First experience of evaluation of the impact of high-matrix size reconstruction in image quality in arterial CT runoff studies of the lower extremities

Objectives To determine whether image reconstruction with a higher matrix size improves image quality for lower extremity CTA studies. Methods Raw data from 50 consecutive lower extremity CTA studies acquired on two MDCT scanners (SOMATOM Flash, Force) in patients evaluated for peripheral arterial d...

Full description

Saved in:
Bibliographic Details
Published in:European radiology 2023-12, Vol.33 (12), p.8745-8753
Main Authors: Schwartz, Fides R., Ronald, James S., Kalisz, Kevin R., Fu, Wanyi, Ramirez-Giraldo, Juan Carlos, Koweek, Lynne M. Hurwitz, Churchill, Susan, Southerland, Kevin W., Marin, Daniele
Format: Article
Language:English
Subjects:
Angiography
Arteries
Confidence
Diagnostic Radiology
Evaluation
Extremities
Image processing
Image quality
Image reconstruction
Imaging
Internal Medicine
Interventional Radiology
Mathematical analysis
Medical imaging
Medicine
Medicine & Public Health
Neuroradiology
Radiology
Stenosis
Ultrasound
Vascular-Interventional
Veins & arteries
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Objectives To determine whether image reconstruction with a higher matrix size improves image quality for lower extremity CTA studies. Methods Raw data from 50 consecutive lower extremity CTA studies acquired on two MDCT scanners (SOMATOM Flash, Force) in patients evaluated for peripheral arterial disease (PAD) were retrospectively collected and reconstructed with standard (512 × 512) and higher resolution (768 × 768, 1024 × 1024) matrix sizes. Five blinded readers reviewed representative transverse images in randomized order (150 total). Readers graded image quality (0 (worst)–100 (best)) for vascular wall definition, image noise, and confidence in stenosis grading. Ten patients’ stenosis scores on CTA images were compared to invasive angiography. Scores were compared using mixed effects linear regression. Results Reconstructions with 1024 × 1024 matrix were ranked significantly better for wall definition (mean score 72, 95% CI = 61–84), noise (74, CI = 59–88), and confidence (70, CI = 59–80) compared to 512 × 512 (wall = 65, CI = 53 × 77; noise = 67, CI = 52 × 81; confidence = 62, CI = 52 × 73; p  = 0.003, p  = 0.01, and p  = 0.004, respectively). Compared to 512 × 512, the 768 × 768 and 1024 × 1024 matrix improved image quality in the tibial arteries (wall = 51 vs 57 and 59, p  
ISSN:1432-1084
0938-7994
1432-1084
DOI:10.1007/s00330-023-09841-4