Deep learning-based reconstruction in ultra-high-resolution computed tomography: Can image noise caused by high definition detector and the miniaturization of matrix element size be improved?

Urikura, Atsushi; Yoshida, Tsukasa; Nakaya, Yoshihiro; Nishimaru, Eiji; Hara, Takanori; Endo, Masahiro

doi:10.1016/j.ejmp.2020.12.006

Cited by 21 publications

(11 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…NR was not included in this figure as AiCE is only available in HR mode. AiCE‐STD resulted in higher detectability compared with AIDR‐FC18 and FBP‐FC18 – a consequence of the superior resolution and noise suppression for AiCE‐STD as shown in previous studies 31,32 . For a 6 mm lumen diameter, the detectability was 48% and 18% higher for AiCE‐STD in comparison to FBP‐FC18 and AIDR‐FC18, respectively.…”

Section: Resultssupporting

confidence: 66%

Performance of high‐resolution CT for detection and discrimination tasks related to stenotic lesions – A phantom study using model observers

et al. 2023

View full text Add to dashboard Cite

Background Accurate detection and grading of atheromatous stenotic lesions within the cardiac, renal, and intracranial vasculature is imperative for early recognition of disease and guiding treatment strategies. Purpose In this work, a stenotic lesion phantom was used to compare high resolution and normal resolution modes on the same CT scanner in terms of detection and size discrimination performance. Materials and Methods The phantom is comprised of three acrylic cylinders (each 15.0 cm in diameter and 1.3 cm thick) with a matching array of holes in each module. The outer two modules contain holes that are slightly larger than the corresponding hole in the central module to simulate stenotic narrowing in vasculature. The stack of modules was submerged in an iodine solution simulating contrast‐enhanced stenotic lesions with a range of lumen diameters (1.32–10.08 mm) and stenosis severity (0%, 50%, 60%, 70%, and 80%). The phantom was imaged on the Canon Aquilion Precision high‐resolution CT scanner in high‐resolution (HR) mode (0.25 mm × 0.50 mm detector element size) and normal‐resolution (NR) mode (0.50 mm × 0.50 mm) using 120 kV and two dose levels (14 and 21 mGy SSDE) with 30 repeat scans acquired for each combination. Filtered back‐projection (FBP) and a hybrid‐iterative reconstruction (AIDR) were used with the FC18 kernel, as well as a deep learning algorithm (AiCE) which is only available for HR. A non‐prewhitening model observer with an eye filter was implemented to quantify performance for detection and size discrimination tasks in the axial plane. Results Detection performance improved with increasing diameter, dose, and for AIDR in comparison to FBP for a fixed resolution mode. Performance in the HR mode was generally higher than NR for the smaller lumen diameters (1–5 mm) with decreasing differences as the diameter increased. Performance in NR mode surpassed HR mode for lumen diameters greater than ∼4 mm and ∼5 mm for 14 mGy and 21 mGy, respectively. AiCE provided consistently higher detection performance compared with AIDR‐FC18 (48% higher for a 6 mm lumen diameter). Discrimination performance increased with increasing nominal diameter, dose, and for larger differences in stenosis severity. When comparing discrimination performance in HR to NR modes, the largest relative differences occur at the smallest nominal diameters and smallest differences in stenosis severity. The AiCE reconstruction algorithm produced the highest overall discrimination performance values, and these were significantly higher than AIDR‐FC18 for nominal diameters of 7.14 and 10.08 mm. Conclusions HR mode outperforms NR for detection up to a specific diameter and the results improve with AiCE and for higher dose levels. For the task of size discrimination, HR mode consistently outperforms NR if AIDR‐FC18 is used for dose levels of at least 21 mGy, and the results improve with AiCE and for the smallest differences in stenosis severity investigated (50% vs. 60%). High‐resolution CT appears to be beneficial for detecting s...

show abstract

Section: Resultssupporting

confidence: 66%

Performance of high‐resolution CT for detection and discrimination tasks related to stenotic lesions – A phantom study using model observers

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Advanced protocols using dual energy CT and/or deep learning-based reconstruction might solve these problems. 11,12 In conclusion, low-tube-voltage slow-injection CTPA is a promising technique for achieving high contrast enhancement while accommodating fragile veins and low injection rates. Although our proposed method does not require a saline flush, artifacts produced by the SVC are minimal.…”

Section: Discussionmentioning

confidence: 98%

“…Advanced protocols using dual energy CT and/or deep learning–based reconstruction might solve these problems. 11 , 12 …”

Section: Discussionmentioning

confidence: 99%

Computed tomographic pulmonary angiography: Three cases of low-tube-voltage acquisition with a slow injection of contrast medium

Urikura

Yoshida

Endo

et al. 2022

Acta Radiologica Open

Self Cite

View full text Add to dashboard Cite

Acute pulmonary thromboembolism occurring during cancer treatment has been increasing with the number of cancer patients and chemotherapy cases. Computed tomographic pulmonary angiography (CTPA) for evaluating the pulmonary artery is generally performed using rapid injection of contrast medium. However, intravenous catheters for contrast medium injection might cause extravasation due to rapid injection. This case series describes three patients who underwent contrast-enhanced computed tomography combined with low-tube-voltage imaging and slow injection. Low-tube-voltage slow-injection CTPA can be an effective technique for obtaining high contrast enhancement while accommodating fragile veins and low injection rates.

show abstract

“…15 Dedicated deep learning reconstruction algorithms have been developed for noise reduction that allow dose neutral high-resolution imaging and maintain image quality in patients with high body mass index. [18][19][20] An example of noise reduction when using a deep learning reconstruction algorithm specifically developed for cardiac UHRCT is provided in Figure…”

Section: J O U R N a L P R E -P R O O Fmentioning

confidence: 99%

CT imaging with ultra-high-resolution: Opportunities for cardiovascular imaging in clinical practice

Schuijf¹,

Lima²,

Boedeker³

et al. 2022

Journal of Cardiovascular Computed Tomography

View full text Add to dashboard Cite

Deep learning-based reconstruction in ultra-high-resolution computed tomography: Can image noise caused by high definition detector and the miniaturization of matrix element size be improved?

Cited by 21 publications

References 25 publications

Performance of high‐resolution CT for detection and discrimination tasks related to stenotic lesions – A phantom study using model observers

Performance of high‐resolution CT for detection and discrimination tasks related to stenotic lesions – A phantom study using model observers

Computed tomographic pulmonary angiography: Three cases of low-tube-voltage acquisition with a slow injection of contrast medium

CT imaging with ultra-high-resolution: Opportunities for cardiovascular imaging in clinical practice

Contact Info

Product

Resources

About