Effectiveness of deep learning reconstruction on standard to ultra-low-dose high-definition chest CT images

Nayu Hamabuchi; Yoshiharu Ohno; Hirona Kimata; Yuya Ito; Kenji Fujii; Naruomi Akino; Daisuke Takenaka; Takeshi Yoshikawa; Yuka Oshima; Takahiro Matsuyama; Hiroyuki Nagata; Takahiro Ueda; Hirotaka Ikeda; Yoshiyuki Ozawa; Hiroshi Toyama

doi:10.1007/s11604-023-01470-7

Effectiveness of deep learning reconstruction on standard to ultra-low-dose high-definition chest CT images

Jpn J Radiol. 2023 Dec;41(12):1373-1388. doi: 10.1007/s11604-023-01470-7. Epub 2023 Jul 27.

Authors

Nayu Hamabuchi¹, Yoshiharu Ohno^{2

3}, Hirona Kimata⁴, Yuya Ito⁴, Kenji Fujii⁴, Naruomi Akino⁴, Daisuke Takenaka^{1

5}, Takeshi Yoshikawa^{1

5}, Yuka Oshima¹, Takahiro Matsuyama¹, Hiroyuki Nagata⁶, Takahiro Ueda¹, Hirotaka Ikeda¹, Yoshiyuki Ozawa¹, Hiroshi Toyama¹

Affiliations

¹ Department of Radiology, Fujita Health University School of Medicine, 1-98 Dengakugakubo, Kutsukake-cho, Toyoake, Aichi, 470-1192, Japan.
² Department of Diagnostic Radiology, Fujita Health University School of Medicine, Toyoake, Aichi, Japan. yohno@fujita-hu.ac.jp.
³ Joint Research Laboratory of Advanced Medical Imaging, Fujita Health University School of Medicine, Toyoake, Aichi, Japan. yohno@fujita-hu.ac.jp.
⁴ Canon Medical Systems Corporation, Otawara, Tochigi, Japan.
⁵ Department of Diagnostic Radiology, Hyogo Cancer Center, Akashi, Hyogo, Japan.
⁶ Joint Research Laboratory of Advanced Medical Imaging, Fujita Health University School of Medicine, Toyoake, Aichi, Japan.

Abstract

Purpose: Deep learning reconstruction (DLR) has been introduced by major vendors, tested for CT examinations of a variety of organs, and compared with other reconstruction methods. The purpose of this study was to compare the capabilities of DLR for image quality improvement and lung texture evaluation with those of hybrid-type iterative reconstruction (IR) for standard-, reduced- and ultra-low-dose CTs (SDCT, RDCT and ULDCT) obtained with high-definition CT (HDCT) and reconstructed at 0.25-mm, 0.5-mm and 1-mm section thicknesses with 512 × 512 or 1024 × 1024 matrixes for patients with various pulmonary diseases.

Materials and methods: Forty age-, gender- and body mass index-matched patients with various pulmonary diseases underwent SDCT (CT dose index volume <CTDI_vol>: mean ± standard deviation, 9.0 ± 1.8 mGy), RDCT (CTDI_vol: 1.7 ± 0.2 mGy) and ULDCT (CTDI_vol: 0.8 ± 0.1 mGy) at a HDCT. All CT data set were then reconstructed with 512 × 512 or 1024 × 1024 matrixes by means of hybrid-type IR and DLR. SNR of lung parenchyma and probabilities of all lung textures were assessed for each CT data set. SNR and detection performance of each lung texture reconstructed with DLR and hybrid-type IR were then compared by means of paired t tests and ROC analyses for all CT data at each section thickness.

Results: Data for each radiation dose showed DLR attained significantly higher SNR than hybrid-type IR for each of the CT data (p < 0.0001). On assessments of all findings except consolidation and nodules or masses, areas under the curve (AUCs) for ULDCT with hybrid-type IR for each section thickness (0.91 ≤ AUC ≤ 0.97) were significantly smaller than those with DLR (0.97 ≤ AUC ≤ 1, p < 0.05) and the standard protocol (0.98 ≤ AUC ≤ 1, p < 0.05).

Conclusion: DLR is potentially more effective for image quality improvement and lung texture evaluation than hybrid-type IR on all radiation dose CTs obtained at HDCT and reconstructed with each section thickness with both matrixes for patients with a variety of pulmonary diseases.

Keywords: CT; Deep learning reconstruction; Iterative reconstruction; Lung; Radiation dose.

MeSH terms

Algorithms
Deep Learning*
Humans
Lung / diagnostic imaging
Lung Diseases* / diagnostic imaging
Radiation Dosage
Radiographic Image Interpretation, Computer-Assisted / methods
Tomography, X-Ray Computed / methods