In atrial fibrillation treatment, it is important to visualize and analyze an accurate left atrial (LA) model from cardiac computed tomography (CT) images. In recent years, 3D-CNNs have been applied to acquire an accurate LA model from CT images. However, due to the hardware limitations, only LA models with low-resolution can be obtained. In this paper, we present a 3D super-resolution method that utilizes the high-resolution original CT volume as a guide by combining features with the same receptive field in layer level. Experimental results show that the proposed method achieves high performance in terms of quantitative and qualitative evaluations.
|Title of host publication||2021 IEEE International Conference on Consumer Electronics, ICCE 2021|
|Publisher||Institute of Electrical and Electronics Engineers Inc.|
|Publication status||Published - 2021 Jan 10|
|Event||2021 IEEE International Conference on Consumer Electronics, ICCE 2021 - Las Vegas, United States|
Duration: 2021 Jan 10 → 2021 Jan 12
|Name||Digest of Technical Papers - IEEE International Conference on Consumer Electronics|
|Conference||2021 IEEE International Conference on Consumer Electronics, ICCE 2021|
|Period||21/1/10 → 21/1/12|
Bibliographical noteFunding Information:
ACKNOWLEDGMENT This work was supported by the Institute for Information & communications Technology Promotion(IITP) grant funded by the Korea government(MSIT) (2017-0-00250, Intelligent Defense Boundary Surveillance Technology Using Collaborative Reinforced Learning of Embedded Edge Camera and Image Analysis).
© 2021 IEEE.
- atrial fibrillation
- computed tomography
- deep learning
- left atrial model
ASJC Scopus subject areas
- Industrial and Manufacturing Engineering
- Electrical and Electronic Engineering