Skip to Main content Skip to Navigation
Conference papers

Linear and Deformable Image Registration with 3D Convolutional Neural Networks

Abstract : Image registration and in particular deformable registration methods are pillars of medical imaging. Inspired by the recent advances in deep learning, we propose in this paper, a novel convolutional neu-ral network architecture that couples linear and deformable registration within a unified architecture endowed with near real-time performance. Our framework is modular with respect to the global transformation component, as well as with respect to the similarity function while it guarantees smooth displacement fields. We evaluate the performance of our network on the challenging problem of MRI lung registration, and demonstrate superior performance with respect to state of the art elastic registration methods. The proposed deformation (between inspiration & expiration) was considered within a clinically relevant task of interstitial lung disease (ILD) classification and showed promising results.
Complete list of metadatas

Cited literature [16 references]  Display  Hide  Download
Contributor : Mihir Sahasrabudhe <>
Submitted on : Monday, November 26, 2018 - 6:48:42 PM
Last modification on : Thursday, July 9, 2020 - 4:06:04 PM
Document(s) archivé(s) le : Wednesday, February 27, 2019 - 3:29:03 PM


Files produced by the author(s)


  • HAL Id : hal-01935607, version 1


Stergios Christodoulidis, Mihir Sahasrabudhe, Maria Vakalopoulou, Guillaume Chassagnon, Marie-Pierre Revel, et al.. Linear and Deformable Image Registration with 3D Convolutional Neural Networks. Reconstruction and Analysis of Moving Body Organs, 21th International Conference on Medical Image Computing and Computer Assisted Intervention 2018, Sep 2018, Grenada, Spain. ⟨hal-01935607⟩



Record views


Files downloads