Deep Learning‑based Ultra-high-resolution CT imaging of Viral Pneumonia at Admission and after Discharge

Deep Learning‑based Ultra-high-resolution CT imaging of Viral Pneumonia at Admission and after Discharge | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Deep Learning‑based Ultra-high-resolution CT imaging of Viral Pneumonia at Admission and after Discharge Yanli Gao, Boyang Pan, Lin Niu, Libo Xu, Ziheng Guo, Weican Liu, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7446633/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 04 Apr, 2026 Read the published version in BMC Medical Imaging → Version 1 posted 10 You are reading this latest preprint version Abstract Background Advancement in deep learning has introduced significant potential for enhancing CT image quality without increasing patient radiation exposure. In this study, we sought to compare deep learning‑based ultra-high-resolution CT (UHRCT-DL) findings of viral pneumonia at admission and after discharge with that of HRCT images. Methods A total of 51 inpatients (mean age 66.78 years; 33 males) of viral pneumonia underwent 102 CT scans at admission and after discharge. A deep learning-based super-resolution model, incorporating a dual-branch architecture for super-resolution and gradient guidance, was used to generate UHRCT-DL. UHRCT-DL and HRCT images were systematically reviewed for viral pneumonia CT findings, including ground-glass opacity (GGO), reticulation, tree-in-bud opacities, consolidation, linear bands, bronchiectasis, and bronchiectasis. Subjective and objective CT image quality was evaluated using a five-point Likert scale (− 2 to 2) and lung signal-to-noise ratios (SNRs). Results The score of clarity of CT findings was significantly higher on UHRCT-DL for all CT findings at admission and after discharge. Compared with HRCT as reference image, the most frequently observed additional/different CT findings on UHRCT-DL at admission were crazy paving pattern (14/51, 27%) and tree-in-bud opacities (8/31, 26%), whereas reticulations (15/51, 29%) and bronchiolectasis (12/44, 27%) were most observed additional/different CT findings after discharge. The subjective and objective image quality of UHRCT-DL was superior to that of HRCT. UHRCT-DL algorithm significantly lowered the level of image noise and improved SNR (19.96 ± 6.46 vs 41.35 ± 11.49, p < 0.001). Conclusions The deep learning‑based UHRCT provided a more precise depiction of CT features of viral pneumonia, that better reflects the inflammatory changes during acute phase and early fibrotic changes during recovery. Multidetector Computed Tomography Deep Learning Pneumonia Viral Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Background High-resolution CT (HRCT) is widely recognized as a primary imaging modality for assessing pulmonary diseases, particularly in diagnosis diffuse lung abnormalities 1 . Despite its efficacy, there is an ongoing clinical demand to improve spatial resolution and image quality to enable more precise delineation of anatomical structures and subtle pathological changes 2 , 3 . Ultra-high-resolution CT (UHRCT), incorporating advanced scanning and image reconstruction algorithms has emerged in response to these clinical demands 4 – 6 . Recently advancement in deep learning has introduced significant potential for enhancing CT image quality without increasing patient radiation exposure 7 , 8 . In recent years, viral pneumonia has gained significant attention due to the emergence of new viruses and the impact of known viruses on global health. CT features of viral pneumonia are complex and diverse 9 , 10 , which underscore the critical role of high-quality imaging in assessing lung injury. Among patients with moderate to severe pneumonia, chest HRCT findings reveal pathological changes of diffuse lung injury with time course change from hospitalization to convalescence 11 . Modern advanced CT imaging holds promise in demonstrating pulmonary manifestations of epithelial and endothelial injuries, narrowing differential diagnoses, aiding in the identification of at-risk patients and offering prognosis for viral pneumonia 12 . However, the potential of UHRCT to improve diagnostic precision and clinical interpretation in viral pneumonia remains largely unexplored. With the rise of artificial intelligence in recent years, there has been an increasingly application of super-resolution techniques in medical imaging, aiming to enhance CT image quality while reducing scan time 13 . Recently, Wei Lu et al. introduced an innovative densely connected network aimed at super-resolution reconstruction of 3D medical images 14 . This network employs a 3D dilated convolution module to expand the receptive field and capture multi-scale information effectively. However, this approach overlooks the possible correlations among 3D medical images. Qiu et al. proposed a deep learning-based super resolution model to improve the resolution of CT images, which utilize a residual dense attention network (RDAN) to better extract features of CT images 15 . However, these research primarily addresses algorithmic improvements, often overlooking the clinical utility of super-resolution reconstructions for improving detection accuracy of viral pneumonia. This study aims to evaluate the effectiveness of deep learning-based UHRCT (UHRCT-DL) compared to routine HRCT in assessment lung abnormalities of viral pneumonia in hospitalized patients. We hypothesized that UHRCT-DL would provide more detailed insights into pathological changes, thereby enhancing clinical interpretation and decision-making. Methods This retrospective study was approved by the institutional review board of Beijing Chao-yang Hospital that waived the requirement for patient written informed consent. Study population Between January 2023 and January 2024, a total of 2,596 patients with viral pneumonia related symptoms underwent chest CT and were diagnosed “pneumonia” via CT in our hospital. Inclusion criteria were: (a) age of at least 18 years; (b) confirmed diagnosis of viral pneumonia, including COVID-19 and Influenza A, via positive polymerase chain reaction (PCR) test and admitted into hospital; (c) underwent chest CT at admission and after discharge respectively. Exclusion criteria consisted of (a) outpatients or no positive PCR test available; (b) critical patients with tracheal intubation; (c) no follow-up CT available / CT at admission>14 days from symptom onset / CT after discharge ≤ 28 days or>3months; (d) history of interstitial lung disease, lung malignancy, lung surgery or other severe lung diseases; (e) obvious breath artifacts affecting the detectability of chest CT. A sample size of 51 patients with 102 chest CT were included and evaluated with routine HRCT and UHRCT based on deep learning technologies. CT protocol Chest CT examinations were acquired at both admission and post-discharge for every patient, using four different multidetector CT systems: Dual-source Force CT scanner (Siemens, Erlangen, Germany) for 22 patients, iCT scanner (Philips, Amsterdam, Netherlands) for 14 patients, Optima CT520 scanner (GE Medical Systems, Milwaukee, WI) for 11 patients and Neuviz Prime scanner (Neusoft Medical Systems, Shenyang, China) for 4 patients. All scans were performed in the supine position with full-inspiration breath hold, without intravenous contrast agent. HRCT images were reconstructed with 0.6-1mm collimation, using high-resolution algorithm (B64 for Force CT, Sharp for iCT, Bone for Optima CT520 and lung 10 for Neuviz Prime), and observed on lung settings (window width, 1500 HU; level, − 700 HU). DL Reconstruction Structure The construction of the RaDynCT model was inspired by the structure-preserving super-resolution (SPSR) model 16 , utilizing gradient guidance from relevant HRCT images to generate clinically relevant UHRCT-DL images that contain richer details and edge information. The generator of the proposed model structure consists of two branches: the SR branch and the gradient branch. The SR branch receives HRCT images as input, while the gradient branch takes the gradient maps of HRCT images as input to guide the SR main branch in learning the detailed features of the HRCT images. This structure facilitates the generation of UHRCT-DL images from HRCT, highlighting clearer structural features. Figure 1 illustrate the details of the proposed model structure. Image Analysis Two radiologists (G.Y. (reader 1, 26 years’ experience) and N.L. (reader 2, 8 years' experience)) independently assessed the CT images, blinded to imaging reports and clinical data. CT findings were evaluated across seven categories according to the Fleischner Society’s glossary 17 , including ground-glass opacity (GGO), reticulation, tree-in-bud opacities (included both vascular and bronchiolar tree-in-bud opacities), consolidation, linear bands, bronchiectasis and bronchiolectasis by 4-point scale (0: absent; 1: probably present; 2: