplicity of the proposed scheme enables efficient modifications, reducing computational complexity to a linear function of the network size and the number of communities. Moreover, to further reduce the complexity of FFMM for very large networks, Multi-cycle FFMM (McFFMM) is proposed.

The proposed McFFMM reduces complexity by breaking networks into multiple sub-networks and applying FFMM to detect their communities. Performance of the proposed techniques are demonstrated with real-world data and the Lancichinetti-Fortunato-Radicchi (LFR) benchmark networks. Moreover, the performance of the proposed techniques is eval- uated versus some state-of-the-art soft overlapping community detection approaches. Results show that the McFFMM produces a remarkable performance in terms of overlapping modularity with fuzzy memberships, computational time, number of detected overlapping nodes, and Overlapping Normalized Mutual Informa- tion (ONMI).

View more info here.


Tim Havens Is Co-author of Article in IEEE Transactions on Fuzzy Systems

Timothy Havens, director of the Institute of Computing and Cybersystems (ICC), is co-author of the article, “A Similarity Measure Based on Bidirectional Subsethood for Intervals,” published in the March 2020 issue of IEEE Transactions on Fuzzy Systems.

Havens’s co-authors are Shaily Kabir, Christian Wagner, and Derek T. Anderson.

Havens is also associate dean for research, College of Computing, and the William and Gloria Jackson Associate Professor of Computer Systems.

Christian Wagner, an affiliated member of the ICC, was an ICC donor-sponsored visiting professor at Michigan Tech in the 2016-17 academic year. He is now with the School of Computer Science at University of Nottingham.

Shaily Kabir is with the School of Computer Science, University of Nottingham. Derek T. Anderson is with the Electrical Engineering and Computer Science Department, University of Missouri, Columbia.

S. Kabir, C. Wagner, T. C. Havens and D. T. Anderson, “A Similarity Measure Based on Bidirectional Subsethood for Intervals,” in IEEE Transactions on Fuzzy Systems.

https://ieeexplore.ieee.org/document/9019656


2020 Undergraduate Research Symposium is March 27

Undergraduate researchers and scholars from all colleges—first-year students to soon-to-graduate seniors—will present a record 76 posters at the 2020 Undergraduate Research Symposium, Friday, March 27, 2020, in the lobby of the Rozsa Center. Two sessions will take place, from 11:00 a.m. to 1:00 p.m., and from 2:00 to 4:00 p.m.

The Symposium, hosted by the Pavlis Honors College, highlights the amazing cutting-edge research being conducted on Michigan Tech’s campus by some of our best and brightest undergraduate students.

All faculty, staff and students are encourage to attend and support our excellent undergraduate researchers. Faculty members who would like to serve as distinguished judges at this year’s symposium may complete this short form

Learn more about the Symposium here.


Faculty Candidate Leo Ureel to Present Lecture March 24

The Colleges of Computing and Engineering invite the campus community to a lecture by faculty candidate Leo C. Ureel II on Tuesday, March 24, 2020, at 3:00 p.m. The title of Ureel’s lecture is, “Critiquing Student Code by Identifying Novice Anti-patterns.”

Join the online Zoom meeting here.

Ureel is a senior lecturer and PhD candidate in the Computer Science department at Michigan Tech. He has been teaching at the college level for 10 years, and he has over 20 years of industry experience in developing software for engineering, artificial intelligence, and education.

Ureel’s research focuses on a constructionist approach to introductory computer science that leverages code critiquers to motivate students to learn computer programming, with less cognitive overhead than is usually associated with learning programming and computation. In particular, he is developing critiques tools designed to provide students with feedback on programming assignments in ways that are similar to human instructors. Critiquer systems can be used to engage students in test-driven agile development methods through small cycles of teaching, coding integrated with testing, and immediate feedback.

Ureel’s work has provided him the opportunity to develop rich collaborations with researchers across the U.S. and in the U.K., Europe, and Africa, and he recently led an ITICSE working group of international researchers. Ureel teaches CS1 and CS2 courses, primarily to first year students, in which he works to broaden students’ views of computing, ground them in a programming language, and teach them problem solving skills. His research has has been supported by NSF, Google, and NCWIT.

Download

Faculty Candidate Vidhya Nagaraju to Present Lecture March 20

The College of Computing invites the campus community to a lecture by faculty candidate Vidhyashree Nagaraju on Friday, March 20, 2020, at 3:00 p.m. The title of Nagaraju’s talk is “Software Reliability Engineering: Algorithms and Tools.”

The lecture will be presented online through a Zoom meeting. Link to the meeting here.

Vidhyashree Nagaraju is a Ph.D. candidate in the Department of Electrical and Computer Engineering at the University of Massachusetts Dartmouth (UMassD), where she received a M.S. in Computer Engineering in 2015. She received a B.E. in electronics and communication engineering from Visvesvaraya Technological University, India, in 2011.

While there are many software reliability models, there are relatively few tools to automatically apply these models. Moreover, these tools are over two decades old and are difficult or impossible to configure on modern operating systems, even with a virtual machine. To overcome this technology gap, Nagaraju is developing an open source software reliability tool for the software and system engineering community. 

A key challenge posed by such a project is the stability of the underlying model fitting algorithms, which must ensure that the parameter estimates of a model are indeed those that best characterize the data. If such model fitting is not achieved, users who lack knowledge of the underlying mathematics may inadvertently use inaccurate predictions. This is potentially dangerous if the model underestimates important measures such as the number of faults remaining or overestimates the mean time to failure (MTTF).

To improve the robustness of the model fitting process, expectation conditional maximization (ECM) algorithms have been developed to compute the maximum likelihood estimates of nonhomogeneous Poisson process (NHPP) software reliability models. Nagaraju ‘s talk will present an implicit ECM algorithm, which eliminates computationally intensive integration from the update rules of the ECM algorithm, thereby achieving a speedup of between 200 and 400 times that of explicit ECM algorithms. The enhanced performance and stability of these algorithms will ultimately benefit the software.

Download

Intel Online Workshop Is Monday, March 2

Intel has invited Michigan Tech students and faculty to join a 4-hour online workshop on Monday, March 2, 2020, at 11:00 a.m. EST. Intel will demonstrate a computer vision workflow using the OpenVINO toolkits, including support for deep learning algorithms that help accelerate Smart Video applications.

The workshop provides an opportunity to learn how to optimize and improve performance, with and without external accelerators, and utilize tools to help us identify the best hardware configuration for our needs.

When: Monday, March 2, March 2020 | 11 am (Houghton Time)

If interested, the event is free but registration is necessary https://iotevents.intel.com/VirtualWorkshop2020March2/

Agenda
20 Minutes – Intel Smart Video/Computer vision Tools Overview
20 Minutes – Model Optimizer
20 Minutes – Inference Engine
10 Minutes – Break
15 Minutes – Intel Movidius„ Inference Accelerator
15 Minutes – FPGA Inference Accelerator
10 Minutes – Register for access to DevCloud
90 Minutes – Labs on Intel DevCloud for the Edge

Who will benefit: Computer vision developers with a basic understanding of machine learning and deep learning techniques who would like to learn optimization and acceleration techniques for industrial, medical, smart city, and smart retail applications.

What’s to gain: Overview and application of Intel computer vision technologies.Understanding of deep learning development using a pre-trained model.7-day access to Intel Devcloud for the Edge, to continue training after the event.

Questions?

Contact Gowtham, PhD
Director of Research Computing, IT
Research Associate Professor, ECE
Michigan Technological University
P: (906) 487-4096
F: (906) 487-2787
g@mtu.edu
https://it.mtu.edu
https://hpc.mtu.edu