Masaki Katsura1, Jiro Sato2, Masaaki Akahane3, Yoko Mise2, Kaoru Sumida2, Osamu Abe2. 1. Department of Radiology, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan. Electronic address: mkatsura-tky@umin.ac.jp. 2. Department of Radiology, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan. 3. Department of Radiology, International University of Health and Welfare Mita Hospital, Tokyo, Japan.
Abstract
OBJECTIVES: To compare image quality characteristics of high-resolution computed tomography (HRCT) in the evaluation of interstitial lung disease using three different reconstruction methods: model-based iterative reconstruction (MBIR), adaptive statistical iterative reconstruction (ASIR), and filtered back projection (FBP). METHODS:Eighty-nine consecutive patients with interstitial lung disease underwentstandard-of-care chest CT with 64-row multi-detector CT. HRCT images were reconstructed in 0.625-mm contiguous axial slices using FBP, ASIR, and MBIR. Two radiologists independently assessed the images in a blinded manner for subjective image noise, streak artifacts, and visualization of normal and pathologic structures. Objective image noise was measured in the lung parenchyma. Spatial resolution was assessed by measuring the modulation transfer function (MTF). RESULTS: MBIR offered significantly lower objective image noise (22.24±4.53, P<0.01 among all pairs, Student's t-test) compared with ASIR (39.76±7.41) and FBP (51.91±9.71). MTF (spatial resolution) was increased using MBIR compared with ASIR and FBP. MBIR showed improvements in visualization of normal and pathologic structures over ASIR and FBP, while ASIR was rated quite similarly to FBP. MBIR significantly improved subjective image noise (P<0.01 among all pairs, the sign test), and streak artifacts (P<0.01 each for MBIR vs. the other 2 image data sets). CONCLUSION: MBIR provides high-quality HRCT images for interstitial lung disease by reducing image noise and streak artifacts and improving spatial resolution compared with ASIR and FBP.
RCT Entities:
OBJECTIVES: To compare image quality characteristics of high-resolution computed tomography (HRCT) in the evaluation of interstitial lung disease using three different reconstruction methods: model-based iterative reconstruction (MBIR), adaptive statistical iterative reconstruction (ASIR), and filtered back projection (FBP). METHODS: Eighty-nine consecutive patients with interstitial lung disease underwent standard-of-care chest CT with 64-row multi-detector CT. HRCT images were reconstructed in 0.625-mm contiguous axial slices using FBP, ASIR, and MBIR. Two radiologists independently assessed the images in a blinded manner for subjective image noise, streak artifacts, and visualization of normal and pathologic structures. Objective image noise was measured in the lung parenchyma. Spatial resolution was assessed by measuring the modulation transfer function (MTF). RESULTS: MBIR offered significantly lower objective image noise (22.24±4.53, P<0.01 among all pairs, Student's t-test) compared with ASIR (39.76±7.41) and FBP (51.91±9.71). MTF (spatial resolution) was increased using MBIR compared with ASIR and FBP. MBIR showed improvements in visualization of normal and pathologic structures over ASIR and FBP, while ASIR was rated quite similarly to FBP. MBIR significantly improved subjective image noise (P<0.01 among all pairs, the sign test), and streak artifacts (P<0.01 each for MBIR vs. the other 2 image data sets). CONCLUSION: MBIR provides high-quality HRCT images for interstitial lung disease by reducing image noise and streak artifacts and improving spatial resolution compared with ASIR and FBP.
Authors: Dinesh Khanna; Oliver Distler; Vincent Cottin; Kevin K Brown; Lorinda Chung; Jonathan G Goldin; Eric L Matteson; Ella A Kazerooni; Simon Lf Walsh; Michael McNitt-Gray; Toby M Maher Journal: J Scleroderma Relat Disord Date: 2022-01-03