present but blurred; 3: present and clear). To assess the detectability of subtle findings with UHRCT-DL, comparison between the UHRCT-DL and HRCT images for the same patient were made for the presence of additional/different CT findings, defined as CT findings in lung areas where HRCT showed no or different CT findings (eg, reticulations on UHRCT-DL images classified as GGO according to HRCT images). Following the same principle, subjective image quality differences were evaluated using a five-point Likert scale: 1) − 2, definitely worse, with probable effect on detectability of CT findings; 2) − 1, definitely worse, without effect on detectability of CT findings; 3) 0, similar, not definitely worse or better; 4)1, definitely better, without effect on detectability of CT findings; 5) 2, definitely better, with probable effect on detectability of CT findings. Analysis of CT findings and subjective image quality were obtained by a consensus between two chest radiologists, who were blinded to the scanner type and clinical conditions. Objective image noise was quantified by placing a circular region of interest (ROIs) with a diameter of 10 mm within the tracheal lumen, just above the level of the aortic arch. The standard deviation of this measurement represented image noise. Signal-to-noise ratio (SNR) was then calculated by dividing the mean Hounsfield unit density of each ROI by its SD. This process was performed three times for each series to obtain mean lung SNRs as assessed by Reader 2. Statistical Analyses Statistical Analyses were performed using SPSS software, version 21 (IBM Corporation, Armonk, NY, USA). Continuous variables were tested for normal distribution using the Shapiro-Wilk test. Normally distributed variables were expressed as mean ± SD, and non-normally distributed variables were expressed as median and IQR. Categorical variables were expressed as absolute numbers and their percentages. The clarity of CT findings was assessed and compared between HRCT and UHRCT-DL images using the Wilcoxon signed-rank test. Differences in subjective image quality between UHRCT-DL images and corresponding HRCT reference images were evaluated for statistical significance using the Wilcoxon signed-rank test. For objective image quality comparisons between HRCT and UHRCT-DL, the paired Student’s t-test was applied when the data followed a normal distribution, otherwise the Wilcoxon signed-rank test was used. A two-sided P value of less than 0.05 was considered indicative of a statistically significant difference. Results Characteristics of the study population The final study sample consisted of 51 patients with viral pneumonia, selected after the exclusion of 2,545 participants (Fig. 2 ). The study population included 33 males and 18 females with a mean age of 66.78 years. The median intervals from symptom onset to CT examination were 7 days at admission and 49 days post-discharge. On HRCT images, ground-glass opacity (GGO) and reticulation were the most common lung abnormalities observed at both admission and post-discharge. Findings of tree-in-bud opacities and consolidations were more frequent at admission, whereas linear bands, bronchiectasis and bronchiolectasis appeared more commonly in post-discharge CT images. Key characteristics of the study population are given in Table 1 . Table 1 Demographic and characteristics Characteristic At Admission (N = 51) After Discharge (N = 51) Male 33 (64.7%) - Female 18 (35.3%) - Mean age, y 66.78 ± 7.38 - Time from onset to CT, d 7 (4,10) 49 (35,60) CT findings GGO 51 (100%) 51 (100%) Consolidations 39 (76.47%) 23 (45.10%) Reticulations 45 (88.24%) 47 (92.16%) TIB 31 (60.78%) 5 (9.80%) Linear bands 6 (11.76%) 30 (58.82%) Bronchiectasis 21 (41.18%) 37 (72.56%) Bronchiolectasis 29 (56.86%) 44 (86.27%) Note: CT findings were evaluated on HRCT GGO: ground glass opacities; TIB: tree-in-bud opacities, including both bronchial and vascular TIB Evaluation of CT findings A total of 51 patients with 102 chest CT scans were evaluated using both routine HRCT and UHRCT-DL. As summarized in Table 2 , image clarity scores were significantly higher on UHRCT-DL for detecting ground-glass opacity (GGO), reticulation, tree-in-bud opacities, consolidations, linear bands, bronchiectasis, and bronchiolectasis at both admission and post-discharge. Notably, 2 cases (2%) of reticulation and 2 cases (3%) of bronchiolectasis were assigned a score of 1 on HRCT, whereas no findings received a score of 1 on UHRCT-DL. Image ratings were enhanced from 2 to 3 on UHRCT-DL for 39 cases (38%) of GGO, 7 cases (11%) of consolidation, 40 cases (43%) of reticulation, 13 cases (36%) of tree-in-bud opacities, 5 cases (14%) of linear band s, 17 cases (29%) of bronchiectasis, and 30 cases (41%) of bronchiolectasis. Table 2 Evaluation of CT findings in HRCT and UHRCT-DL CT findings HRCT (N = 102) UHRCT-DL (N = 102) Value P value GGO (102) 2 (2,3) 3 (3,3) Z =-6.245 0.000 Score 0 - - Score 1 - - Score 2 45 6 Score 3 57 96 Consolidations (62) 3 (0,3) 3 (0,3) Z =-2.646 0.008 Score 0 40 40 Score 1 - - Score 2 9 2 Score 3 53 60 Reticulations (92) 2 (2,3) 3 (3,3) Z =-6.557 0.000 Score 0 10 10 Score 1 2 - Score 2 44 6 Score 3 46 86 TIB (36) 0 (0,2) 0 (0,3) Z =-3.606 0.000 Score 0 66 66 Score 1 - - Score 2 17 4 Score 3 19 32 Linear bands (36) 0 (0,3) 0 (0,3) Z =-2.236 0.025 Score 0 66 66 Score 1 - - Score 2 6 1 Score 3 30 35 Bronchiectasis (58) 2 (0,3) 2 (0,3) Z =-4.123 0.000 Score 0 44 44 Score 1 - - Score 2 27 10 Score 3 31 48 Bronchiolectasis (73) 2 (0,2) 3 (0,3) Z =-5.657 0.000 Score 0 29 29 Score 1 2 - Score 2 46 18 Score 3 25 55 GGO: ground glass opacities; TIB: tree-in-bud opacities Table 3 Additional or different CT findings detected with UHRCT-DL HRCT UHRCT-DL At Admission (N = 51) UHRCT-DL After Discharge (N = 51) abnormalities Additional Different Total Additional Different Total GGO - 14 (27.45%) 51 - 15 (29.41%) 51 Consolidation - - 39 - - 23 Reticulations - - 45 - - 47 TIB 8 (25.81%) - 31 1 (20.00%) - 5 Linear bands - - 6 - - 30 Bronchiectasis - - 21 - - 37 Bronchiolectasis 6 (20.69%) - 29 12 (27.27%) - 44 GGO: ground glass opacities; TIB: Tree-in-bud opacities UHRCT-DL revealed additional/different CT findings in 23 patients (45%) and in 31 (30%) CT scans (15 + 16 = 31), compared to HRCT as the reference. In 29 CT scans (28%), areas identified as ground-glass opacities (GGO) on HRCT exhibited fine reticulations on UHRCT-DL, which were categorized as distinct CT findings of reticulations or crazy paving pattern when they overlapped with GGO 17 . The most frequently observed additional/different CT findings on UHRCT-DL at admission were crazy paving pattern (14/51, 27%) and tree-in-bud opacities (8/31, 26%), whereas reticulations (15/51, 29%) and bronchiolectasis (12/44, 27%) were most observed additional/different CT findings after discharge (Fig. 3 – 5 ). No additional GGO, consolidations, reticulations, bronchiectasis or linear bands were found on UHRCT-DL. Table 3 showed detailed additional/different CT findings detected with UHRCT-DL. Table 4 Subjective and objective image quality of UHRCT-DL in comparison with HRCT Image Quality UHRCT-DL (N = 102) HRCT(N = 102) Value P value Subjective Noise magnitude Z =-9.221 0.000 -2 0 (0%) 0 -1 0 (0%) 0 0 0 (0%) 0 1 73 (71.57%) 0 2 29 (28.43%) 0 Streak artifact Z =-6.856 0.000 -2 0 (0%) 0 -1 0 (0%) 0 0 55 (53.92%) 0 1 47 (46.08%) 0 2 0 (0%) 0 Image sharpness Z =-8.702 0.000 -2 0 (0%) 0 -1 0 (0%) 0 0 10 (9.80%) 0 1 62 (60.78%) 0 2 30 (29.41%) 0 Objective (HU) Mean 971.69 ± 14.01 964.12 ± 13.32 t = 12.743 0.000 SD 52.96 ± 14.49 25.20 ± 7.26 t = 31.638 0.000 SNR 19.96 ± 6.46 41.35 ± 11.49 t =-30.088 0.000 Table 4 summarizes subjective and objective image quality of UHRCT-DL in comparison with HRCT. On UHRCT-DL, the subjective assessment of image quality was rated as equivalent (score 0) or superior (scores + 1 and + 2) compared to HRCT, achieving the highest cumulative scores of ≥ + 1 for noise magnitude (102/102, 100%). For image sharpness and streak artifact, 90.20% (92/102) and 46.08% (47/102) UHRCT-DL images were evaluated as definitely better than HRCT images respectively. As to objective image quality, UHRCT-DL algorithm significantly lowered the mean level of image noise and improved SNR ( p < 0.001), mean CT value was slightly increased (8/964HU, 0.78%) at the same time ( p < 0.001). Discussion In this retrospective study, we investigated the application of deep learning-based UHRCT to assess pulmonary abnormalities associated with viral pneumonia. A total of 102 CT scans from 51 patients admitted into hospital with viral pneumonia were analyzed, comparing images acquired at admission and following discharge with those obtained from HRCT. Our findings reveal that UHRCT-DL demonstrates superior image quality and clarity for identifying various parenchymal and interstitial abnormalities compared to HRCT. The additional/different UHRCT-DL findings of crazy paving pattern and tree-in-bud opacities at admission indicated the inflammatory changes of interstitium, bronchioles and micro vessels during acute phase, whereas reticulations and bronchiolectasis after discharge characterized fibrosis changes during recovery 12 , 18 . Thus, the improvement of image quality by UHRCT-DL may better reflect pathological change and benefit the clinical performance of viral pneumonia. In recent years, UHRCT has been explored using advanced scanning techniques 19 – 22 . For example, Prayer F, et al found that photon-counting detector (PCD) CT revealed subtle lung abnormalities in symptomatic participants with persistent post- viral pneumonia symptoms 8 . Yanagawa M, et al reported that super-high-resolution mode with 0.25-mm section thickness and a 2048 matrix have high diagnostic performance to predict invasiveness of lung adenocarcinoma 23 . To some extent, both methods are not yet extensively facilitated in most hospitals and may involve increased radiation doses. Deep learning-based reconstruction techniques offer an alternative by enhancing image quality through post-processing without additional radiation exposure or specialized equipment. 