Hybrid U-Net and Swin-transformer network for limited-angle cardiac computed tomography.
Yongshun XuShuo HanDayang WangG E WangJonathan S MaltzHengyong YuPublished in: Physics in medicine and biology (2024)
Cardiac computed tomography is widely used for diagnosis of cardiovascular disease, the leading cause of morbidity and mortality in the world. Diagnostic performance depends strongly on the temporal resolution of the CT images. To image the beating heart, one can reduce the scanning time by acquiring limited-angle projections. However, this leads to increased image noise and limited-angle-related artifacts. The ability to reconstruct high quality images from limited-angle projections is highly desirable and remains a major challenge. With the development of deep learning networks, such as U-Net and transformer networks, progresses have been reached on image reconstruction and processing. Here we propose a hybrid model based on the U-Net and Swin-transformer (U-Swin) networks. The U-Net has the potential to restore structural information due to missing projection data and related artifacts, then the Swin-transformer can gather a detailed global feature distribution. Using synthetic XCAT and clinical cardiac COCA datasets, we demonstrate that our proposed method outperforms the state-of-the-art deep learning-based methods.
Keyphrases
- deep learning
- computed tomography
- image quality
- high resolution
- convolutional neural network
- artificial intelligence
- cardiovascular disease
- dual energy
- positron emission tomography
- left ventricular
- machine learning
- magnetic resonance imaging
- contrast enhanced
- big data
- heart failure
- electronic health record
- atrial fibrillation
- magnetic resonance
- rna seq
- cardiovascular events
- electron microscopy
- single cell