Data Science Research

Project Title: Revolution through Evolution: A Controls Approach to Improve How Society Interacts with Electricity

Investigator: Laura Brown

Sponsor: National Science Foundation

Overview: This CRISP (Computer Retrieval of Information on Scientific Projects) project addresses the challenges associated with the rapid evolution of the electricity grid to a highly distributed infrastructure. The keystone of this research is the transformation of power distribution feeders, from relatively passive channels for delivering electricity to customers, to distribution microgrids, entities that actively manage local production, storage and use of electricity, with participation from individual customers. Distribution microgrids combine the advantages of the traditional electricity grid with the advantages of emerging distributed technologies, including the ability to produce and use power locally in the event of grid outages.

The project will result in a unified model that incorporates key aspects of power generation and delivery, information flow, market design and human behavior. The model predictions can be used by policymakers to guide a transition to clean energy via distribution microgrids. The expectation is to enable at least 50 percent of electric power to come from renewable resources. This cannot be done with either the traditional grid, due to its limited capacity to accommodate intermittent renewable power sources, or with fully decentralized approaches, which would not be affordable for most utility customers.

Project Title: Multi-modality Image Fusion to Improve Coronary Revascularization in Patients with Stable Coronary Artery Disease

Investigator: Weihua Zhou

Sponsors: National Heart, Lung, and Blood Institute (NHLBI)/ National Institutes of Health (NIH)

Overview: This project aims to develop advanced multi-modality image fusion and machine learning technologies to improve clinical decision-making for coronary revascularization in patients with stable coronary artery disease (CAD). By integrating physiological information from gated SPECT myocardial perfusion imaging (MPI) with anatomical data from ICA, this project seeks to provide a comprehensive, patient-specific assessment of coronary artery lesions and myocardial function to better identify functionally significant stenoses.

Leveraging state-of-the-art medical image analysis, deep learning, graph-based modeling, and reinforcement learning, the proposed research will automate coronary artery extraction, generate accurate 3D SPECT–ICA fusion maps, and develop intelligent decision-support models to assist clinicians in selecting the most appropriate targets for revascularization. Built upon strong preliminary clinical validation and extensive prior work by the investigative team, this project addresses a critical unmet need in cardiovascular care by reducing user variability, improving reproducibility, and enhancing patient outcomes.

In addition to its clinical impact, the project integrates research with education by training undergraduate and graduate students in interdisciplinary data science, medical imaging, and informatics, thereby strengthening Michigan Tech’s role in translational biomedical research.

Project Title: Integrative analysis of electrical and mechanical dyssynchrony to improve cardiac resynchronization therapy

Investigator: Weihua Zhou

Sponsors: National Heart, Lung, and Blood Institute (NHLBI)/ National Institutes of Health (NIH)

Overview: Cardiac resynchronization therapy (CRT) is an established treatment for patients with heart failure; however, up to one-third of patients do not respond favorably to therapy. A major limitation of current clinical practice is that CRT candidacy and lead placement decisions are primarily based on electrical dyssynchrony assessed by ECG, which does not fully capture the underlying mechanical behavior of the failing heart. This project aims to develop an integrative, data-driven framework that combines electrical and mechanical dyssynchrony to improve patient selection, therapy planning, and outcome prediction for CRT.

Leveraging advanced medical imaging and machine learning techniques, this research will integrate electrical information from surface ECG and invasive electroanatomic measurements with mechanical dyssynchrony derived from gated cardiac imaging, particularly SPECT myocardial perfusion imaging. By jointly characterizing the spatial and temporal patterns of ventricular activation and contraction, the proposed methods will enable a more comprehensive, patient-specific assessment of ventricular synchrony. Built on extensive prior work by the investigative team in image-guided CRT, multimodality fusion, and quantitative cardiac imaging, this project seeks to reduce non-responder rates and improve long-term clinical outcomes.

Beyond its clinical impact, the project advances medical imaging and informatics by developing reproducible, automated analysis tools and interpretable machine-learning models for complex clinical decision-making. The research is tightly integrated with education, providing interdisciplinary training opportunities for undergraduate and graduate students in data science, biomedical engineering, and health informatics, and reinforcing Michigan Tech’s role in translational cardiovascular research supported by the NIH.

