Image quality improvement and radiation dose reduction in ultrahigh-resolution CT with iterative reconstruction

2023 Fiscal Year Final Research Report

• Project/Area Number: 18K07643
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 52040:Radiological sciences-related
• Research Institution: Tokyo Women's Medical University (2021-2023); Kyorin University (2018-2020)
• Principal Investigator: Machida Haruhiko 東京女子医科大学, 医学部, 教授 (70439834)
• Co-Investigator(Kenkyū-buntansha): 横山 健一 杏林大学, 医学部, 教授 (20383680)
• Project Period (FY): 2018-04-01 – 2024-03-31
• Keywords: 超高精細CT / 逐次近似再構成 / 深層学習再構成 / 高空間分解能 / 被ばく低減 / 造影剤減量

Outline of Final Research Achievements

Increased spatial resolution by ultrahigh-resolution CT (UHRCT) improved the delineation of fine structures in the middle and inner ears at CT of the temporal bones and peripheral bronchi at CT virtual bronchoscopy. UHRCT with iterative reconstruction improved the delineation of small vessels, such as perforating arteries of the brain, at CT angiography. UHRCT with deep learning reconstruction, compared with iterative reconstruction, improved subjective acceptance by radiologists to accurately detect fine recurrent, disseminated, and/or metastatic lesions while reducing the doses of ionizing radiation and contrast medium and preserving image quality at contrast-enhanced body CT for oncologic follow-up.

Free Research Field

放射線科学

Academic Significance and Societal Importance of the Research Achievements

本研究により、超高精細CTは中耳・内耳、肺・気管支などの画像コントラストが高い領域だけでなく、逐次近似再構成法や深層学習再構成法を適宜併用することでコントラストが低い領域でも十分な画質を維持しつつ空間分解能を向上し、微細な構造・病変の描出を改善できることが判明した。よって、超高精細CTは従来CTより様々な領域で診断能を改善し、侵襲的な検査を省略可能である。逐次近似再構成法や深層学習再構成法を併用することで、被ばくや造影剤量も低減しうるため、臨床的に大変有用と考えられる。