Image quality and dose reduction opportunity of deep learning image reconstruction algorithm for CT: a phantom study

Joël Greffier; Aymeric Hamard; Fabricio Pereira; Corinne Barrau; Hugo Pasquier; Jean Paul Beregi; Julien Frandon

doi:10.1007/s00330-020-06724-w

Image quality and dose reduction opportunity of deep learning image reconstruction algorithm for CT: a phantom study

Eur Radiol. 2020 Jul;30(7):3951-3959. doi: 10.1007/s00330-020-06724-w. Epub 2020 Feb 25.

Authors

Joël Greffier^{1

2}, Aymeric Hamard³, Fabricio Pereira³, Corinne Barrau⁴, Hugo Pasquier⁵, Jean Paul Beregi³, Julien Frandon³

Affiliations

¹ Department of Medical Imaging, CHU Nimes, Medical Imaging Group Nimes, Univ Montpellier, EA 2415, Bd Prof Robert Debré, 30029, Nîmes Cedex 9, France. joel.greffier@chu-nimes.fr.
² Department of Medical Physics, CHU Nimes, Univ Montpellier, Montpellier, France. joel.greffier@chu-nimes.fr.
³ Department of Medical Imaging, CHU Nimes, Medical Imaging Group Nimes, Univ Montpellier, EA 2415, Bd Prof Robert Debré, 30029, Nîmes Cedex 9, France.
⁴ Department of Medical Physics, CHU Nimes, Univ Montpellier, Montpellier, France.
⁵ GE Healthcare, Buc, France.

PMID: 32100091
DOI: 10.1007/s00330-020-06724-w

Abstract

Objectives: To assess the impact on image quality and dose reduction of a new deep learning image reconstruction (DLIR) algorithm compared with a hybrid iterative reconstruction (IR) algorithm.

Methods: Data acquisitions were performed at seven dose levels (CTDI_vol : 15/10/7.5/5/2.5/1/0.5 mGy) using a standard phantom designed for image quality assessment. Raw data were reconstructed using the filtered back projection (FBP), two levels of IR (ASiR-V50% (AV50); ASiR-V100% (AV100)), and three levels of DLIR (TrueFidelity™ low, medium, high). Noise power spectrum (NPS) and task-based transfer function (TTF) were computed. Detectability index (d') was computed to model a large mass in the liver, a small calcification, and a small subtle lesion with low contrast.

Results: NPS peaks were higher with AV50 than with all DLIR levels and only higher with DLIR-H than with AV100. The average NPS spatial frequencies were higher with DLIR than with IR. For all DLIR levels, TTF_50% obtained with DLIR was higher than that with IR. d' was higher with DLIR than with AV50 but lower with DLIR-L and DLIR-M than with AV100. d' values were higher with DLIR-H than with AV100 for the small low-contrast lesion (10 ± 4%) and in the same range for the other simulated lesions.

Conclusions: New DLIR algorithm reduced noise and improved spatial resolution and detectability without changing the noise texture. Images obtained with DLIR seem to indicate a greater potential for dose optimization than those with hybrid IR.

Key points: • This study assessed the impact on image quality and radiation dose of a new deep learning image reconstruction (DLIR) algorithm as compared with hybrid iterative reconstruction (IR) algorithm. • The new DLIR algorithm reduced noise and improved spatial resolution and detectability without perceived alteration of the texture, commonly reported with IR. • As compared with IR, DLIR seems to open further possibility of dose optimization.

Keywords: Deep learning; Image enhancement; Image reconstruction; Multidetector computed tomography.

MeSH terms

Algorithms*
Deep Learning*
Humans
Phantoms, Imaging*
Radiation Dosage
Radiographic Image Interpretation, Computer-Assisted / methods*
Tomography, X-Ray Computed / methods*