We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

eess.IV

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Electrical Engineering and Systems Science > Image and Video Processing

Title: TransEM:Residual Swin-Transformer based regularized PET image reconstruction

Abstract: Positron emission tomography(PET) image reconstruction is an ill-posed inverse problem and suffers from high level of noise due to limited counts received. Recently deep neural networks especially convolutional neural networks(CNN) have been successfully applied to PET image reconstruction. However, the local characteristics of the convolution operator potentially limit the image quality obtained by current CNN-based PET image reconstruction methods. In this paper, we propose a residual swin-transformer based regularizer(RSTR) to incorporate regularization into the iterative reconstruction framework. Specifically, a convolution layer is firstly adopted to extract shallow features, then the deep feature extraction is accomplished by the swin-transformer layer. At last, both deep and shallow features are fused with a residual operation and another convolution layer. Validations on the realistic 3D brain simulated low-count data show that our proposed method outperforms the state-of-the-art methods in both qualitative and quantitative measures.
Comments: Accepted to 25th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) 2022
Subjects: Image and Video Processing (eess.IV)
Journal reference: 2022 25th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI)
DOI: 10.1007/978-3-031-16440-8_18
Cite as: arXiv:2205.04204 [eess.IV]
  (or arXiv:2205.04204v2 [eess.IV] for this version)

Submission history

From: Rui Hu [view email]
[v1] Mon, 9 May 2022 11:46:03 GMT (1029kb,D)
[v2] Mon, 24 Oct 2022 13:28:15 GMT (1399kb,D)

Link back to: arXiv, form interface, contact.