24 – 26 . Several studies have highlighted the potential of deep learning in improving the quality of CT imaging, especially in lung examination. Hata et al. proposed a method combining deep learning–based denoising (DLD) and iterative reconstruction (IR) to improve the image quality of lung CT imaging, which utilize the ultra-low-dose CT to reconstruct the standard-does CT 27 . It can effectively reduce noise in low-dose CT images and improve spatial resolution. Nagayama et al. proposed a deep-learning-based reconstruction (SR-DLR) method to evaluate the image quality of coronary CT angiography (CCTA), which can effectively improve image clarity and noise characteristics while reducing halo artifacts caused by calcified plaques 6 , 28 . However, these approaches did not explore the comparative benefits of UHRCT over HRCT in clinical diagnosis. In contrast, our study focuses specifically on the advantages of UHRCT-DL for detecting pulmonary abnormalities in viral pneumonia, providing evidence of its clinical utility. Our results find that UHRCT-DL achieved consistently higher subjective image quality scores, with all scans rated either as equivalent (score 0) or superior (scores + 1 and + 2) to HRCT. Regarding objective image quality, the UHRCT-DL algorithm notably reduced mean image noise and improved the SNR, both with high statistical significance ( p < 0.001). Additionally, UHRCT-DL demonstrated superior clarity for subtle CT findings, such as fine reticulations, tree-in-bud opacities, and bronchiolectasis, which were less distinguishable on HRCT. Additionally, in 28% CT scans, the detail of GGO were shown on UHRCT-DL as crazy paving pattern or reticulations which reflects interstitial changes of viral pneumonia not shown on routine HRCT. Furthermore, 2 cases (2%) of reticulation and 2 cases (3%) of bronchiolectasis were rated as “probably present” (score 1) on HRCT, suggesting that the elimination of ambiguous findings (score 1) on UHRCT-DL may enhance diagnostic confidence for clinicians. At admission, the most common additional/different CT findings on UHRCT-DL included crazy paving patterns and tree-in-bud opacities. The crazy paving pattern, characterized by focal area of GGO with septal thickening and intralobular reticular lines, indicate alveolar septal inflammation and fibroblast proliferation in the early stage of viral pneumonia. The different CT findings of crazy paving patterns on UHRCT-DL reflects the inflammation of both parenchyma and interstitium 12 . The tree-in-bud pattern is conventionally used to describe infection and inflammation of the bronchioles. Viral pneumonia may affect both respiratory and circulatory system 29 , 30 . Patel et al first introduced the vascular tree-in-bud pattern in severe COVID-19 as a characteristic pattern 18 . Interestingly, this CT finding was found to be correlated with longer ventilation and hospitalization, highlighting its prognostic role 12 . After patients’ discharge, reticulations and bronchiolectasis were most observed additional/different CT findings. In areas that had been classified as GGO at HRCT, UHRCT-DL showed fine reticulations, which are typical manifestation of interstitial lung abnormalities in 15 cases. Traction bronchiolectasis, defined as dilatation of bronchioles within the central portions of the secondary pulmonary lobule, is typically associated with fibrosis during recovery. Therefore, the distinctive CT findings identified by deep learning-based UHRCT may more accurately reflect underlying pathological changes, enhancing clinical insight and management of viral pneumonia. Our study has several limitations. First, the inclusion of only hospitalized patients with complete medical records limits the generalizability of findings to individuals with mild or long-term pulmonary abnormalities. Second, the single-center, retrospective design and the relatively small sample size underscore the need for larger, multicenter studies to confirm our findings across a broader patient population. Finally, images were acquired from various CT scanners using different reconstruction algorithms at conventional radiation doses. Future studies exploring UHRCT-DL with low-dose protocols may provide further insights into optimizing image quality while minimizing radiation exposure. Conclusion Deep learning-based UHRCT significantly enhanced image quality, offering a more precise depiction of viral pneumonia related CT features. The unique findings observed on UHRCT-DL may more accurately reflect inflammatory changes during the acute phase and early fibrotic changes during recovery. These advancements could provide a methodological reference for future research on CT image reconstruction in viral pneumonia and other diffuse lung diseases. Declarations Ethics approval and consent to participate This study was performed in accordance with the ethical principles of Declaration of Helsinki. Consent for publication Not applicable. Competing Interests Author B.P. and L.X. are stockholder of RadioDynamic Medical Funding This study has received funding by Beijing Hospitals Authority Clinical Medicine Development of Special Funding Support (ZLRK202306), Natural Science Foundation of Xiamen, China (3502Z202373090) and National Natural Science Foundation of China (82000103). Author Contribution Yanli Gao: Writing – original draft, Writing – review & editing, Visualization, Validation, Resources, Project administration, Methodology, Investigation, Formal analysis, Conceptualization.Boyang Pan: Data curation, Writing – review & editing, ValidationLin Niu: Formal analysis, InvestigationLibo Xu: Data curationZiheng Guo: Data curationWeican Liu: Data curationPenghui Sun: ResourcesYanyan Zhang: ResourcesXiaoli Xu: Funding acquisitionNanjie Gong: Methodology, Software, Funding acquisitionQi Yang: Supervision, Project administration, Conceptualization, Funding acquisition Acknowledgements Not applicable. Data Availability The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request. References Brixey AG, Oh AS, Alsamarraie A, Chung JH. Pictorial Review of Fibrotic Interstitial Lung Disease on High-Resolution CT Scan and Updated Classification. Chest 2024;165(4):908-923. DOI: 10.1016/j.chest.2023.11.037. Marton N, Gyebnar J, Fritsch K, et al. Photon-counting computed tomography in the assessment of rheumatoid arthritis-associated interstitial lung disease: an initial experience. Diagn Interv Radiol 2023;29(2):291-299. DOI: 10.4274/dir.2023.221959. Ohno Y, Akino N, Fujisawa Y, et al. Comparison of lung CT number and airway dimension evaluation capabilities of ultra-high-resolution CT, using different scan modes and reconstruction methods including deep learning reconstruction, with those of multi-detector CT in a QIBA phantom study. Eur Radiol 2023;33(1):368-379. DOI: 10.1007/s00330-022-08983-1. Tatsugami F, Higaki T, Kawashita I, et al. Improvement of Spatial Resolution on Coronary CT Angiography by Using Super-Resolution Deep Learning Reconstruction. Acad Radiol 2023;30(11):2497-2504. DOI: 10.1016/j.acra.2022.12.044. Nagayama Y, Iwashita K, Maruyama N, et al. Deep learning-based reconstruction can improve the image quality of low radiation dose head CT. Eur Radiol 2023;33(5):3253-3265. DOI: 10.1007/s00330-023-09559-3. Sun Y, Pan B, Li Q, et al. Clinical ultra-high resolution CT scans enabled by using a generative adversarial network. Med Phys 2023;50(6):3612-3622. DOI: 10.1002/mp.16172. Gaillandre Y, Duhamel A, Flohr T, et al. Ultra-high resolution CT imaging of interstitial lung disease: impact of photon-counting CT in 112 patients. Eur Radiol 2023;33(8):5528-5539. DOI: 10.1007/s00330-023-09616-x. Prayer F, Kienast P, Strassl A, et al. Detection of Post-COVID-19 Lung Abnormalities: Photon-counting CT versus Same-Day Energy-integrating Detector