Causal simulation of physiological systems

Interdisciplinary Areas: Engineering and Healthcare/Medicine/Biology, Data/Information/Computation

Project Description

This project aims to study the causal model of physical systems, such as fluid dynamics models using structure causal models (SCM). Structure causal model requires the problem domain represented by a graphical representation, where nodes represent variables and directional edges represent cause-effect relationship. More specifically, the project will study simulation in causal inference with networks-based structural equation models [1]. One particular area of the study will be reviewing the simcausal package in R [2], and establish protocols and graphical models for generating data using fluid dynamics models of heart. The goal is to better understand the mechanism of heart and generate data from the graphical model similar to the distribution of the data generating mechanism. 

Start Date

Summer/Fall 2019

Postdoc Qualifications

PhD in Computer Science or Statistics or related quantitative field 


Paul Griffin, Professor, Industrial Engineering

Pavlos Vlachos, Professor, Mechanical Engineering
1. J. Pearl, Causality: Cambridge university press, 2009.
2. E. Bareinboim and J. Pearl, "Causal inference and the data-fusion problem," Proceedings of the National Academy of Sciences, vol. 113, pp. 7345-7352, 2016.
3. M. Bikak, M. Adibuzzaman, Y. Jung, Y. Yih, and E. Bareinboim, "Regenerating evidence from landmark trials in ARDS using Structural Causal Models on Electronic Health Record," in American Thoracic Society Conference (ATS 2018), 2018.
4. O. Sofrygin, R. Neugebauer, and M. J. van der Laan, "Conducting Simulations in Causal Inference with Networks-Based Structural Equation Models," arXiv preprint arXiv:1705.10376, 2017.
5. O. Sofrygin, M. J. van der Laan, and R. Neugebauer, "simcausal: Simulating Longitudinal Data with Causal Inference Applications," R package version 0.4, URL http://CRAN. R-project. org/package= simcausal, 2015