Label-Free Physics-Informed Image Sequence Reconstruction with Disentangled Spatial-Temporal Modeling

Xiajun Jiang, Ryan Missel, Maryam Toloubidokhti, Zhiyuan Li, Omar Gharbia, John L. Sapp, Linwei Wang

Producción científica: Capítulo en Libro/Reporte/Acta de conferenciaContribución a la conferencia

5 Citas (Scopus)

Resumen

Traditional approaches to image reconstruction uses physics-based loss with data-efficient inference, although the difficulty to properly model the inverse solution precludes learning the reconstruction across a distribution of data. Modern deep learning approaches enable expressive modeling but rely on a large number of reconstructed images (labeled data) that are often not available in practice. To combine the best of the above two lines of works, we present a novel label-free image reconstruction network that is supervised by physics-based forward operators rather than labeled data. We further present an expressive yet disentangled spatial-temporal modeling of the inverse solution, where its latent dynamics is modeled by neural ordinary differential equations and its emission over non-Euclidean geometrical domains by graph convolutional neural networks. We applied the presented method to reconstruct electrical activity on the heart surface from body-surface potential. In simulation and real-data experiments in comparison to both traditional physics-based and modern data-driven reconstruction methods, we demonstrated the ability of the presented method to learn how to reconstruct using observational data without any corresponding labels.

Idioma originalEnglish
Título de la publicación alojadaMedical Image Computing and Computer Assisted Intervention – MICCAI 2021 - 24th International Conference, Proceedings
EditoresMarleen de Bruijne, Marleen de Bruijne, Philippe C. Cattin, Stéphane Cotin, Nicolas Padoy, Stefanie Speidel, Yefeng Zheng, Caroline Essert
EditorialSpringer Science and Business Media Deutschland GmbH
Páginas361-371
Número de páginas11
ISBN (versión impresa)9783030872304
DOI
EstadoPublished - 2021
Evento24th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2021 - Virtual, Online
Duración: sep. 27 2021oct. 1 2021

Serie de la publicación

NombreLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volumen12906 LNCS
ISSN (versión impresa)0302-9743
ISSN (versión digital)1611-3349

Conference

Conference24th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2021
CiudadVirtual, Online
Período9/27/2110/1/21

Nota bibliográfica

Funding Information:
Acknowledgement. This work is supported by the National Institutes of Health

Funding Information:
This work is supported by the National Institutes of Health (NIH) under Award Number R01HL145590 and R01NR018301.

Publisher Copyright:
© 2021, Springer Nature Switzerland AG.

ASJC Scopus Subject Areas

  • Theoretical Computer Science
  • General Computer Science

Huella

Profundice en los temas de investigación de 'Label-Free Physics-Informed Image Sequence Reconstruction with Disentangled Spatial-Temporal Modeling'. En conjunto forman una huella única.

Citar esto