Nabil Kahouadji
Northeastern Illinois University, IL·Mathematics & Statistics
Where people live, and what they are exposed to there, may shape their risk of disease. Emerging evidence suggests that neighborhood stresses, such as exposure to violence, can trigger a lasting toxic stress response that drives biological changes linked to cancer.
- Fields
Mathematics & Statistics
Public Health & Epidemiology
Computer Science
- Topics
machine learning
predictive modeling
triple discriminant scoring
- Grants
- 1 known grant, through Jul 2027
What they work on
- Test the triple discriminant scoring methodology, a classification method built on many simulations across subsets of variables, for predicting colorectal cancer outcomes from Chicago electronic medical records and socio-environmental data.
- Apply the same machine learning framework to predict lung cancer outcomes, comparing its accuracy and bias with Extreme Gradient Boosting (XGBoost) and neural networks.
- Predict postpartum health outcomes from electronic medical records and neighborhood data, and test how exposure to violence relates to health risk.
- Run a mathematical, statistical and empirical comparison of classification methods across the three datasets, including how robust each is when the training data distribution shifts.
More about the research
This project asks whether machine learning (computer methods that learn patterns from data) can predict cancer and other health outcomes accurately and fairly. The goal is to help people assess their own risk and catch disease earlier.
Why it matters
More accurate and less biased prediction models could help Chicago patients and underserved communities find out about colorectal cancer, lung cancer and postpartum health risks sooner and catch disease earlier. The work may also clarify how exposure to violence and other neighborhood factors affect health. It could inform fairer use of predictive modeling in healthcare more broadly.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Daniela Raicu
DePaul University, IL·Computer Science
Develop a Semantic Deep-learning Neural Network (SDNN), a deep-learning model enhanced with knowledge of what visual features mean, that generates visual patterns and example images resembling the concepts radiologists use to read lung nodules.
Jeanne Shen
Stanford University, CA·Medicine & Clinical Research
Build a publicly available, expert-annotated multimodal colorectal cancer dataset. It will combine pathology whole-slide images, preoperative CT and MRI scans, and structured electronic medical record data.
Fergal Fleming
University of Rochester, NY·Medicine & Clinical Research
Validate whether ICD-10 codes Z65.1 and Y92.14x and an admission-source code accurately identify incarcerated patients in electronic medical records and claims databases.
The school
Northeastern Illinois University: 1 labs on PI Pages.
Reach out
A start for your email. Fill in the brackets.