Type:
Educational Exhibit
Keywords:
Cardiac, Computer applications, CT, CT-Angiography, CT-High Resolution, eLearning, Education and training
Authors:
M. Ohana, F. Tatsugami, A. Labani, W. Fukumoto, A. Taniguchi, K. Haioun, K. Awai, C. Roy
DOI:
10.26044/ecr2023/C-10591
Background
Ultra-High Resolution (UHR) coronary-CTA (CCTA) acquisitions are available since 2017 (Aquilion Precision), with :
- 0.25mm detector size
- 160 rows
- 0.4 x 0.5 mm focal spot
- 1024 and 2048 matrix sizes
- Deep Learning Reconstructions are feasible
Publications such as [1] have demonstrated significant advances in CCTA diagnostic performances with UHR-CT in patients with heavily calcified arteries (Figure 1) and/or stents.
However, UHR CT is – in 2023 – still largely under-deployed [2], due to
- High cost
- Inadequate PACS infrastructure
- Some technical disadvantages in the field of CCTA
- No wide-area detector - only 0.25mm x 160 = 4cm coverage
- No fast gantry rotation - only 0.35 sec
- Recommended usage of 120kV, resulting in higher radiation dose.
The goal of Super Resolution Deep Learning Reconstruction (SR-DLR) is to overcome these limitations by bringing the UHR-CT advantages (higher spatial resolution, lower image noise, etc...) to conventional wide-area detector CT, so as to keep some of its strategic benefits such as one-beat volume acquisition.
Deep Learning Reconstruction (DLR) was introduced in 2018-2019: it uses modern Artificial Intelligence techniques to denoise reconstructed CT images.
In 2023, DLR can be considered as the current reference standard in CT image reconstruction:
- Regardless of the type of CT examination
- Regardless of the anatomical location
- Except for extremely specific quantification tasks (emphysema)
Super Resolution DLR “SR-DLR” is based on the same principle of DLR, which involves 3D-Convolutionnal Neural Network [3], trained using a database - UHR CT and paired simulated conventional CT - in the factory (Figure 2) and applied onsite to enhance conventional CT images (Figure 3).
In a routine cardiac CT rotation, SR-DLR has many advantages
- Image reconstruction is done on-site - no need to upload to a cloud solution
- Technology is transparent for the end user
- Reconstruction speed is ≈ 2 minutes for a volume of 640 images
- Impact on workflow is minimal
- Usage is therefore possible for 100% of cardiac CT