Lung visualisation on PET/MRI: implementing a protocol with a short echo-time and low flip-angle volumetric interpolated breath-hold examination sequence.
Journal
Clinical radiology
ISSN: 1365-229X
Titre abrégé: Clin Radiol
Pays: England
ID NLM: 1306016
Informations de publication
Date de publication:
03 2020
03 2020
Historique:
received:
10
06
2019
accepted:
30
10
2019
pubmed:
6
12
2019
medline:
29
9
2020
entrez:
6
12
2019
Statut:
ppublish
Résumé
To assess the diagnostic performance in detecting lung lesions of a short echo-time (TE) and low flip-angle (FA) volumetric interpolated breath-hold examination (VIBE) sequence included in the integrated positron-emission tomography (PET)/magnetic resonance imaging (MRI) protocol. Thirty-seven oncological patients who underwent computed tomography (CT) and PET/MRI, including both a dedicated short TE, low FA VIBE (modified VIBE) and a standard VIBE of the lung, were enrolled. Modified VIBE images were reviewed retrospectively and independently by three raters, to detect pulmonary nodules, parenchymal consolidation, and bands. Three other groups examined standard VIBE, PET, and CT images. MRI and PET findings were compared to CT using Krippendorff's alpha using patient-based and a lesion-based analysis. Krippendorff's alpha was calculated to assess the interobserver agreement among the three raters of the modified VIBE. In the patient-based analysis (positivity ≥1 lesion), the comparison of modified VIBE with CT showed an alpha of 0.54 for nodules <6 mm (versus 0.41 for standard VIBE and 0.09 for PET) and an alpha of 0.88 for nodules ≥6 mm (versus 0.74 for standard VIBE and 0.42 for PET). On a lesion-based analysis (presence/absence of each lesion), modified VIBE compared to CT showed an alpha of0.58 for nodules <6 mm (versus 0.44 for standard VIBE and 0.09 for PET) and an alpha of 0.90 for nodules ≥6 mm (versus 0.79 for standard VIBE and 0.50 for PET). The alpha value for the interobserver agreement was 0.90 for nodules <6 mm, 0.91 for nodules ≥6 mm, 1.00 for consolidations, and 0.95 for bands in the patient-based analysis and 0.89, 0.93, 1.00, and 0.95 in the lesion-based analysis. Modified VIBE proved to be reproducible, showed better accuracy than standard VIBE and PET, and very good concordance with CT in assessing lung nodules ≥6 mm, whereas the agreement was less satisfactory for smaller nodules.
Identifiants
pubmed: 31801658
pii: S0009-9260(19)30642-7
doi: 10.1016/j.crad.2019.10.024
pii:
doi:
Substances chimiques
Radiopharmaceuticals
0
Fluorodeoxyglucose F18
0Z5B2CJX4D
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
239.e15-239.e21Informations de copyright
Copyright © 2019 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.