CT. Radiology 2023;307(1):e222087. DOI: 10.1148/radiol.222087. Koo HJ, Lim S, Choe J, Choi S-H, Sung H, Do K-H. Radiographic and CT Features of Viral Pneumonia. RadioGraphics 2018;38(3):719-739. DOI: 10.1148/rg.2018170048. Koo HJ, Choi S-H, Sung H, Choe J, Do K-H. RadioGraphics Update: Radiographic and CT Features of Viral Pneumonia. RadioGraphics 2020;40(4):E8-E15. DOI: 10.1148/rg.2020200097. Pan F, Ye T, Sun P, et al. Time Course of Lung Changes at Chest CT during Recovery from Coronavirus Disease 2019 (COVID-19). Radiology 2020;295(3):715-721. DOI: 10.1148/radiol.2020200370. Lee JH, Koh J, Jeon YK, Goo JM, Yoon SH. An Integrated Radiologic-Pathologic Understanding of COVID-19 Pneumonia. Radiology 2023;306(2):e222600. DOI: 10.1148/radiol.222600. Sandino CM, Cheng JY, Chen F, Mardani M, Pauly JM, Vasanawala SS. Compressed Sensing: From Research to Clinical Practice with Deep Neural Networks. IEEE Signal Process Mag 2020;37(1):111-127. DOI: 10.1109/MSP.2019.2950433. Wei Lu ZS, Jinghui Chu. A novel 3D medical image super-resolution method based on densely connected network. Biomedical Signal Processing and Control 2020;62: 102120. D. Qiu YCaXW. Residual Dense Attention Networks for COVID-19 Computed Tomography Images Super Resolution. IEEE Transactions on Cognitive and Developmental Systems 2023;15(2):904-913. Ma C, Rao Y, Lu J, Zhou J. Structure-Preserving Image Super-Resolution. IEEE Trans Pattern Anal Mach Intell 2022;44(11):7898-7911. DOI: 10.1109/TPAMI.2021.3114428. Bankier AA, MacMahon H, Colby T, et al. Fleischner Society: Glossary of Terms for Thoracic Imaging. Radiology 2024;310(2):e232558. DOI: 10.1148/radiol.232558. Patel BV, Arachchillage DJ, Ridge CA, et al. Pulmonary Angiopathy in Severe COVID-19: Physiologic, Imaging, and Hematologic Observations. Am J Respir Crit Care Med 2020;202(5):690-699. DOI: 10.1164/rccm.202004-1412OC. Inoue A, Johnson TF, Walkoff LA, et al. Lung Cancer Screening Using Clinical Photon-Counting Detector Computed Tomography and Energy-Integrating-Detector Computed Tomography: A Prospective Patient Study. J Comput Assist Tomogr 2023;47(2):229-235. DOI: 10.1097/RCT.0000000000001419. Dunning CAS, Marsh JF, Jr., Winfree T, et al. Accuracy of Nodule Volume and Airway Wall Thickness Measurement Using Low-Dose Chest CT on a Photon-Counting Detector CT Scanner. Invest Radiol 2023;58(4):283-292. DOI: 10.1097/RLI.0000000000000933. Chamberlin JH, Smith CD, Maisuria D, et al. Ultra-high-resolution photon-counting detector computed tomography of the lungs: Phantom and clinical assessment of radiation dose and image quality. Clin Imaging 2023;104:110008. DOI: 10.1016/j.clinimag.2023.110008. Hata A, Yanagawa M, Ninomiya K, et al. Photon-Counting Detector CT Radiological-Histological Correlation in Cadaveric Human Lung Nodules and Airways. Invest Radiol 2025;60(2):151-160. DOI: 10.1097/RLI.0000000000001117. Yanagawa M, Tsubamoto M, Satoh Y, et al. Lung Adenocarcinoma at CT with 0.25-mm Section Thickness and a 2048 Matrix: High-Spatial-Resolution Imaging for Predicting Invasiveness. Radiology 2020;297(2):462-471. DOI: 10.1148/radiol.2020201911. Sakai Y, Hida T, Matsuura Y, et al. Impact of a new deep-learning-based reconstruction algorithm on image quality in ultra-high-resolution CT: clinical observational and phantom studies. Br J Radiol 2023;96(1141):20220731. DOI: 10.1259/bjr.20220731. Nakamura Y, Narita K, Higaki T, Akagi M, Honda Y, Awai K. Diagnostic value of deep learning reconstruction for radiation dose reduction at abdominal ultra-high-resolution CT. Eur Radiol 2021;31(7):4700-4709. DOI: 10.1007/s00330-020-07566-2. Wiebe M, Haston C, Lamey M, Narayan A, Rajapakshe R. The effect of spatial resolution on deep learning classification of lung cancer histopathology. BJR Open 2023;5(1):20230008. DOI: 10.1259/bjro.20230008. Hata A, Yanagawa M, Yoshida Y, et al. Combination of Deep Learning-Based Denoising and Iterative Reconstruction for Ultra-Low-Dose CT of the Chest: Image Quality and Lung-RADS Evaluation. AJR Am J Roentgenol 2020;215(6):1321-1328. DOI: 10.2214/AJR.19.22680. Nagayama Y, Emoto T, Kato Y, et al. Improving image quality with super-resolution deep-learning-based reconstruction in coronary CT angiography. Eur Radiol 2023;33(12):8488-8500. DOI: 10.1007/s00330-023-09888-3. Chen X, Long Z, Lei Y, et al. CT Differentiation and Prognostic Modeling in COVID-19 and Influenza A Pneumonia. Acad Radiol. 2025 Jul;32(7):4289-4300. doi: 10.1016/j.acra.2025.02.004. Jin L, Chen J, Wu L, et al. Central artery pulse pressure, not central arterial stiffness impact on all-cause mortality in patients with viral pneumonia infection. BMC Infectious Diseases 2024;24(1). DOI: 10.1186/s12879-024-10091-y. Additional Declarations Competing interest reported. Author B.P. and L.X. are stockholder of RadioDynamic Medical Cite Share Download PDF Status: Published Journal Publication published 04 Apr, 2026 Read the published version in BMC Medical Imaging → Version 1 posted Editorial decision: Revision requested 06 Mar, 2026 Reviews received at journal 05 Mar, 2026 Reviewers agreed at journal 17 Feb, 2026 Reviews received at journal 08 Nov, 2025 Reviewers agreed at journal 02 Nov, 2025 Reviewers invited by journal 22 Sep, 2025 Editor assigned by journal 17 Sep, 2025 Editor invited by journal 08 Sep, 2025 Submission checks completed at journal 05 Sep, 2025 First submitted to journal 05 Sep, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7446633","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":524106640,"identity":"83cd8c05-364c-467d-a2ef-d75632d7adc5","order_by":0,"name":"Yanli Gao","email":"","orcid":"","institution":"Beijing Chao-Yang Hospital, Capital Medical University","correspondingAuthor":false,"prefix":"","firstName":"Yanli","middleName":"","lastName":"Gao","suffix":""},{"id":524106641,"identity":"9a4d16b0-ba43-49d7-9e42-b48e9b4d48dc","order_by":1,"name":"Boyang Pan","email":"","orcid":"","institution":"Laboratory for Intelligent Medical Imaging, Tsinghua Cross-Strait Research Institute,","correspondingAuthor":false,"prefix":"","firstName":"Boyang","middleName":"","lastName":"Pan","suffix":""},{"id":524106642,"identity":"c2bf1adb-b7e4-44db-b4df-9bd13625298b","order_by":2,"name":"Lin Niu","email":"","orcid":"","institution":"Beijing Chao-Yang Hospital, Capital Medical University","correspondingAuthor":false,"prefix":"","firstName":"Lin","middleName":"","lastName":"Niu","suffix":""},{"id":524106643,"identity":"610ce83d-8045-4a21-bead-1f2a352ad212","order_by":3,"name":"Libo Xu","email":"","orcid":"","institution":"RadioDynamic Medical","correspondingAuthor":false,"prefix":"","firstName":"Libo","middleName":"","lastName":"Xu","suffix":""},{"id":524106644,"identity":"8b3c991b-9ea1-42e7-b1f4-3b2daf5ed831","order_by":4,"name":"Ziheng Guo","email":"","orcid":"","institution":"RadioDynamic Medical","correspondingAuthor":false,"prefix":"","firstName":"Ziheng","middleName":"","lastName":"Guo","suffix":""},{"id":524106645,"identity":"b165d101-5561-4c88-9f7b-1056f6c5c721","order_by":5,"name":"Weican Liu","email":"","orcid":"","institution":"RadioDynamic Medical","correspondingAuthor":false,"prefix":"","firstName":"Weican","middleName":"","lastName":"Liu","suffix":""},{"id":524106646,"identity":"66c29a0a-bf13-4491-9a31-ff069777ba8c","order_by":6,"name":"Penghui Sun","email":"","orcid":"","institution":"Beijing Chao-Yang Hospital, Capital Medical University","correspondingAuthor":false,"prefix":"","firstName":"Penghui","middleName":"","lastName":"Sun","suffix":""},{"id":524106647,"identity":"6912d98a-47d1-46ae-9eac-252cf87509d3","order_by":7,"name":"Yanyan Zhang","email":"","orcid":"","institution":"Beijing Chao-Yang Hospital, Capital Medical University","correspondingAuthor":false,"prefix":"","firstName":"Yanyan","middleName":"","lastName":"Zhang","suffix":""},{"id":524106648,"identity":"486c9575-bfd8-43a2-a444-0ee8652426d0","order_by":8,"name":"Xiaoli Xu","email":"","orcid":"","institution":"Beijing Chao-Yang Hospital, Capital Medical University","correspondingAuthor":false,"prefix":"","firstName":"Xiaoli","middleName":"","lastName":"Xu","suffix":""},{"id":524106649,"identity":"1ee02019-c9fb-4ebe-9700-d5b7d26e756b","order_by":9,"name":"Nan-jie Gong","email":"","orcid":"","institution":"Shanghai Key Laboratory of Magnetic Resonance, School of Physics and Electronic Science, East China Normal University","correspondingAuthor":false,"prefix":"","firstName":"Nan-jie","middleName":"","lastName":"Gong","suffix":""},{"id":524106650,"identity":"a9498d76-00c7-4836-b764-b5f4c894d298","order_by":10,"name":"Qi Yang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA/ElEQVRIiWNgGAWjYFADZgaGAx+ANBs7KVoOzgBpYSbJIh6oXrzA4PjZw695au7Y9R1n3njY5tc2eT5mBsYPH3PwaDmTl2bNc+xZ8szDbAWHc/tuG7YxMzBLztyGR8uBHDNjHrbDyQaHeQwO5/bcZgRqYWPmxafl/Bugln9QLZY9t+0Ja7mRY/yYt+2wHVgLw4/biQS1SN54Y8Y4t+9wgiTQLwd7G24ntzEzNuP1C9/5HOMPb74dtuc7f3jzhx9/btvOb28++OEjHi0KBxjYpIDRkdhwgMGAgbENJMbYgFs9EMg3MDB//MHAYM8A0sLwB6/iUTAKRsEoGKEAABGCWgQoaiY6AAAAAElFTkSuQmCC","orcid":"","institution":"Beijing