Projects

Court Watch NOLA is a non-profit organization dedicated to promoting transparency, equity, and justice in the criminal court system. By training volunteers to observe and report on thousands of criminal court cases a year, Court Watch NOLA works to ensure judges, prosecutors, public defenders, sheriff deputies, police officers, and other criminal justice actors are doing their jobs professionally, transparently, fairly, and economically.
We work with Court Watch NOLA to build AI systems that collect, search, and report on criminal court data to improve transparency and oversight.
2023-2024 | 2022-2023 Poster | 2021-2022 Poster

sawt.us is an AI system that provide intelligent access to City Council transcripts in order to help citizens be better informed about important issues facing them.
2023-2024 Poster

The Data Center is a fully independent, neutral nonprofit organization that brings together data together from multiple sources to support rigorous analysis on issues that matter most to government, business, nonprofit, and community leaders in Southeast Louisiana.

The City of New Orleans Office of Information Technology and Innovation facilitates effective, cost efficient use of technology by spearheading the assessment and deployment of technology based business management solutions, and service delivery strategies. BlightWatch NOLA is a project focused on publicly available blighted property data in New Orleans. Blight, also known as urban decay, refers to properties that have been abandoned by their owners and left to decay. This includes homes, commercial buildings, and even empty lots. Cities experience blight primarily due to economic distress, natural disasters, and depopulation in general. While blight is a symptom of larger societal problems, its presence also causes problems for residents where it is prevalent. Economically, blighted properties decrease the property values of nearby residents and discourage investment in effected neighborhoods.

Families Helping Families is a family-directed resource center that provides information and referral, training and education, and peer-to-peer support on issues related to disability. FHF of GNO is also home to the Louisiana Parent Training and Information Center, a federal education grant that provides training and support to families throughout Louisiana on special education and transition topics.
Our work with FHF was to develop a chatbot tool to enable easier access to information for parents of children with disabilities.
2021 Poster

PI: Ibrahim Demir, Michael A. Fitts Presidential Chair in Environmental Informatics and Artificial Intelligence, Professor, Tulane University, River-Coastal Science and Engineering
Project Overview: This project will design and launch an open-access platform where communities can directly engage with artificial intelligence in urban and environmental decision-making. Still in its early prototype stage, the project will use this grant to build out core features, develop a library of real-world scenarios, and involve community partners in shaping how the system works.
When complete, CrowdSimAI will allow anyone—residents, advocacy groups, policymakers, and educators—to simulate how diverse “crowds” of AI agents deliberate on civic dilemmas, from flood mitigation trade-offs to green infrastructure planning. By openly sharing data, results, and community feedback, the project aims to increase transparency in AI, support public learning, and give communities a stronger voice in how technology shapes civic life.

PI: Martin Nwadiugwu, PhD student, Biomedical Science, Tulane University School of Medicine
Project Overview: This project will use artificial intelligence to uncover the earliest molecular signatures of Alzheimer’s disease (AD). Led by Tulane School of Medicine researcher Martin Nwadiugwu, the project will analyze single-nucleus RNA sequencing data from post-mortem brain samples, alongside clinical and cognitive assessments, to identify the key genes and regulatory networks linked to AD.
With this award, the team will build machine learning models capable of predicting cognitive decline from gene expression data—moving beyond traditional symptom-based assessments. The project’s goal is to lay the groundwork for earlier detection, more precise risk prediction, and future diagnostic tools that could be adapted for both advanced clinical environments and resource-limited settings. By integrating biomedical data with AI, this research aims to improve how we diagnose, manage, and ultimately intervene in Alzheimer’s disease.

PI: Xin Jiang, Associate Professor of Sociology, Tulane University
Project Overview: This project explores how digital platforms are reshaping migration and community life. Led by Tulane researchers, the study focuses on the experiences of undocumented Chinese migrants, a population that has grown rapidly in recent years. The team will analyze more than half a million Telegram messages from migrant networks using multimodal AI models, while also conducting in-depth interviews with migrants in Louisiana and Texas.
With this award, the researchers will combine cutting-edge data science with traditional fieldwork to better understand how migrants access information, emotional support, and community ties in both online and offline spaces. Partnering with the American Chinese United Association (Louisiana Chapter), the project will also share findings in Chinese and English with local organizations, aiming to inform community services, safety efforts, and policy discussions.

