Ultra-low-dose hepatic computed tomography with a novel real-time deep learning-based noise reduction algorithm: a prospective cross-sectional analysis of image quality and lesion detection

Abstract

Background: Contrast-enhanced computed tomography (CT) is essential for tumor assessment, but the detection of low-contrast liver lesions remains challenging. Reducing the radiation dose increases image noise, compromising image quality and diagnostic accuracy. Iterative reconstruction (IR) algorithms can reduce noise; however, they can also alter image texture and limit lesion detection. Deep-learning image reconstruction (DLIR) represents a promising alternative, but its efficacy in ultra-low-dose (ULD) hepatic CT for detecting small, low-contrast lesions remains underexplored. Thus, this study aimed to evaluate a novel real-time DLIR algorithm in ULD hepatic CT, focusing on image quality and lesion detection.

Methods: In total, 65 patients with hepatic lesions underwent both standard-dose and ULD abdominal CT scans during the portal venous phase. The standard-dose protocol (group A) used 120 kV with a signal-to-noise ratio (SNR) of 1.0, and the images were reconstructed using 50% IR. The ULD protocol (group B) used 120 kV with an SNR of 0.5, and the images were reconstructed using 50% IR and 50% DLIR (groups B1 and B2, respectively). The quantitative and qualitative image quality parameters were assessed. The lesion detection rates were evaluated by lesion type and size using the metrics of detection rate, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV).

Results: Group B showed a 73.3% reduction in the radiation dose compared to group A (1.5±0.8 vs. 6.9±2.0 mSv, P<0.001). Image noise differed significantly across the protocols: group B1 had the highest noise [10.05±2.94 Hounsfield units (HU)], followed by groups A (8.29±2.82 HU) and B2 (8.04±2.71 HU; all pairwise P<0.05 except group A vs. group B2: P=0.625). The CT values and contrast-to-noise ratios (CNRs) were comparable between groups B2 and A (all P>0.05), while group B2 had a 29.9-42.2% higher CNR than group B1 (all P<0.001). The qualitative assessments confirmed that the image quality and diagnostic acceptability of groups B2 (100%) and A (all P>0.05) were comparable, while the images of group B1 were diagnostically unacceptable (all scores <3). Overall, lesion detection was comparable in groups B2 (90.5%, 133/147) and A (98.0%, 144/147; P>0.05). However, group B2 had a significantly lower detection rate for small lesions (<0.5 cm: 77.8%, 42/54) compared to group A (P<0.05), but outperformed group B1 (57.4%, 31/54; P<0.05). Group B2 also had a significantly improved lesion detection rate and sensitivity for low-contrast lesions (87.2%, 95/109) compared to group B1 (75.2%, 82/109; P<0.05). The novel DLIR algorithm achieved a reconstruction speed of 60 images per second (ips), which was significantly faster than that of other DLIR approaches, while maintaining comparable performance to the IR algorithm.

Conclusions: The combination of tube current reduction with a novel real-time DLIR algorithm enabled ULD abdominal CT to achieve a 73.3% reduction in the radiation dose while maintaining image quality and diagnostic performance for detecting hepatic lesions larger than 0.5 cm.

Keywords: Deep learning; abdomen; computed tomography (CT); radiation dose.

Figure 1

Flowchart of the study design. Group A: standard-dose protocol with 50% CV; group B1: ultra-low-dose protocol with 50% CV; group B2: ultra-low-dose protocol with 50% CI. CI, ClearInfinity; CT, computed tomography; CV, ClearView.

Figure 2

Illustrative diagram of the ClearInfinity algorithm based on DLIR. (A) The network structure diagram of DLIR. (B) The schematic diagram of DLIR. CT, computed tomography; DLIR, deep-learning image reconstruction.

Figure 3

Box plots of image noise among the three image groups. The central line represents the median, the box boundaries indicate the first and third quartiles, and the whisker boundaries extend to 1.5 quartiles. Other points outside the whisker boundaries are plotted as outliers. Group A: standard-dose protocol with 50% CV; group B1: ultra-low-dose protocol with 50% CV; group B2: ultra-low-dose protocol with 50% CI. CI, ClearInfinity; CV, ClearView; HU, Hounsfield units; SD, standard deviation.

Figure 4

Comparison of lesion detection rates among the three groups. (A) Comparison of lesion detection rates by lesion type. (B) Comparison of lesion detection rates by lesion size. *, P<0.05; **, P<0.001. Group A: standard-dose protocol with 50% CV; group B1: ultra-low-dose protocol with 50% CV; group B2: ultra-low-dose protocol with 50% CI. CI, ClearInfinity; CV, ClearView.

Figure 5

A 42-year-old female with hepatic metastases from cervical cancer was evaluated using the following three protocols: (A) SDCT with 50% CV reconstruction; (B) ULD-CT with 50% CV reconstruction; (C) ULD-CT with 50% CI reconstruction. An arrow indicates the hepatic metastatic lesion in each image. The image noise levels for the three groups were 9.17, 12.66, and 9.38 HU, respectively. Compared to the SDCT image (A), the ULD-CT image with 50% CI reconstruction (C) showed a 76.1% reduction in the radiation dose, while preserving the lesion’s structure, details and clear boundaries to meet the clinical diagnostic requirements, while the ULD-CT with 50% CV (B) had higher image noise and less clear lesion boundaries. CI, ClearInfinity; CT, computed tomography; CV, ClearView; HU, Hounsfield units; SDCT, standard-dose CT; ULD-CT, ultra-low-dose CT.