Improved image quality in abdominal computed tomography reconstructed with a novel Deep Learning Image Reconstruction technique - initial clinical experience.

Tormund Njølstad,Helga M Brøgger,Johannes C Godt,Anselm Schulz,Cathrine K Johansen,Anne Catrine T Martinsen,Hilde K Andersen

doi:10.1177/20584601211008391

Tormund Njølstad, Helga M Brøgger + Show 5 more

Open Access

https://doi.org/10.1177/20584601211008391

Copy DOI

Abstract

BackgroundA novel Deep Learning Image Reconstruction (DLIR) technique for computed tomography has recently received clinical approval.PurposeTo assess image quality in abdominal computed tomography reconstructed with DLIR, and compare with standardly applied iterative reconstruction.Material and methodsTen abdominal computed tomography scans were reconstructed with iterative reconstruction and DLIR of medium and high strength, with 0.625 mm and 2.5 mm slice thickness. Image quality was assessed using eight visual grading criteria in a side-by-side comparative setting. All series were presented twice to evaluate intraobserver agreement. Reader scores were compared using univariate logistic regression. Image noise and contrast-to-noise ratio were calculated for quantitative analyses.ResultsFor 2.5 mm slice thickness, DLIR images were more frequently perceived as equal or better than iterative reconstruction across all visual grading criteria (for both DLIR of medium and high strength, p < 0.001). Correspondingly, DLIR images were more frequently perceived as better (as opposed to equal or in favor of iterative reconstruction) for visual reproduction of liver parenchyma, intrahepatic vascular structures as well as overall impression of image noise and texture (p < 0.001). This improved image quality was also observed for 0.625 mm slice images reconstructed with DLIR of high strength when directly comparing to traditional iterative reconstruction in 2.5 mm slices. Image noise was significantly lower and contrast-to-noise ratio measurements significantly higher for images reconstructed with DLIR compared to iterative reconstruction (p < 0.01).ConclusionsAbdominal computed tomography images reconstructed using a DLIR technique shows improved image quality when compared to standardly applied iterative reconstruction across a variety of clinical image quality criteria.

Highlights

Computed tomography (CT) has become an indispensable clinical tool in contemporary medicine, with widespread availability, demonstrated safety, and non-invasive ability to rapidly image large anatomical volumes.[1]
Deep Learning Image Reconstruction (DLIR) image quality was more frequently perceived as slightly better or clearly better across all visual grading criteria compared to iterative reconstruction (IR) for all six image comparison sets, except the set comparing 0.625 mm DLIR of medium strength with 2.5 mm IR (Fig. 1)
When dichotomizing visual scores as DLIR perceived as equal or better than IR as opposed to worse, DLIR was significantly more frequently perceived as equal or better than IR for all visual grading criteria across all tested image sets, except for a selection of criteria in the set comparing 0.625 mm DLIR of medium strength to 2.5 mm IR (Table 2)

Summary

Introduction

Computed tomography (CT) has become an indispensable clinical tool in contemporary medicine, with widespread availability, demonstrated safety, and non-invasive ability to rapidly image large anatomical volumes.[1]. Results: For 2.5 mm slice thickness, DLIR images were more frequently perceived as equal or better than iterative reconstruction across all visual grading criteria (for both DLIR of medium and high strength, p < 0.001). DLIR images were more frequently perceived as better (as opposed to equal or in favor of iterative reconstruction) for visual reproduction of liver parenchyma, intrahepatic vascular structures as well as overall impression of image noise and texture (p < 0.001). This improved image quality was observed for 0.625 mm slice images reconstructed with DLIR of high strength when directly comparing to traditional iterative reconstruction in 2.5 mm slices. Conclusions: Abdominal computed tomography images reconstructed using a DLIR technique shows improved image quality when compared to standardly applied iterative reconstruction across a variety of clinical image quality criteria

Methods

Results

Conclusion