Effect of iterative reconstruction and temporal averaging on contour sharpness in dynamic myocardial CT perfusion: Sub-analysis of the prospective 4D CT perfusion pilot study

Abstract

PurposeMyocardial computed tomography perfusion (CTP) allows the assessment of the functional relevance of coronary artery stenosis. This study investigates to what extent the contour sharpness of sequences acquired by dynamic myocardial CTP is influenced by the following noise reduction methods: temporal averaging and adaptive iterative dose reduction 3D (AIDR 3D).Materials and methodsDynamic myocardial CT perfusion was conducted in 29 patients at a dose level of 9.5±2.0 mSv and was reconstructed with both filtered back projection (FBP) and strong levels of AIDR 3D. Temporal averaging to reduce noise was performed as a post-processing step by combining two, three, four, six and eight original consecutive 3D datasets. We evaluated the contour sharpness at four distinct edges of the left-ventricular myocardium based on two different approaches: the distance between 25% and 75% of the maximal grey value (d) and the slope in the contour (m).ResultsIterative reconstruction reduced contour sharpness: both measures of contour sharpness performed better for FBP than for AIDR 3D (d = 1.7±0.4 mm versus 2.0±0.5 mm, p>0.059 at all edges; m = 255.9±123.9 HU/mm versus 160.6±123.5 HU/mm; p<0.023 for all edges). Increasing levels of temporal averaging degraded contour sharpness. When FBP reconstruction was applied, contour sharpness was best without temporal averaging (d = 1.7±0.4 mm, m = 255.9±123.9 HU/mm) and poorest for the strongest levels of temporal averaging (d = 2.1±0.3 mm, m = 142.2±104.9 HU/mm; comparison between lowest and highest temporal averaging level: for d p>0.052 at all edges and for m p<0.001 at all edges).ConclusionThe use of both temporal averaging and iterative reconstruction degrades objective contour sharpness parameters of dynamic myocardial CTP. Thus, further advances in image processing are needed to optimise contour sharpness of 4D myocardial CTP.

Deep learning image reconstruction algorithm for abdominal multidetector CT at different tube voltages: assessment of image quality and radiation dose in a phantom study

“…The degradation of image sharpness resulting in a blotchy appearance of the solid organs could compromise the diagnostic accuracy if it caused a well-defined lesion to appear less well-defined. Such degradation can make it difficult to differentiate a small hepatic cyst from metastasis in abdominal CT interpretation for radiologists and had been regarded as an important drawback of ASIR [32][33][34]. At the beginning of this study, we expected that DLIR has a great advantage in improving the noise and sharpness simultaneously, which was regarded to be the weakest point of ASIR-V in terms of image sharpness.…”

Noise reduction and motion elimination in low-dose 4D myocardial computed tomography perfusion (CTP): preliminary clinical evaluation of the ASTRA4D algorithm

“…Image quality peaked when using 3 combinations and motion classified as little motion [29]. Contour sharpness decreased monotonically with increasing number of averaging volumes reconstructed by filtered backprojection (FBP), while iterative reconstructions (AIDR3D) reduced contour sharpness compared to FBP [28]. Only the use of 3 combinations improved diagnostic accuracy in the detection of perfusion defects relative the original sequence, resulting in a per-patient sensitivity of 82% (9/11) and specificity 100% (4/4), and on a persegment sensitivity of 31% (10/32) and specificity of 97% (216/223), considering MR perfusion as the reference standard [29].…”

“…While objective image quality (noise, signal-tonoise and contrast-to-noise ratio) was improved, mainly due to the nature of the averaging filter, subjective image quality could be improved only when combining three cardiac phases with little cardiac motion. Without morphological alignment, moderate to severe motion led to blurring of edges and fine structures [28], spatially, and temporal averaging erased hypoperfused regions, which differ from healthy myocardial tissue only by 17-50 HU [17] and thus in the same range as noise (Figure 8).…”

“…The goal of the dissertation is to assess and to improve image quality and diagnostic accuracy of the dynamic CT perfusion sequence obtained from the CT perfusion pilot study (n=30) for the visual detection of myocardial perfusion deficits with MR myocardial perfusion imaging as the reference standard. In a first step, the notion of temporal averaging of consecutive 3D CTP datasets, without prior motion correction, as a method of noise reduction was introduced [28,29]. In a second step, we proposed a unified framework comprising both motion elimination and an ex-ante approach to temporal denoising [30] on the basis of exploratory principal component-based alignment [22].…”