The influence of a deep learning image reconstruction algorithm on the image quality and auto-analysis of pulmonary nodules at ultra-low dose chest CT: a phantom study

Yao, Yue; Guo, Baolei; Li, Jianying; Yang, Quanxin; Li, Xiaohui; Deng, Lei

doi:10.21037/qims-21-815

Cited by 10 publications

(3 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Detectability of simulated lung lesions was best with the smoothest level in DLR; a dose reduction potential of 81% to 94% was assumed. An overview of recently published articles on deep learning–based image reconstruction 5,9,47–87 is given in Table 5.…”

Section: Image Reconstructionmentioning

confidence: 99%

Computed Tomography 2.0

Lell

Kachelrieß

2023

Invest Radiol

View full text Add to dashboard Cite

Computed tomography (CT) dramatically improved the capabilities of diagnostic and interventional radiology. Starting in the early 1970s, this imaging modality is still evolving, although tremendous improvements in scan speed, volume coverage, spatial and soft tissue resolution, as well as dose reduction have been achieved. Tube current modulation, automated exposure control, anatomy-based tube voltage (kV) selection, advanced x-ray beam filtration, and iterative image reconstruction techniques improved image quality and decreased radiation exposure. Cardiac imaging triggered the demand for high temporal resolution, volume acquisition, and high pitch modes with electrocardiogram synchronization. Plaque imaging in cardiac CT as well as lung and bone imaging demand for high spatial resolution. Today, we see a transition of photon-counting detectors from experimental and research prototype setups into commercially available systems integrated in patient care. Moreover, with respect to CT technology and CT image formation, artificial intelligence is increasingly used in patient positioning, protocol adjustment, and image reconstruction, but also in image preprocessing or postprocessing. The aim of this article is to give an overview of the technical specifications of up-to-date available whole-body and dedicated CT systems, as well as hardware and software innovations for CT systems in the near future.

show abstract

Section: Image Reconstructionmentioning

confidence: 99%

Computed Tomography 2.0

Lell

Kachelrieß

2023

Invest Radiol

View full text Add to dashboard Cite

show abstract

“…To assess the CT performance in the imaging of the lung, studies utilize a range of phantoms, from simplified image quality phantoms to more advanced anthropomorphic phantoms [16,17]. In the specific context of evaluating lung nodules, several recent studies have effectively employed an anthropomorphic thorax phantom (Lungman, Kyoto Kagaku, Kyoto, Japan) to evaluate the performance of both EID-CT and PCD-CT systems [18,19]. Recent studies have also shown the potential of PCD-CT to reduce radiation exposure in low-dose lung examinations compared to EID-CT without compromising image quality [20,21].…”

Section: Introductionmentioning

confidence: 99%

Detectability and Volumetric Accuracy of Pulmonary Nodules in Low-Dose Photon-Counting Detector Computed Tomography: An Anthropomorphic Phantom Study

Hop,

Walstra,

Pelgrim

et al. 2023

Diagnostics

View full text Add to dashboard Cite

The aim of this phantom study was to assess the detectability and volumetric accuracy of pulmonary nodules on photon-counting detector CT (PCD-CT) at different low-dose levels compared to conventional energy-integrating detector CT (EID-CT). In-house fabricated artificial nodules of different shapes (spherical, lobulated, spiculated), sizes (2.5–10 mm and 5–1222 mm3), and densities (−330 HU and 100 HU) were randomly inserted into an anthropomorphic thorax phantom. The phantom was scanned with a low-dose chest protocol with PCD-CT and EID-CT, in which the dose with PCD-CT was lowered from 100% to 10% with respect to the EID-CT reference dose. Two blinded observers independently assessed the CT examinations of the nodules. A third observer measured the nodule volumes using commercial software. The influence of the scanner type, dose, observer, physical nodule volume, shape, and density on the detectability and volumetric accuracy was assessed by a multivariable regression analysis. In 120 CT examinations, 642 nodules were present. Observer 1 and 2 detected 367 (57%) and 289 nodules (45%), respectively. With PCD-CT and EID-CT, the nodule detectability was similar. The physical nodule volumes were underestimated by 20% (range 8–52%) with PCD-CT and 24% (range 9–52%) with EID-CT. With PCD-CT, no significant decrease in the detectability and volumetric accuracy was found at dose reductions down to 10% of the reference dose (p > 0.05). The detectability and volumetric accuracy were significantly influenced by the observer, nodule volume, and a spiculated nodule shape (p < 0.05), but not by dose, CT scanner type, and nodule density (p > 0.05). Low-dose PCD-CT demonstrates potential to detect and assess the volumes of pulmonary nodules, even with a radiation dose reduction of up to 90%.

