Welcome to the AUDITION (AUtism DIgital TwIn fOuNdations) project
This project aims to develop the mathematical foundations for a digital twin (DT) system for individuals with autism spectrum disorder (ASD), focusing on dynamic modeling, prediction, uncertainty quantification, and treatment or intervention recommendation through DT-based optimization. This project is supported by NSF award at the George Washington University (NSF DMS-2436216) and George Mason University (NSF DMS-2436217).
The specific goals of this project include:
- Develop computational models based on conditional variational auto-encoders (CVAE) and longitudinal CVAE to analyze brain activities, and neurodevelopmental processes.
- Create a novel bilevel formulation for fine-tuning foundational models to predict ASD outcomes.
- Develop a model-free conformal prediction procedure to ensemble predictions, integrating various types of uncertainties.
- Develop a DT-based reinforcement learning framework for personalized treatment plans to improve clinical outcomes.