Research Areas

The CSDA Lab develops statistical learning, trustworthy AI, and decision-support methodologies for applications in health, engineering, and society. Our research combines statistical rigor, uncertainty quantification, machine learning, and domain expertise to improve prediction, monitoring, and decision making in complex systems.

Research Themes

Our work is organized around three interconnected research themes:

  • Monitoring, quality analytics, and uncertainty quantification
  • Precision health and biomedical analytics
  • Trustworthy AI and decision-support systems

Together, these themes support the development of reliable, interpretable, and data-driven solutions for complex real-world problems.

Research Program I: Monitoring, Quality Analytics, and Uncertainty Quantification

  1. Statistical process monitoring and quality control
  2. Missing data and imputation effects on Hotelling T² monitoring procedures
  3. Monitoring mass spectrometry and high-dimensional biological data
  4. Monitoring IoT and streaming-data environments
  5. Design and evaluation of multivariate monitoring systems
  6. Wavelet-based statistical methodologies
  7. Uncertainty quantification and robust statistical inference

Research Program II: Precision Health and Biomedical Analytics

  1. Cancer Survivorship Analytics
  2. ACL injury prediction and rehabilitation analytics
  3. Predictive models for clinical trial enrollment
  4. Uncertainty quantification for cancer flow cytometry
  5. Causal inference using All of Us Research Program data
  6. Cancer, metabolic syndromes, mental health, and vaccination analytics . Ecological and environmental predictive analytics

Research Program III: Trustworthy AI and Decision Support Systems

  1. Retrieval-Augmented Generation (RAG) systems
  2. Agentic AI and multi-agent architectures
  3. AI-enabled advising and decision-support systems
  4. Large language model applications
  5. 3D point-cloud segmentation and classification
  6. Engineering and computational intelligence applications

Student Research Projects

Statistical Methodology

  • Confidence intervals for binomial proportions
  • Second-generation p-values
  • Latent class analysis and sample size determination
  • Conformal Prediction set size random variable

Machine Learning and Predictive Analytics

  • Cherry blossom prediction using machine learning and regression
  • Conformal Predictions
  • Load forecasting models
  • R package development
  • R Shiny application development

The CSDA Lab encourages undergraduate and graduate students to participate in interdisciplinary research, conference presentations, software development, and collaborative publications. Students work closely with faculty and external collaborators to develop both methodological and applied research skills.

Student UWF Grant Resources