show abstract

“…The typical applications of AI in pulmonary imaging are cancer detection, characterization, classification [ 10 , 11 ], and lung cancer screening [ 12 , 13 , 14 ]. As one of the emerging technologies of AI, deep learning image reconstruction (DLIR) has accelerated the development of ultra-low-dose CT (ULDCT) by improving the signal-to-noise ratio, contrast-to-noise ratio, and lung nodule detection rate [ 15 ]. At present, an iterative reconstruction (IR) algorithm is commonly used for image reconstruction, which refers to starting from an image assumption and comparing it with the real-time measured value while constantly adjusting until the two agree [ 16 ].…”

Section: Introductionmentioning

confidence: 99%

Measurement Accuracy and Repeatability of RECIST-Defined Pulmonary Lesions and Lymph Nodes in Ultra-Low-Dose CT Based on Deep Learning Image Reconstruction

Zhao

Jiang

Zhang

et al. 2022

Cancers

View full text Add to dashboard Cite

Background: Deep learning image reconstruction (DLIR) improves image quality. We aimed to compare the measured diameter of pulmonary lesions and lymph nodes between DLIR-based ultra-low-dose CT (ULDCT) and contrast-enhanced CT. Methods: The consecutive adult patients with noncontrast chest ULDCT (0.07–0.14 mSv) and contrast-enhanced CT (2.38 mSv) were prospectively enrolled. Patients with poor image quality and body mass index ≥ 30 kg/m2 were excluded. The diameter of pulmonary target lesions and lymph nodes defined by Response Evaluation Criteria in Solid Tumors (RECIST) was measured. The measurement variability between ULDCT and enhanced CT was evaluated by Bland-Altman analysis. Results: The 141 enrolled patients (62 ± 12 years) had 89 RECIST-defined measurable pulmonary target lesions (including 30 malignant lesions, mainly adenocarcinomas) and 45 measurable mediastinal lymph nodes (12 malignant). The measurement variation of pulmonary lesions between high-strength DLIR (DLIR-H) images of ULDCT and contrast-enhanced CT was 2.2% (95% CI: 1.7% to 2.6%) and the variation of lymph nodes was 1.4% (1.0% to 1.9%). Conclusions: The measured diameters of pulmonary lesions and lymph nodes in DLIR-H images of ULDCT are highly close to those of contrast-enhanced CT. DLIR-based ULDCT may facilitate evaluating target lesions with greatly reduced radiation exposure in tumor evaluation and lung cancer screening.

show abstract

The influence of a deep learning image reconstruction algorithm on the image quality and auto-analysis of pulmonary nodules at ultra-low dose chest CT: a phantom study

Cited by 10 publications

References 28 publications

Computed Tomography 2.0

Computed Tomography 2.0

Detectability and Volumetric Accuracy of Pulmonary Nodules in Low-Dose Photon-Counting Detector Computed Tomography: An Anthropomorphic Phantom Study

Measurement Accuracy and Repeatability of RECIST-Defined Pulmonary Lesions and Lymph Nodes in Ultra-Low-Dose CT Based on Deep Learning Image Reconstruction

Contact Info

Product

Resources

About