Project Title: AngioReady: A Digital Simulation Platform with Mixed Reality for Medical Trainees and Staff to Learn Invasive Coronary Angiography

Investigator: Weihua Zhou

Sponsor: Michigan Economic Development Corporation

Overview: AngioReady is a digital simulation and mixed-reality (MR) training platform designed to enhance education and workforce readiness in invasive coronary angiography (ICA) for medical trainees and clinical staff. ICA remains a cornerstone procedure in the diagnosis and management of coronary artery disease, yet training opportunities are limited by patient safety concerns, procedural complexity, and variability in case exposure. AngioReady addresses these challenges by creating an immersive, interactive learning environment that allows users to safely practice angiographic workflows, understand coronary anatomy, and develop procedural proficiency without risk to patients.

Building on advances in medical imaging, computer vision, large language models, and extended reality technologies, the platform integrates patient-specific coronary anatomy reconstructed from real angiographic data with physics-based procedural simulation and mixed-reality visualization. Trainees can interact with virtual catheters, C-arms, and contrast injections while receiving real-time feedback on imaging views, catheter positioning, and anatomical interpretation. The system supports progressive learning—from fundamental angiographic concepts to complex clinical scenarios—and enables objective assessment of technical skills and decision-making.

Project Title: Enabling the Future of Great Lakes Biological Assessment

Investigators: Timothy C. Havens, Michael Sayers, Ashraf Saleem

Sponsor: USGS

Overview: This project advances next-generation biological assessment of the Great Lakes by integrating autonomous platforms, advanced imaging, and machine learning to dramatically improve the scale, accuracy, and efficiency of ecosystem monitoring. Building on a strong USGS–MTU/MTRI partnership, the effort develops and deploys robotic sensing systems, automated image and acoustic analysis pipelines, and high-performance computational workflows to enable lake-wide surveys of fisheries, habitats, and water quality.

The project will deliver novel tools for interpreting underwater and remote sensing data, generate long-term ecological datasets, and produce open-source software and products accessible to resource managers and stakeholders. Collectively, these innovations will enhance decision-making for Great Lakes management by providing more comprehensive, timely, and actionable biological information.

Project Title: Robust Algorithms for Complex Autonomous Robot Systems

Investigator: Timothy C. Havens

Sponsor: US Navy

Overview: This project advances the state-of-the-art in robust autonomous robot systems by developing algorithms and system architectures that can withstand adversarial attacks and sensor disruptions in complex, real-world environments. Focusing on both terrestrial and marine platforms, the effort systematically investigates vulnerabilities across the autonomy stack—from sensing and perception to decision-making—and quantifies how uncertainty propagates through these systems.

The team will design and validate novel defense strategies, adaptive autonomy algorithms, and resilient sensing frameworks using a combination of high-fidelity simulation and real-world experimentation. Deliverables include a flexible autonomy codebase, a comprehensive “playbook” of attack and mitigation strategies, and a prototype robust autonomy architecture capable of maintaining performance under contested conditions. This work will enhance the reliability and operational effectiveness of military autonomous systems while training the next generation of engineers in resilient AI and autonomy.

Project Title: High-Frequency Radar in the Straits of Mackinac, Michigan

Investigators: Timothy C. Havens, Evan Lucas, Walker Nelson

Sponsors: Great Lakes Observing System / NOAA

More About High Frequency Radar Research

 

Project Title: Gener8tor Great Lakes Resilience Accelerator Testbed

Investigators: Hayden Henderson, Timothy C. Havens

Sponsors: Great Lakes Observing Systems / NOAA

Overview: The gener8tor Great Lakes Resilience Accelerator Testbed establishes a first-of-its-kind, end-to-end validation environment for emerging sensor technologies aimed at advancing environmental monitoring and resilience across the Great Lakes. Led by Michigan Technological University’s Great Lakes Research Center in partnership with GLOS, the testbed integrates state-of-the-art facilities, autonomous platforms, high-performance computing, and interdisciplinary engineering expertise to support startup companies from prototype through field deployment.

The effort will develop rigorous, application-driven evaluation protocols that account for real-world environmental conditions, data standards, and operational constraints, while leveraging a broad network of academic, federal, and industry partners. By accelerating the testing, validation, and commercialization of innovative sensing technologies, this initiative will strengthen regional observing systems and enable more informed, data-driven decision-making for Great Lakes resilience.

Visit the gener8tor Great Lakes Innovation Accelerator website for more information.