Low-contrast detectability and potential for radiation dose reduction using deep learning image reconstruction-A 20-reader study on a semi-anthropomorphic liver phantom

Tormund Njølstad; Kristin Jensen; Anniken Dybwad; Øyvind Salvesen; Hilde K Andersen; Anselm Schulz

doi:10.1016/j.ejro.2022.100418

Low-contrast detectability and potential for radiation dose reduction using deep learning image reconstruction-A 20-reader study on a semi-anthropomorphic liver phantom

Eur J Radiol Open. 2022 Apr 2:9:100418. doi: 10.1016/j.ejro.2022.100418. eCollection 2022.

Authors

Tormund Njølstad^{1

2}, Kristin Jensen³, Anniken Dybwad³, Øyvind Salvesen⁴, Hilde K Andersen³, Anselm Schulz¹

Affiliations

¹ Department of Radiology and Nuclear Medicine, Oslo University Hospital Ullevål, Oslo, Norway.
² Department of Radiology, Haukeland University Hospital, Bergen, Norway.
³ Department of Diagnostic Physics, Oslo University Hospital, Oslo, Norway.
⁴ Department of Clinical and Molecular Medicine, Faculty of Medicine and Health Sciences, Norwegian University of Science and Technology (NTNU), Trondheim, Norway.

Abstract

Background: A novel deep learning image reconstruction (DLIR) algorithm for CT has recently been clinically approved.

Purpose: To assess low-contrast detectability and dose reduction potential for CT images reconstructed with the DLIR algorithm and compare with filtered back projection (FBP) and hybrid iterative reconstruction (IR).

Material and methods: A customized upper-abdomen phantom containing four cylindrical liver inserts with low-contrast lesions was scanned at CT dose indexes of 5, 10, 15, 20 and 25 mGy. Images were reconstructed with FBP, 50% hybrid IR (IR50), and DLIR of low strength (DLL), medium strength (DLM) and high strength (DLH). Detectability was assessed by 20 independent readers using a two-alternative forced choice approach. Dose reduction potential was estimated separately for each strength of DLIR using a fitted model, with the detectability performance of FBP and IR50 as reference.

Results: For the investigated dose levels of 5 and 10 mGy, DLM improved detectability compared to FBP by 5.8 and 6.9 percentage points (p.p.), and DLH improved detectability by 9.6 and 12.3 p.p., respectively (all p < .007). With IR50 as reference, DLH improved detectability by 5.2 and 9.8 p.p. for the 5 and 10 mGy dose level, respectively (p < .03). With respect to this low-contrast detectability task, average dose reduction potential relative to FBP was estimated to 39% for DLM and 55% for DLH. Relative to IR50, average dose reduction potential was estimated to 21% for DLM and 42% for DLH.

Conclusions: Low-contrast detectability performance is improved when applying a DLIR algorithm, with potential for radiation dose reduction.

Keywords: CT; DLH, deep learning image reconstruction of high strength; DLIR, deep learning image reconstruction; DLL, deep learning image reconstruction of low strength; DLM, deep learning image reconstruction of medium strength; Deep learning image reconstruction; FBP, filtered back projection; IR, iterative reconstruction; Low-contrast detectability.