The Predictive Science Laboratory, founded in 2014 by Dr. Ilias Bilionis at Purdue University, advances scientific machine learning for engineering innovation.
Our research bridges mathematics, statistics, and engineering through Bayesian methods and causal modeling.
We develop predictive frameworks for uncertainty quantification and intelligent simulation across diverse scientific domains.
Led by Dr. Ilias Bilionis, Professor of Mechanical Engineering, our laboratory brings together a diverse and talented group of Research Scientists, Postdoctoral Researchers, and Graduate and Undergraduate students.
Our team’s expertise spans the cutting edge of modern engineering and data science, including Scientific Machine Learning, Uncertainty Quantification, and Physics-Informed Deep Learning. From advancing Medical Image Analysis and 4D Flow MRI to developing Agentic AI for scientific workflows and Organ-on-chip models, we collaborate to solve complex inverse problems and create intelligent predictive frameworks for the future.
Developing autonomous AI agents to automate complex scientific workflows—from simulation to analysis—allowing researchers to focus on decision-making rather than process management.
Our research is published in leading peer-reviewed journals and presented at top-tier international conferences. We focus on developing robust mathematical frameworks that integrate physical laws with data-driven insights.
Our projects apply Scientific Machine Learning and Bayesian Inference to solve high-impact challenges in medicine and aerospace. By integrating physical laws with advanced data analytics, we develop predictive frameworks that provide not just answers, but quantified uncertainty for critical decision-making.
Key Initiatives:
SMURF: Unsupervised 4D Flow MRI Reconstruction – Integrating cardiac geometry and velocity fields directly from MRI data for advanced hemodynamics.
Aneurysm Growth Detection – A probabilistic Bayesian framework for detecting intracranial aneurysm growth from longitudinal medical imaging.
Spacecraft Thermal Protection – Calibrating thermal conductivity models for mission-critical aerospace systems.
