Physical-statistical modeling and regularization of high-dimensional dynamical systems

Bing Yao, Hui Yang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This paper presents a novel physics-driven spatiotemporal regularization (STRE) method for high-dimensional predictive modeling. This model not only captures the physics-based interrelationship between time-varying explanatory and response variables that are distributed in the space, but also addresses the spatial and temporal regularizations to improve the prediction performance. The STRE model is implemented to predict time-varying distributions of electric potentials on the heart surface based on the electrocardiogram (ECG) data on the body surface. The model performance is evaluated and validated in both a simulated two-sphere geometry and a realistic torso-heart geometry. Experimental results show that the STRE model significantly outperforms other models widely used in current practice such as Tikhonov zero-order, Tikhonov first-order and L1 first-order methods.

Original languageEnglish (US)
Title of host publication67th Annual Conference and Expo of the Institute of Industrial Engineers 2017
EditorsHarriet B. Nembhard, Katie Coperich, Elizabeth Cudney
PublisherInstitute of Industrial Engineers
Pages710-715
Number of pages6
ISBN (Electronic)9780983762461
StatePublished - Jan 1 2017
Event67th Annual Conference and Expo of the Institute of Industrial Engineers 2017 - Pittsburgh, United States
Duration: May 20 2017May 23 2017

Publication series

Name67th Annual Conference and Expo of the Institute of Industrial Engineers 2017

Other

Other67th Annual Conference and Expo of the Institute of Industrial Engineers 2017
CountryUnited States
CityPittsburgh
Period5/20/175/23/17

All Science Journal Classification (ASJC) codes

  • Industrial and Manufacturing Engineering

Fingerprint Dive into the research topics of 'Physical-statistical modeling and regularization of high-dimensional dynamical systems'. Together they form a unique fingerprint.

Cite this