Chao-Yang Hospital, Capital Medical University","correspondingAuthor":true,"prefix":"","firstName":"Qi","middleName":"","lastName":"Yang","suffix":""}],"badges":[],"createdAt":"2025-08-24 14:08:13","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7446633/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7446633/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s12880-026-02320-4","type":"published","date":"2026-04-04T15:57:56+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":93006321,"identity":"ab38e76f-fe4e-47d5-8a95-b13787320092","added_by":"auto","created_at":"2025-10-08 06:46:53","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2236185,"visible":true,"origin":"","legend":"","description":"","filename":"BMJmedimagingDeepLearningbasedUltrahighresolutionCTImagingofViralPneumoniaatAdmissionandafterDischarge.docx","url":"https://assets-eu.researchsquare.com/files/rs-7446633/v1/91b6e0494bf32939e5627c6e.docx"},{"id":93006316,"identity":"ad7b5f09-00f4-4eec-8b24-829ca3039d68","added_by":"auto","created_at":"2025-10-08 06:46:53","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":11752,"visible":true,"origin":"","legend":"","description":"","filename":"02711755dc58475cb1852794dfc326ee.json","url":"https://assets-eu.researchsquare.com/files/rs-7446633/v1/69eef9b8e1c64235d95ef1a2.json"},{"id":93006320,"identity":"1303306d-c828-411a-844b-b1c3ad72df26","added_by":"auto","created_at":"2025-10-08 06:46:53","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":116357,"visible":true,"origin":"","legend":"","description":"","filename":"02711755dc58475cb1852794dfc326ee1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7446633/v1/f158f1440f7c216a38691e45.xml"},{"id":93006323,"identity":"b586aad7-c3b6-43a1-9fc9-717529ef1c7e","added_by":"auto","created_at":"2025-10-08 06:46:53","extension":"png","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":127207,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7446633/v1/00aab304dc741bc773a317af.png"},{"id":93006318,"identity":"041ccc1a-df0b-441a-b3a2-f1215f45ed21","added_by":"auto","created_at":"2025-10-08 06:46:53","extension":"jpeg","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1074,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7446633/v1/306a7fde10a50f3b901d7124.jpeg"},{"id":93006334,"identity":"7c1a2557-4b94-46c5-882f-a2dfd93b76c1","added_by":"auto","created_at":"2025-10-08 06:46:53","extension":"jpeg","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":401960,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7446633/v1/fdf7214f23be820447378aa8.jpeg"},{"id":93006336,"identity":"6bd3fb22-f464-4109-86ba-7f2048613fc0","added_by":"auto","created_at":"2025-10-08 06:46:53","extension":"jpeg","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":367065,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7446633/v1/38b8f40c11e82659e6e027d7.jpeg"},{"id":93006337,"identity":"d2ec88f5-919e-474e-abe1-34531f7b99ee","added_by":"auto","created_at":"2025-10-08 06:46:53","extension":"jpeg","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":509790,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7446633/v1/0067923fa493c764aff5c078.jpeg"},{"id":93006326,"identity":"633b02cf-922b-47f5-96f1-82c0edbfefa6","added_by":"auto","created_at":"2025-10-08 06:46:53","extension":"jpeg","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":58597,"visible":true,"origin":"","legend":"","description":"","filename":"groupimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7446633/v1/77340986cac2ce5f150d8fbb.jpeg"},{"id":93007680,"identity":"090c031d-2e4b-40ca-8826-ad93e126d4e5","added_by":"auto","created_at":"2025-10-08 06:54:53","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":41224,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7446633/v1/1e0da16f293c8e3476ed0263.png"},{"id":93006325,"identity":"a5069ac0-7b3f-4db4-9314-96dc23d5ddb2","added_by":"auto","created_at":"2025-10-08 06:46:53","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":935,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7446633/v1/e4680ed5b3901beb99582e3f.png"},{"id":93006330,"identity":"89f14465-afd7-4276-a102-d5eb0be76a99","added_by":"auto","created_at":"2025-10-08 06:46:53","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":146645,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7446633/v1/d833a0d2422deec33341940a.png"},{"id":93006332,"identity":"3b48c7a3-7142-4d81-a9ec-cd07a7e448b4","added_by":"auto","created_at":"2025-10-08 06:46:53","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":544237,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7446633/v1/b4213c3dee8a4c52b5b9e15b.png"},{"id":93007684,"identity":"40c434b6-7fbf-41cf-8cf7-1ffc4d1c5e71","added_by":"auto","created_at":"2025-10-08 06:54:53","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":766170,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7446633/v1/7bb1aabc65d72fa71ccde219.png"},{"id":93007682,"identity":"de81118a-e9c9-47ea-bf7a-18172135d24d","added_by":"auto","created_at":"2025-10-08 06:54:53","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":33057,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinegroupimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7446633/v1/0f8f9023ce0d777bc02cb13b.png"},{"id":93007683,"identity":"844b8f7a-8ff7-4985-abc3-286e12070a9b","added_by":"auto","created_at":"2025-10-08 06:54:53","extension":"xml","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":114575,"visible":true,"origin":"","legend":"","description":"","filename":"02711755dc58475cb1852794dfc326ee1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7446633/v1/1c0f95c20ed02df3ca78bdd0.xml"},{"id":93006333,"identity":"45d665a9-621d-4582-911b-7c29f5de0abb","added_by":"auto","created_at":"2025-10-08 06:46:53","extension":"html","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":124652,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7446633/v1/774e071d5b4d138ca3095068.html"},{"id":93007677,"identity":"d6070f09-41fc-4271-a5cf-f6c00a93aeb7","added_by":"auto","created_at":"2025-10-08 06:54:53","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":141873,"visible":true,"origin":"","legend":"\u003cp\u003eIllustration of Our Proposed Framework for UHRCT-DL.\u003c/p\u003e","description":"","filename":"1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7446633/v1/589cb015781a8bf05e3962c6.jpg"},{"id":93006317,"identity":"89955dde-eb2c-458b-bd3b-5830931c9033","added_by":"auto","created_at":"2025-10-08 06:46:53","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":127971,"visible":true,"origin":"","legend":"\u003cp\u003eFlowchart of the study population\u003c/p\u003e","description":"","filename":"2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7446633/v1/b2fda927fd874088e8001193.jpg"},{"id":93007679,"identity":"3dc54dfe-05e1-4220-a039-95c597f1f700","added_by":"auto","created_at":"2025-10-08 06:54:53","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":105271,"visible":true,"origin":"","legend":"\u003cp\u003eHRCT and UHRCT-DL images in a male patient with viral pneumonia at admission into hospital. (A) HRCT image obtained at 0.75-mm collimation. (B) UHRCT-DL image at the same level. Ground-glass opacity detected on A was found to contain reticulations on B (black arrows), defined as different CT findings of crazy paving pattern.\u003c/p\u003e","description":"","filename":"3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7446633/v1/3578ed99b4114568755d570b.jpg"},{"id":93007678,"identity":"c4bd0468-1130-4f36-970b-ce7c920db19d","added_by":"auto","created_at":"2025-10-08 06:54:53","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":95186,"visible":true,"origin":"","legend":"\u003cp\u003eHRCT and UHRCT-DL images in a male patient with viral pneumonia at admission into hospital. (A) HRCT image obtained at 1mm collimation. (B) UHRCT-DL image at the same level. Some fine vascular tree-in-bud opacities (black arrow) were neglected at HRCT but detected at UHRCT-DL with noise reduction.\u003c/p\u003e","description":"","filename":"4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7446633/v1/bb4be7240105ae82c66ca0aa.jpg"},{"id":93006329,"identity":"a58dab00-dadf-4b59-9ea4-8a3bd900c02a","added_by":"auto","created_at":"2025-10-08 06:46:53","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":135904,"visible":true,"origin":"","legend":"\u003cp\u003eHRCT and UHRCT-DL images in a male patient with viral pneumonia after discharge from hospital. (A) HRCT image obtained at 0.7mm collimation. (B) UHRCT-DL image at the same level. Bronchiolectasis and reticulation (black arrow) was not distinguished at HRCT but was found with UHRCT-DL.