The CHIME-in volunteer program matches students with public health agencies that need help with informatics and data science projects. Some external partners have included the World Health Organization (WHO), Pan-American Health Organization (PAHO), Illinois Department of Public Health (IDPH), Uganda Ministry of Health, and Champaign-Urbana Public Health District (CUPHD).
Our current projects include but may not be limited to the following. If you are interested in joining a specific research team or in learning more about volunteering for the Center for Health Informatics, please complete our Student Interest Form (external link).
AI Tools for Managing Infodemics (Health Misinformation in Video-based Social Media)
An infodemic is “too much information including false or misleading information in digital and physical environments during a disease outbreak” (WHO). CHI is partnering with WHO to address a need to identify and combat current and future infodemics online.
In this multi-semester project, student researchers are building and strengthening tools to detect misinformation in video and image-based social media. The goal is to provide WHO and partner agencies with scalable, trustworthy pipelines for monitoring misinformation without requiring humans to manually watch hours of content.
Suggested skills and interests for research volunteers:
- Prior machine learning or data science coursework is recommended but not required.
- Experience managing APIs
- UX/UI design
- Natural language processing in languages other than English
- Evaluating predictive models
- Familiarity with ethical issues regarding public health data
Global Health Ministry Online Presence Assessment
During COVID, WHO discovered many Ministries of Health lacked functional online communication. UIUC’s Center for Health Informatics is partnering with WHO to assess the online presence of these Ministries of Health.
This project applies a survey instrument to systematically evaluate the web and social presence of ministries worldwide. Students will research whether ministries maintain websites, provide timely updates, and communicate in their national languages. The long-term goal is to build a global dataset highlighting strengths and gaps.
Suggested skills and interests for research volunteers:
- Basic Excel knowledge
- Social media
- Good internet research skills
- Experience conducting web surveys and analyzing survey data
- Proficiency or fluency in non-English languages
Detecting Drug Supply Chain Risks in the Americas
This exploratory project asks whether signals of drug shortages can be identified through social media or other nontraditional sources. Students will scan posts, scholarly research, government reports, and community complaints to spot patterns that could predict supply disruption (e.g., stockouts of insulin or antibiotics). Insights will be shared with Pan-American Health Organization (PAHO) officials for decision support.
Suggested skills and interests for research volunteers:
- Text analysis and descriptive statistics
- Social media mining, keyword/hashtag development
- Exploratory data analysis and dashboards
- Working with messy datasets
- Creativity and problem-solving orientation
Systematic Literature Review on AI in Epidemic Management
In this project, a student or students will conduct a structured literature review of AI applications in epidemic management. Students will develop a search strategy, apply inclusion/exclusion criteria, chart their findings (tasks, models, gaps), and synthesize results into a paper for possible publication.
Suggested skills and interests for research volunteers:
- Strong academic writing skills
- Figure/table creation for publication
- Attention to proper citation practices
Drug Shipping Cost Prediction Tool
PAHO buys drugs in bulk for member states but previously used rough percentage rates (+20% of drug cost) to estimate drug shipping costs. Students on this project team will (1) refine models using new features (distance, ports, fuel indexes, congestion), and (2) productionize the models into a usable software tool (API/web app) that PAHO and member states can use directly.
Suggested skills and interests for research volunteers:
- Predictive modeling
- Software engineering for tools
- Data visualization and dashboards
- Coursework in statistics/ML recommended
Forecasting Drug Demand under Climate and Migration Pressures
This project models how external pressures—like climate change and migration—will affect national drug needs. Teams will combine migration and climate data with drug utilization and develop predictive models to estimate demand under various scenarios.
Suggested skills and interests for research volunteers:
- Predictive/time series modeling
- Geospatial analysis and migration flows
- Communication of uncertainty and scenario analysis
Data Cleaning and Standardization of National Drug Stockpiles
PAHO member states report stockpiles in messy free-text formats. This project aims to standardize data into a usable database for PAHO. Students will develop normalization methods using fuzzy matching, unit harmonization, and ontology mapping. Exploration of LLMs for semi-automated mapping will also be tested with human oversight.
Suggested skills and interests for research volunteers:
- Multilingual handling (Spanish/Portuguese/English)
- Text normalization (regex, fuzzy matching libraries)
- Patience with “messy data” problems
Extreme Weather, Climate, and Health Outcomes in Illinois
This is a flagship project with multiple subteams, working with the Illinois Department of Public Health. Students will explore how extreme weather (heat, cold, floods, poor air quality) impacts morbidity and mortality in Illinois. The aim is to create predictive models that local hospitals can use to anticipate surges. Subteams focus on all-cause mortality, equity effects, heat metrics, and air quality impacts (e.g., wildfire smoke). A secondary stream explores how to leverage social media as an early-warning layer.
Suggested skills and interests for research volunteers:
- Epidemiology and public health analytics
- Predictive/time series modelling
- Handling sensitive data responsibly (secure environments)
- Statistics background recommended but not required
Social Media Insights into Climate and Health Discourse
This project investigates whether Illinois residents’ online discussions of climate, weather, and health provide actionable signals. Students will identify relevant conversations, categorize sentiment and themes, and align spikes with events (e.g., heat waves, wildfire smoke days).
Suggested skills and interests for research volunteers:
- Social media searching/APIs
- NLP for sentiment/topic modeling
- Dashboard creation and reporting
- Exploratory, open-ended research
- Curiosity and adaptability
GIS Mapping of Disease Prevalence and Specialist Access
The aim is to produce state maps that visualize disease prevalence (e.g., diabetes rates) alongside distance to specialists (e.g., endocrinologists). This helps IDPH identify underserved areas and plan recruitment or subsidies for physicians.
Suggested skills and interests for research volunteers:
- GIS software
- Spatial analysis (drive-time, buffers)
Opioid Overdose Surveillance in Champaign-Urbana
This project explores ways to detect clusters of opioid overdoses, despite reduced ER data visibility due to Narcan distribution. Students will integrate multiple data sources, test geospatial/time-series methods, and create visualizations or alerting mechanisms for CUPHD to improve local interventions.
Suggested skills and interests for research volunteers:
- Descriptive statistics and time-series analysis
- Geospatial methods (cluster detection, scan statistics)
- Public health data integration
- Secure data handling requirements
AI-Powered Training for Medical Education
This project applies LLMs to support medical students’ interview training. Models will analyze patient interview transcripts to check whether essential questions were asked, highlight gaps, and provide feedback. A key focus is migrating the pipeline from English to Portuguese, requiring multilingual adaptation.
Suggested skills and interests for research volunteers:
- LLM prompting and evaluation
- ASR pipeline integration
- Educational technology and rubric design
- Bilingual skills (English/Portuguese) are a strong advantage
- Ethical awareness in education/medical contexts
Generative AI for Personalized Nutrition in Elderly Care
Many elderly patients entering long-term care come from non-Western backgrounds with unfamiliar diets. This project builds a generative AI tool that creates photorealistic food images from verbal descriptions, enabling dietitians to recognize foods and analyze nutritional content. Work includes UI prototyping and testing with real-world dietitians.
Suggested skills and interests for research volunteers:
- Generative AI image models
- Prompt design and dataset curation
- Basic nutrition knowledge
- Sensitivity with regard to nutrition, food, and culture
- UI/UX prototyping and usability testing