A Comparative Study of Target Reconstruction of Ultra-High-Resolution CT for Patients with Corona-Virus Disease 2019 (COVID-19)
Lv, S.; Lin, Y.; Kang, J.; Duan, S.; Zhang, W.; Wang, J.-a.
Show abstract
BackgroundThe corona-virus disease 2019 (COVID-19) pandemic has caused a serious public health risk. Compared with conventional high-resolution CT (C-HRCT, matrix 512), ultra-high resolution CT (U-HRCT, matrix 1024) can increase the effective pixel per unit volume by about 4 times. Our study is to evaluate the value of target reconstruction of U-HRCT in the accurate diagnosis of COVID-19. MethodsA total of 13 COVID-19 cases, 44 cases of other pneumonias, and 6 cases of ground-glass nodules were retrospectively analyzed. The data were categorized into groups A (C-HRCT) and B (U-HRCT), following which iDose4-3 and iDose4-5 were used for target reconstruction, respectively. CT value, noise, and signal-to-noise ratio (SNR) in different reconstructed images were measured. Two senior imaging doctors scored the image quality and the structure of the lesions on a 5-point scale. Chi-square test, variance analysis, and binarylogistic regression analysis were used for statistical analysis. ResultsU-HRCT image can reduce noise and improve SNR with an increase of the iterative reconstruction level. The SNR of U-HRCT image was lower than that of the C-HRCT image of the same iDose4level, and the noise of U-HRCT was higher than that of C-HRCT image; the difference was statistically significant (P< 0.05). Logistic regression analysis showed thatperipleural distribution, thickening of blood vessels and interlobular septum, and crazy-paving pattern were independent indictors of the COVID-19 on U-HRCT. U-HRCT was superior to C-HRCT in showing the blood vessels, bronchial wall, and interlobular septum in the ground-glass opacities; the difference was statistically significant (P < 0.05). ConclusionsPeripleural distribution, thickening of blood vessels and interlobular septum, and crazy-paving pattern on U-HRCT are favorable signs for COVID-19. U-HRCT is superior to C-HRCT in displaying the blood vessels, bronchial walls, and interlobular septum for evaluating COVID-19.
Matching journals
The top 5 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Predicting EGFR mutation status in lung adenocarcinoma presenting as ground-glass opacity: utilizing radiomics model in clinical translation 94%
- From Community Acquired Pneumonia to COVID-19: A Deep Learning Based Method for Quantitative Analysis of COVID-19 on thick-section CT Scans 93%
- First-generation clinical dual-source photon-counting CT: ultra-low dose quantitative spectral imaging 92%
Similar papers in this journal
- Quantitative analysis of chest computed tomography of COVID-19 pneumonia using a software widely used in Japan 94%
- Accuracy of deep learning based computed tomography diagnostic system of COVID-19: a consecutive sampling external validation cohort study 93%
- A metal artifact reduction method for small field of view CT imaging. 93%
Similar papers in this journal
- Computed Tomography Features of COVID-19 in Children: A Systematic Review and Meta-analysis 90%
- Upregulation of ARHGAP9 is correlated with poor prognosis and immune infiltration in clear cell renal cell carcinoma 90%
- The PBL teaching method in Neurology Education in the Traditional Chinese Medicine undergraduate students: An Observational Study 89%
Similar papers in this journal
- Deep learning-based model for detecting 2019 novel coronavirus pneumonia on high-resolution computed tomography: a prospective study 93%
- High-Dimensional Multinomial Multiclass Severity Scoring of COVID-19 Pneumonia Using CT Radiomics Features and Machine Learning Algorithms 93%
- Multi-center Clinical Study Using Optical Coherence Tomography for Evaluation of Cervical Lesions In-vivo 93%
Similar papers in this journal
- Fully Automated Explainable Abdominal CT Contrast Media Phase Classification Using Organ Segmentation and Machine Learning 94%
- Phase Recognition in Contrast-Enhanced CT Scans based on Deep Learning and Random Sampling 91%
- Selective ensemble methods for deep learning segmentation of major vessels in invasive coronary angiography 89%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.