Table 2.
Comparison of reconstruction time, image noise, signal-to-noise ratio, contrast-to-noise ratio, and visual evaluation of coronary arteries between super-resolution deep learning reconstruction and model-based iterative reconstruction
| Parameter | SR-DLR | MBIR | p-value |
|---|---|---|---|
| Reconstruction time, s | 97 (88–109) | 173 (164–187) | < 0.01 |
| Image noise, HU | |||
| Ascending aorta | 22.5 (20.5–31.8) | 29.1 (26.2–36.6) | < 0.01 |
| Left atrium | 22.4 ± 4.0 | 30.1 ± 5.4 | < 0.01 |
| Septal wall of the ventricle | 20.5 ± 3.6 | 23.7 ± 3.8 | < 0.01 |
| All locations | 22.1 (19.3–24.9) | 27.4 (24.2–31.2) | < 0.01 |
| SNR | 16.3 (11.8–21.8) | 13.7 (9.9–18.4) | 0.01 |
| CNR | 24.4 (15.5–30.2) | 19.2 (14.1–23.2) | < 0.01 |
| Overall quality (scale 1–4) | 4.0 (4.0–4.0) | 3.0 (3.0–4.0) | < 0.01 |
SR-DLR, super-resolution deep learning reconstruction; MBIR, model-based iterative reconstruction; SNR, signal-to-noise ratio; CNR, contrast-to-noise ratio