Four-dimensional noise reduction using the time series of medical computed tomography datasets with short interval times: a static-phantom study.

Tatsuya Nishii,Noriyuki Negi,Wakiko Tani,Kazuro Sugimura,Atsushi K Kono,Erina Suehiro,Satoru Takahashi

doi:10.7717/peerj.1680

Abstract

Backgrounds. This study examines the hypothesis that four-dimensional noise reduction (4DNR) with short interval times reduces noise in cardiac computed tomography (CCT) using “padding” phases. Furthermore, the capability of reducing the reduction dose in CCT using this post-processing technique was assessed.Methods. Using base and quarter radiation doses for CCT (456 and 114 mAs/rot with 120 kVp), a static phantom was scanned ten times with retrospective electrocardiogram gating, and 4DNR with short interval times (50 ms) was performed using a post-processing technique. Differences in the computed tomography (CT) attenuation, contrast-to-noise ratio (CNR) and spatial resolution with modulation transfer function in each dose image obtained with and without 4DNR were assessed by conducting a Tukey–Kramer’s test and non-inferiority test.Results. For the base dose, by using 4DNR, the CNR was improved from 1.18 ± 0.15 to 2.08 ± 0.20 (P = 0.001), while the CT attenuation and spatial resolution of the image of 4DNR did not were significantly inferior to those of reference image (P < 0.001). CNRs of the quarter-dose image in 4DNR also improved to 1.28 ± 0.11, and were not inferior to those of the non-4DNR images of the base dose (P < 0.001).Conclusions. 4DNR with short interval times significantly reduced noise. Furthermore, applying this method to CCT would have the potential of reducing the radiation dose by 75%, while maintaining a similar image noise level.

Highlights

Four-dimensional noise reduction (4DNR) is a powerful post-processing method of reducing noise by employing spatial–temporal analysis
The interval times were decided by the result of the ex vivo preliminary study, that the contrast-to-noise ratio (CNR) reached a plateau at an interval time of 50 ms
The legato images obtained with the same radiation dose were compared with the reference image in terms of the computed tomography (CT) attenuation, CNR, modulation transfer function (MTF), and noise power spectrum (NPS)

Summary

Introduction

Four-dimensional noise reduction (4DNR) is a powerful post-processing method of reducing noise by employing spatial–temporal analysis. We developed a method of applying 4DNR to datasets having short interval times (≤50 ms), which we refer to as the method of obtaining a noiseless image by adaptive phase-shifted topological coherence analysis or legato. The present study conducts quantitative image quality analysis using static ex vivo phantoms and in vivo retrospective analysis to examine the hypothesis that post processing with legato reduces the noise in CCT images and permits a lower radiation dose when using ‘‘padding’’ phases.

Results

Conclusion