Danny Smyl , Tyler N. Tallman , Laura Homa , Chenoa Flournoy , Sarah J. Hamilton , John Wertz
{"title":"Physics Informed Neural Networks for Electrical Impedance Tomography","authors":"Danny Smyl , Tyler N. Tallman , Laura Homa , Chenoa Flournoy , Sarah J. Hamilton , John Wertz","doi":"10.1016/j.neunet.2025.107410","DOIUrl":null,"url":null,"abstract":"<div><div>Electrical Impedance Tomography (EIT) is an imaging modality used to reconstruct the internal conductivity distribution of a domain via boundary voltage measurements. In this paper, we present a novel EIT approach for integrated sensing of composite structures utilizing Physics Informed Neural Networks (PINNs). Unlike traditional data-driven only models, PINNs incorporate underlying physical principles governing EIT directly into the learning process, enabling precise and rapid reconstructions. We demonstrate the effectiveness of PINNs with a variety of physical constraints for integrated sensing. The proposed approach has potential to enhance material characterization and condition monitoring, offering a robust alternative to classical EIT approaches.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"188 ","pages":"Article 107410"},"PeriodicalIF":6.0000,"publicationDate":"2025-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Neural Networks","FirstCategoryId":"94","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0893608025002898","RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE","Score":null,"Total":0}
引用次数: 0
Abstract
Electrical Impedance Tomography (EIT) is an imaging modality used to reconstruct the internal conductivity distribution of a domain via boundary voltage measurements. In this paper, we present a novel EIT approach for integrated sensing of composite structures utilizing Physics Informed Neural Networks (PINNs). Unlike traditional data-driven only models, PINNs incorporate underlying physical principles governing EIT directly into the learning process, enabling precise and rapid reconstructions. We demonstrate the effectiveness of PINNs with a variety of physical constraints for integrated sensing. The proposed approach has potential to enhance material characterization and condition monitoring, offering a robust alternative to classical EIT approaches.
期刊介绍:
Neural Networks is a platform that aims to foster an international community of scholars and practitioners interested in neural networks, deep learning, and other approaches to artificial intelligence and machine learning. Our journal invites submissions covering various aspects of neural networks research, from computational neuroscience and cognitive modeling to mathematical analyses and engineering applications. By providing a forum for interdisciplinary discussions between biology and technology, we aim to encourage the development of biologically-inspired artificial intelligence.