Exploring the Low-Dose Limit for Focal Hepatic Lesion Detection with a Deep Learning-Based CT Reconstruction Algorithm: A Simulation Study on Patient Images

Abstract

This study aims to investigate the maximum achievable dose reduction for applying a new deep learning-based reconstruction algorithm, namely the artificial intelligence iterative reconstruction (AIIR), in computed tomography (CT) for hepatic lesion detection. A total of 40 patients with 98 clinically confirmed hepatic lesions were retrospectively included. The mean volume CT dose index was 13.66 ± 1.73 mGy in routine-dose portal venous CT examinations, where the images were originally obtained with hybrid iterative reconstruction (HIR). Low-dose simulations were performed in projection domain for 40%-, 20%-, and 10%-dose levels, followed by reconstruction using both HIR and AIIR. Two radiologists were asked to detect hepatic lesion on each set of low-dose image in separate sessions. Qualitative metrics including lesion conspicuity, diagnostic confidence, and overall image quality were evaluated using a 5-point scale. The contrast-to-noise ratio (CNR) for lesion was also calculated for quantitative assessment. The lesion CNR on AIIR at reduced doses were significantly higher than that on routine-dose HIR (all p < 0.05). Lower qualitative image quality was observed as the radiation dose reduced, while there were no significant differences between 40%-dose AIIR and routine-dose HIR images. The lesion detection rate was 100%, 98% (96/98), and 73.5% (72/98) on 40%-, 20%-, and 10%-dose AIIR, respectively, whereas it was 98% (96/98), 73.5% (72/98), and 40% (39/98) on the corresponding low-dose HIR, respectively. AIIR outperformed HIR in simulated low-dose CT examinations of the liver. The use of AIIR allows up to 60% dose reduction for lesion detection while maintaining comparable image quality to routine-dose HIR.

Keywords: Artificial intelligence iterative reconstruction; Computed tomography; Dose reduction; Hepatic lesion.

Fig. 1

The flowchart of the study design

Fig. 2

Comparison of hepatic CT images at 4 dose levels in an 83-year-old male with hepatic cysts and metastases. (a1–d1 and a2) images reconstructed with HIR. (b2–d2) images reconstructed with AIIR. Green circles indicate reference lesions, and yellow arrows indicate hepatic lesions detected by readers. All lesions can be detected on 40%-dose and 20%-dose AIIR, with the lesion conspicuity being scored as 5 and 4, respectively. However, a tiny hepatic cyst (3.4 mm) was missed on 20%-dose HIR, where the lesion conspicuity was scored as 2. On 10%-dose HIR and AIIR, 3 and 2 out of 4 lesions were failed to detect by readers, respectively

Fig. 3

Comparison of hepatic CT images at 4 dose levels in a 64-year-old male with an 8.1 mm hepatic hemangioma (green circle). (a1–d1 and a2) images reconstructed with HIR. (b2–d2) images reconstructed with AIIR. The hemangioma was detected on both AIIR and HIR at 40%- and 20%-dose level (yellow arrows), while it appears more conspicuous with AIIR than with HIR. Two readers failed to detect the hemangioma on both 10%-dose HIR and AIIR, where the lesion CNR were 0.22 and 0.51, respectively. The lesion conspicuity was scored as 1 for two images, due to the blurring of lesion boundary and loss of imaging detail

Fig. 4

Bar graphs show lesion CNR and detection rate in relation to radiation dose, reconstruction algorithm, and lesion size