\u003c/p\u003e","description":"","filename":"5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7446633/v1/ea64437d53582d6a31706a39.jpg"},{"id":106343362,"identity":"45e14058-366d-4bdc-a121-e0493f7fa62c","added_by":"auto","created_at":"2026-04-07 16:03:08","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1581539,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7446633/v1/7e718a3d-de62-4a67-ae38-672b485926b3.pdf"}],"financialInterests":"Competing interest reported. Author B.P. and L.X. are stockholder of RadioDynamic Medical","formattedTitle":"Deep Learning‑based Ultra-high-resolution CT imaging of Viral Pneumonia at Admission and after Discharge","fulltext":[{"header":"Background","content":"\u003cp\u003eHigh-resolution CT (HRCT) is widely recognized as a primary imaging modality for assessing pulmonary diseases, particularly in diagnosis diffuse lung abnormalities \u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e. Despite its efficacy, there is an ongoing clinical demand to improve spatial resolution and image quality to enable more precise delineation of anatomical structures and subtle pathological changes\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e,\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e. Ultra-high-resolution CT (UHRCT), incorporating advanced scanning and image reconstruction algorithms has emerged in response to these clinical demands\u003csup\u003e\u003cspan additionalcitationids=\"CR5\" citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. Recently advancement in deep learning has introduced significant potential for enhancing CT image quality without increasing patient radiation exposure\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e,\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eIn recent years, viral pneumonia has gained significant attention due to the emergence of new viruses and the impact of known viruses on global health. CT features of viral pneumonia are complex and diverse\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e,\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e, which underscore the critical role of high-quality imaging in assessing lung injury. Among patients with moderate to severe pneumonia, chest HRCT findings reveal pathological changes of diffuse lung injury with time course change from hospitalization to convalescence \u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e. Modern advanced CT imaging holds promise in demonstrating pulmonary manifestations of epithelial and endothelial injuries, narrowing differential diagnoses, aiding in the identification of at-risk patients and offering prognosis for viral pneumonia \u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. However, the potential of UHRCT to improve diagnostic precision and clinical interpretation in viral pneumonia remains largely unexplored.\u003c/p\u003e\u003cp\u003eWith the rise of artificial intelligence in recent years, there has been an increasingly application of super-resolution techniques in medical imaging, aiming to enhance CT image quality while reducing scan time\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e. Recently, Wei Lu et al. introduced an innovative densely connected network aimed at super-resolution reconstruction of 3D medical images\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e. This network employs a 3D dilated convolution module to expand the receptive field and capture multi-scale information effectively. However, this approach overlooks the possible correlations among 3D medical images. Qiu et al. proposed a deep learning-based super resolution model to improve the resolution of CT images, which utilize a residual dense attention network (RDAN) to better extract features of CT images\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e. However, these research primarily addresses algorithmic improvements, often overlooking the clinical utility of super-resolution reconstructions for improving detection accuracy of viral pneumonia.\u003c/p\u003e\u003cp\u003eThis study aims to evaluate the effectiveness of deep learning-based UHRCT (UHRCT-DL) compared to routine HRCT in assessment lung abnormalities of viral pneumonia in hospitalized patients. We hypothesized that UHRCT-DL would provide more detailed insights into pathological changes, thereby enhancing clinical interpretation and decision-making.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e This retrospective study was approved by the institutional review board of Beijing Chao-yang Hospital that waived the requirement for patient written informed consent.\u003c/p\u003e\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eStudy population\u003c/h2\u003e\u003cp\u003eBetween January 2023 and January 2024, a total of 2,596 patients with viral pneumonia related symptoms underwent chest CT and were diagnosed \u0026ldquo;pneumonia\u0026rdquo; via CT in our hospital. Inclusion criteria were: (a) age of at least 18 years; (b) confirmed diagnosis of viral pneumonia, including COVID-19 and Influenza A, via positive polymerase chain reaction (PCR) test and admitted into hospital; (c) underwent chest CT at admission and after discharge respectively. Exclusion criteria consisted of (a) outpatients or no positive PCR test available; (b) critical patients with tracheal intubation; (c) no follow-up CT available / CT at admission\u0026gt;14 days from symptom onset / CT after discharge\u0026thinsp;\u0026le;\u0026thinsp;28 days or\u0026gt;3months; (d) history of interstitial lung disease, lung malignancy, lung surgery or other severe lung diseases; (e) obvious breath artifacts affecting the detectability of chest CT. A sample size of 51 patients with 102 chest CT were included and evaluated with routine HRCT and UHRCT based on deep learning technologies.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eCT protocol\u003c/h3\u003e\n\u003cp\u003eChest CT examinations were acquired at both admission and post-discharge for every patient, using four different multidetector CT systems: Dual-source Force CT scanner (Siemens, Erlangen, Germany) for 22 patients, iCT scanner (Philips, Amsterdam, Netherlands) for 14 patients, Optima CT520 scanner (GE Medical Systems, Milwaukee, WI) for 11 patients and Neuviz Prime scanner (Neusoft Medical Systems, Shenyang, China) for 4 patients. All scans were performed in the supine position with full-inspiration breath hold, without intravenous contrast agent. HRCT images were reconstructed with 0.6-1mm collimation, using high-resolution algorithm (B64 for Force CT, Sharp for iCT, Bone for Optima CT520 and lung 10 for Neuviz Prime), and observed on lung settings (window width, 1500 HU; level, \u0026minus;\u0026thinsp;700 HU).\u003c/p\u003e\n\u003ch3\u003eDL Reconstruction Structure\u003c/h3\u003e\n\u003cp\u003eThe construction of the RaDynCT model was inspired by the structure-preserving super-resolution (SPSR) model\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e, utilizing gradient guidance from relevant HRCT images to generate clinically relevant UHRCT-DL images that contain richer details and edge information. The generator of the proposed model structure consists of two branches: the SR branch and the gradient branch. The SR branch receives HRCT images as input, while the gradient branch takes the gradient maps of HRCT images as input to guide the SR main branch in learning the detailed features of the HRCT images. This structure facilitates the generation of UHRCT-DL images from HRCT, highlighting clearer structural features. Figure\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e illustrate the details of the proposed model structure.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\n\u003ch3\u003eImage Analysis\u003c/h3\u003e\n\u003cp\u003eTwo radiologists (G.Y. (reader 1, 26 years\u0026rsquo; experience) and N.L. (reader 2, 8 years' experience)) independently assessed the CT images, blinded to imaging reports and clinical data. CT findings were evaluated across seven categories according to the Fleischner Society\u0026rsquo;s glossary\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e, including ground-glass opacity (GGO), reticulation, tree-in-bud opacities (included both vascular and bronchiolar tree-in-bud opacities), consolidation, linear bands, bronchiectasis and bronchiolectasis by 4-point scale (0: absent; 1: probably present; 2: present but blurred; 3: present and clear). To assess the detectability of subtle findings with UHRCT-DL, comparison between the UHRCT-DL and HRCT images for the same patient were made for the presence of additional/different CT findings, defined as CT findings in lung areas where HRCT showed no or different CT findings (eg, reticulations on UHRCT-DL images classified as GGO according to HRCT images).\u003c/p\u003e\u003cp\u003eFollowing the same principle, subjective image quality differences were evaluated using a five-point Likert scale: 1)\u0026thinsp;\u0026minus;\u0026thinsp;2, definitely worse, with probable effect on detectability of CT findings; 2)\u0026thinsp;\u0026minus;\u0026thinsp;1, definitely worse, without effect on detectability of CT findings; 3) 0, similar, not definitely worse or better; 4)1, definitely better, without effect on detectability of CT findings; 5) 2, definitely better, with probable effect on detectability of CT findings. Analysis of CT findings and subjective image quality were obtained by a consensus between two chest radiologists, who were blinded to the scanner type and clinical conditions.