Nonlinear Dynamics and Machine Learning for Accurate Detection of Early-stage Atrial Fibrillation


Guest post by Dr. Changqing Cheng, Binghamton University, State University of New York

This Success Story is a report on the results of the Northeast Big Data Innovation Hub’s 2020 Seed Fund program.


The overarching goal of this Seed Fund project was to develop an integrated platform to integrate nonlinear dynamics analysis and data science for incipient-stage Atrial Fibrillation (AF)  detection. 

With the support from this Seed Fund award, the PI and his team accomplished the following: (1) designed data science algorithms to process electrocardiogram (ECG) data and unbalanced data learning; (2) developed a Mobile App to monitor cardiac dynamics, which was ranked the second place in the Mobile App Competition organized by the Institute of Industrial and Systems Engineers (IISE)  in May 2021; and (3) engaging in development of a master’s thesis to integrate the developed algorithm on AF detection. 

This project has kickstarted other initiatives, including the formation of a new research team and the beginnings of a new NSF proposal, which will mention the preliminary results of this Seed Fund initiative.

Several publications have been generated from this Seed Fund research:

Y. Shu, W. Dan and C. Cheng, “Spatiotemporal regularization in effective reconstruction of epicardial potential,” IEEE EMBS International Conference on Biomedical and Health Informatics (BHI), 2021. 

Y. Che, Z. Guo, and C. Cheng, “Generalized polynomial chaos-informed efficient stochastic kriging,” Journal of Computational Physics, Vol. 445, 110598, 2021. DOI: https://doi.org/10.1016/j.jcp.2021.110598

Y. Che and C. Cheng, “Active learning and relevance vector machine in efficient estimate of basin stability for large-scale dynamic networks,” Chaos: An Interdisciplinary Journal of Nonlinear Science, Vol. 31, 053129, 2021. DOI: https://doi.org/10.1063/5.0044899

Y. Shu, C. Cheng, T. G. Smith, “A novel mobile application for Tele-ICU monitoring using electrocardiographic imaging (ECGI),” 21st Triennial Congress of the International Ergonomics Association, Vancouver, June 13 – 18, 2021.


Lead PI: Changqing Cheng

Changqing Cheng is an Assistant Professor in the Department of Systems Science and Industrial Engineering at State University of New York at Binghamton.

Website: https://ccheng686.wixsite.com/ccheng