PI: Liam Guest, MS Student, Public Health, Tulane University
Project Overview: AURA (AI for Urban Resilience & Alerts) is a project to develop GulfGuard—a public-facing app designed to help New Orleans residents prepare for hurricanes. Using open data from FEMA, NOAA, the U.S. Census, and the City of New Orleans, the project will build an explainable AI model that assigns transparent risk scores to neighborhoods. These scores will then power an easy-to-use app that delivers tailored preparedness resources, evacuation routes, and real-time alerts.
With this award, the project team will develop the prototype, host co-design workshops with local partners such as NOLAReady and the Lower 9th Ward Center for Sustainable Engagement and Development, and incorporate community feedback into the tool’s final design. GulfGuard aims to save lives, strengthen local resilience, and serve as a model for other coastal cities facing climate risks.

PI: Michael Hoerger, Associate Professor of Psychology and Psychiatry, Tulane University School of Medicine and School of Science and Engineering
Project Overview: SCOPE-AI (Stakeholder-Centered Optimization of Predictive Epidemiology using Artificial Intelligence) builds on the widely used Pandemic Mitigation Collaborative (PMC) COVID-19 Dashboard, which provides real-time transmission data for patients with high-risk medical conditions and their families. With this award, the team will use AI to enhance the dashboard’s forecasting accuracy and incorporate stakeholder feedback to make the tool more transparent, reliable, and useful.
The updated model will help patients and families better plan essential activities around waves of COVID-19 transmission—supporting decisions like when to schedule medical appointments, social visits, or travel. By combining advanced modeling with direct input from patients, caregivers, and clinicians, SCOPE-AI will both improve public health preparedness and strengthen community trust in AI-driven epidemiology.

PI: MD Mostafized Rahman, Postdoctoral Scholar, Computer Science, Tulane University
Project Overview: AIPTA is an AI-powered assistant for evidence-based physical therapy. By grounding recommendations in clinical guidelines and testing in senior care communities, the project aims to improve patient outcomes and support therapists with trustworthy, personalized care tools.

PI: Fallon Aidoo, Assistant Professor of Real Estate & Historic Preservation, Tulane University, School of Architecture
Project Overview: This project aims to develop AI tools to identify and protect unregistered national historic landmarks from real estate development threats, preserving cultural heritage and history. The project will use AI to map and analyze the development of historic properties in the Oak Bluffs Highlands Heritage Project of the African American Heritage Trail of Martha’s Vineyard. By developing AI-enhancements to text recognition and table analysis software, the team will guide community-driven preservation of that ethnic heritage.

PI: Jordan Karubian, Professor, Tulane University, Department of Ecology & Evolutionary Biology
Project Overview: This initiative leverages AI to aid local conservation efforts in one of the world's most biodiverse areas, enhancing the protection of critical ecosystems. The project will co-develop a localized forest monitoring system called Chocó Forest Watch with Fundación para la Conservación de los Andes Tropicales (FCAT), an Ecuadorian grassroots NGO that manages a community-run reserve in the highly threatened Chocó rainforests of Ecuador. Through a human-centered design approach that engages local community members, the project will create a user-friendly, low-cost, and locally-adapted tool that supports FCAT to monitor and respond to deforestation.

PI: Audrey Hang Hai, Assistant Professor, Tulane University, School of Social Work
Project Overview: This research seeks to integrate AI into existing interventions to provide enhanced support for young adults struggling with substance use disorders, aiming to improve outcomes and accessibility. Partnering with the CADA Prevention & Recovery Center, the team will conduct focus groups to assess attitudes towards the use of AI to help guide individuals to available resources to deal with substance use disorders.