\u003c/p\u003e\u003cp\u003eObjective image noise was quantified by placing a circular region of interest (ROIs) with a diameter of 10 mm within the tracheal lumen, just above the level of the aortic arch. The standard deviation of this measurement represented image noise. Signal-to-noise ratio (SNR) was then calculated by dividing the mean Hounsfield unit density of each ROI by its SD. This process was performed three times for each series to obtain mean lung SNRs as assessed by Reader 2.\u003c/p\u003e\n\u003ch3\u003eStatistical Analyses\u003c/h3\u003e\n\u003cp\u003eStatistical Analyses were performed using SPSS software, version 21 (IBM Corporation, Armonk, NY, USA). Continuous variables were tested for normal distribution using the Shapiro-Wilk test. Normally distributed variables were expressed as mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, and non-normally distributed variables were expressed as median and IQR. Categorical variables were expressed as absolute numbers and their percentages.\u003c/p\u003e\u003cp\u003eThe clarity of CT findings was assessed and compared between HRCT and UHRCT-DL images using the Wilcoxon signed-rank test. Differences in subjective image quality between UHRCT-DL images and corresponding HRCT reference images were evaluated for statistical significance using the Wilcoxon signed-rank test. For objective image quality comparisons between HRCT and UHRCT-DL, the paired Student\u0026rsquo;s t-test was applied when the data followed a normal distribution, otherwise the Wilcoxon signed-rank test was used. A two-sided \u003cem\u003eP\u003c/em\u003e value of less than 0.05 was considered indicative of a statistically significant difference.\u003c/p\u003e"},{"header":"Results","content":"\u003ch3\u003eCharacteristics of the study population\u003c/h3\u003e\n\u003cp\u003eThe final study sample consisted of 51 patients with viral pneumonia, selected after the exclusion of 2,545 participants (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). The study population included 33 males and 18 females with a mean age of 66.78 years. The median intervals from symptom onset to CT examination were 7 days at admission and 49 days post-discharge. On HRCT images, ground-glass opacity (GGO) and reticulation were the most common lung abnormalities observed at both admission and post-discharge. Findings of tree-in-bud opacities and consolidations were more frequent at admission, whereas linear bands, bronchiectasis and bronchiolectasis appeared more commonly in post-discharge CT images. Key characteristics of the study population are given in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eDemographic and characteristics\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"3\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCharacteristic\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAt Admission (N\u0026thinsp;=\u0026thinsp;51)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eAfter Discharge (N\u0026thinsp;=\u0026thinsp;51)\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eMale\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e33 (64.7%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eFemale\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e18 (35.3%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eMean age, y\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e66.78\u0026thinsp;\u0026plusmn;\u0026thinsp;7.38\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eTime from onset to CT, d\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e7 (4,10)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e49 (35,60)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eCT findings\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGGO\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e51 (100%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e51 (100%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eConsolidations\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e39 (76.47%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e23 (45.10%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eReticulations\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e45 (88.24%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e47 (92.16%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTIB\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e31 (60.78%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e5 (9.80%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLinear bands\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e6 (11.76%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e30 (58.82%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBronchiectasis\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e21 (41.18%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e37 (72.56%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBronchiolectasis\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e29 (56.86%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e44 (86.27%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"3\"\u003eNote: CT findings were evaluated on HRCT\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd colspan=\"3\"\u003eGGO: ground glass opacities; TIB: tree-in-bud opacities, including both bronchial and vascular TIB\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\n\u003ch3\u003eEvaluation of CT findings\u003c/h3\u003e\n\u003cp\u003eA total of 51 patients with 102 chest CT scans were evaluated using both routine HRCT and UHRCT-DL. As summarized in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, image clarity scores were significantly higher on UHRCT-DL for detecting ground-glass opacity (GGO), reticulation, tree-in-bud opacities, consolidations, linear bands, bronchiectasis, and bronchiolectasis at both admission and post-discharge.\u003c/p\u003e\u003cp\u003eNotably, 2 cases (2%) of reticulation and 2 cases (3%) of bronchiolectasis were assigned a score of 1 on HRCT, whereas no findings received a score of 1 on UHRCT-DL. Image ratings were enhanced from 2 to 3 on UHRCT-DL for 39 cases (38%) of GGO, 7 cases (11%) of consolidation, 40 cases (43%) of reticulation, 13 cases (36%) of tree-in-bud opacities, 5 cases (14%) of linear band s, 17 cases (29%) of bronchiectasis, and 30 cases (41%) of bronchiolectasis.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eEvaluation of CT findings in HRCT and UHRCT-DL\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"5\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCT findings\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eHRCT (N\u0026thinsp;=\u0026thinsp;102)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eUHRCT-DL (N\u0026thinsp;=\u0026thinsp;102)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eValue\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003eP\u003c/em\u003e value\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eGGO (102)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2 (2,3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3 (3,3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cem\u003eZ\u003c/em\u003e=-6.245\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.000\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScore 0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScore 1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScore 2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e45\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScore 3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e57\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e96\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eConsolidations (62)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e3 (0,3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3 (0,3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cem\u003eZ\u003c/em\u003e=-2.646\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.008\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScore 0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e40\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e40\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScore 1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScore 2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScore 3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e53\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e60\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eReticulations (92)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2 (2,3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3 (3,3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cem\u003eZ\u003c/em\u003e=-6.557\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.000\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScore 