Dear website visitor: for any access requests/demos, please email ioannisk@uh.edu, and we will send you instructions and login information.

Active Projects

Title: AI-Response - Developing AI-powered tools to optimize emergency food distribution by processing real-time reports from charities and coordinating resources during disasters

Website: proj002.cbl-uh.org

Description: The project creates an AI-driven platform to improve emergency food distribution. It collects real-time reports from charities, analyzes demand and supply, and coordinates deliveries to ensure food reaches those in need efficiently. By leveraging AI, the system predicts shortages, identifies high-need areas, and optimizes resource allocation during disasters. The platform aims to reduce waste, accelerate response times, and support equitable access to food for affected communities.

Sponsor: NIFA

Title: Financial and Network Disruptions in Illicit and Counterfeit Medicines Trade

Website: ai-snips.io

Description: AI-SNIPS is a forward-thinking initiative and comprehensive applied research project designed to combat illicit drug sales through innovative methods. The critical components of the project include web scraping, machine learning, and network analysis, each contributing to a robust data pipeline. AI-SNIPS seeks to provide a unique and physics-based perspective for identifying and disrupting networks of illicit pharmaceutical sellers. By combining web scraping, machine learning, and network analysis, the project aims to uncover patterns and connections that may not be apparent through traditional methods. AI-SNIPS emphasizes the importance of clear communication through data visualization by producing visually compelling representations of the modeling decisions made during the analysis. This will enable stakeholders, including pharmaceutical companies and law enforcement, to quickly understand the insights derived from the data. Lastly, AI-SNIPS recognizes the dynamic nature of illicit networks and incorporates performance analysis into its framework. By continually assessing the network dynamics, the project aims to enhance its machine learning models, ensuring they remain effective in identifying optimal points for disruption.

Sponsor: NSF

Title: CRASA - Community Responsive Algorithms for Social Accountability

Website: crasa.ai

Description: CRASA is a multidisciplinary, community-based participatory research program to develop an algorithm accountability benchmark to meet societal and legal needs and guide best practices. The research program involves several key steps. Firstly, stakeholders are interviewed, and a Community Advisory Board is established to gain insight into public policy needs. Legal frameworks have been reviewed to shape accountability for AI algorithms. Following this, an algorithm accountability benchmark is being developed that outlines specific standards for the community, auditors, and legal evaluation. Behavioral experiments are conducted to ensure alignment between the algorithm design and social concerns. Finally, software scoring tools are being developed based on the accountability benchmark to assess public policy applications such as criminal recidivism assessment and facial recognition.

Sponsor: NSF

Title: Healthcare Dataset Harmonization

Website: http://3.89.127.155

Description: The Healthcare Dataset Harmonization project is a pioneering web-based platform that semi-automates the harmonization and linking of EHR datasets through a human-in-the-loop framework, guiding users with large language models (LLMs). Our solution is a two-stage harmonization pipeline that keeps schema metadata processing online while handling patient-level data locally, to align with HIPAA data privacy principles. In the first stage, users harmonize and link only non-identifiable schema information. In the second stage, sensitive value-level harmonization occurs entirely on the user’s system, so no private and protected health information ever leaves their environment. Across both stages, we expect LLM-powered suggestions to accelerate the harmonization and linking processes.

Sponsor: National Center for Advancing Translational Sciences (NCATS), NIH

Title: FAIR Principles AI Chatbot

Website: FAIR Principles AI Chatbot

Description: The FAIR Principles AI Chatbot project is a pioneering platform dedicated to advancing FAIR (Findable, Accessible, Interoperable, Reusable) data principles for researchers, educators, and students in scientific research and data management. It features an AI-powered chatbot leveraging Retrieval-Augmented Generation (RAG) architecture to deliver personalized, contextually relevant guidance, explanations, examples, and implementation strategies for applying FAIR standards in real-world workflows, ensuring responses are grounded in up-to-date domain-specific knowledge. Complementing the chatbot, the site curates high-quality resources, including courses, blogs, videos, and podcasts, to support diverse learning needs and foster a collaborative community committed to making research outputs more discoverable, accessible, and reusable across disciplines.

Sponsor: National Center for Advancing Translational Sciences (NCATS), NIH

Expired Projects

Title: AI-FEED - Connecting Food Charities to End Hunger: Artificial-Intelligence-Based Decision Support for Equitable Food and Nutrition Security in the Houston Area

Website: ai-feed.ai

Description: Combating Food Insecurity Through AI: Although a well-resourced nation, 12% of Americans lack nutritious food to live a healthy life. AI-FEED is an innovative artificial intelligence platform to fix the broken food charity ecosystem. Beginning in Texas and moving across the U.S., AI-FEED will be a game-changer for significant action to support nutrition security. It connects food charities, donors, clients, and community leaders to optimize food resources and support the flourishing of the greater community.

Sponsor: NSF

Title: Differentiation among Multisystem Inflammatory Syndrome in Children, Kawasaki Disease, Endemic Typhus, and Non-Specific Febrile Illnesses in Pediatric Patients

Website: aiheat.ai

Description: The AI-HEAT project represents the second iteration and evolution of our pioneering AI-MET framework, advancing Deep Learning–based clinical decision support to distinguish among four clinically overlapping pediatric febrile conditions: Multisystem Inflammatory Syndrome in Children (MIS-C), Kawasaki disease, Endemic Typhus, and non-specific febrile illnesses. Building directly on AI-MET's success in differentiating MIS-C from typhus using early Emergency Department data, AI-HEAT expands this capability with a compact Triplet Loss Siamese Network architecture that leverages just eight routinely available clinical and laboratory features within the first six hours of presentation. These disorders share deceptive early symptoms such as fever, rash, conjunctivitis, and gastrointestinal issues, yet require distinct treatments and carry different risks; AI-HEAT empowers medical teams with timely, probabilistic diagnostics to minimize misclassification, optimize interventions, and improve critical pediatric outcomes.

Sponsor: NIH

Title: Artificial-Intelligence-Based Decision Support for Equitable and Resilient Food Distribution During Pandemics and Extreme Weather Events

Website: ai-serve.org

Description: The AI-SERVE project stands at the forefront of innovation, aiming to transform the emergency management landscape. Leveraging the power of Artificial Intelligence, AI-SERVE seeks to revolutionize how we approach food distribution during pandemics and extreme weather events. Its mission is clear: to bring automation technology to the heart of crisis response, ensuring equitable and resilient food distribution for all, especially those in dire need.

Sponsor: NSF

Title: AIM-AHEAD Resource Center of Excellence at UH for Data Curation, Linkages, and Harmonization of Datasets

Website: arch-d2h.ai

Description: The project aims to establish the AIM-AHEAD Resource Center of Excellence at the University of Houston (UH) for Data Curation, Linkages, and Harmonization of Datasets. We will assess UH's data and infrastructure capacity, provide AI/ML training for staff, and build multidisciplinary and multi-institution partnerships.

Sponsor: NIH

Title:

Website: ireadusa.co

Description: iReadUSA is an AI-powered project that compiles and verifies health data from various sources. It creates a website tailored to the specific needs of a Texas county (or the entire USA), ensuring vulnerable populations have access to accurate health information.

Sponsor: NIH