0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScore 1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScore 2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e44\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScore 3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e46\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e86\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eTIB (36)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0 (0,2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0 (0,3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cem\u003eZ\u003c/em\u003e=-3.606\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.000\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScore 0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e66\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e66\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScore 1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScore 2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e17\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScore 3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e19\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e32\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eLinear bands (36)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0 (0,3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0 (0,3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cem\u003eZ\u003c/em\u003e=-2.236\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.025\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScore 0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e66\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e66\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScore 1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScore 2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScore 3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e30\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e35\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eBronchiectasis (58)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2 (0,3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e2 (0,3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cem\u003eZ\u003c/em\u003e=-4.123\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.000\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScore 0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e44\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e44\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScore 1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScore 2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e27\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScore 3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e31\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e48\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eBronchiolectasis (73)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2 (0,2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3 (0,3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cem\u003eZ\u003c/em\u003e=-5.657\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.000\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScore 0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e29\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e29\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScore 1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScore 2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e46\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e18\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScore 3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e25\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e55\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"5\"\u003eGGO: ground glass opacities; TIB: tree-in-bud opacities\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eAdditional or different CT findings detected with UHRCT-DL\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"8\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHRCT\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e\u003cp\u003eUHRCT-DL At Admission (N\u0026thinsp;=\u0026thinsp;51)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e\u003cp\u003eUHRCT-DL After Discharge (N\u0026thinsp;=\u0026thinsp;51)\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eabnormalities\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAdditional\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eDifferent\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eTotal\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eAdditional\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eDifferent\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003eTotal\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eGGO\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e14 (27.45%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e51\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e15 (29.41%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e51\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eConsolidation\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e39\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e23\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eReticulations\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e45\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e47\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eTIB\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e8 (25.81%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e31\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1 (20.00%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e5\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eLinear bands\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e30\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eBronchiectasis\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e21\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e37\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eBronchiolectasis\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e6 (20.69%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e29\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e12 (27.27%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e44\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"8\"\u003eGGO: ground glass opacities; TIB: Tree-in-bud opacities\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eUHRCT-DL revealed additional/different CT findings in 23 patients (45%) and in 31 (30%) CT scans (15\u0026thinsp;+\u0026thinsp;16\u0026thinsp;=\u0026thinsp;31), compared to HRCT as the reference. In 29 CT scans (28%), areas identified as ground-glass opacities (GGO) on HRCT exhibited fine reticulations on UHRCT-DL, which were categorized as distinct CT findings of reticulations or crazy paving pattern when they overlapped with GGO\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e. The most frequently observed additional/different CT findings on UHRCT-DL at admission were crazy paving pattern (14/51, 27%) and tree-in-bud opacities (8/31, 26%), whereas reticulations (15/51, 29%) and bronchiolectasis (12/44, 27%) were most observed additional/different CT findings after discharge (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). No additional GGO, consolidations, reticulations, bronchiectasis or linear bands were found on UHRCT-DL. Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e showed detailed additional/different CT findings detected with UHRCT-DL.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eSubjective and objective image quality of UHRCT-DL in comparison with HRCT\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"5\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eImage Quality\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eUHRCT-DL (N\u0026thinsp;=\u0026thinsp;102)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eHRCT(N\u0026thinsp;=\u0026thinsp;102)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eValue\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003eP\u003c/em\u003e value\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSubjective\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNoise magnitude\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cem\u003eZ\u003c/em\u003e=-9.221\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.000\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e-2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0 (0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e-1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0 (0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0 (0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e73 (71.57%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e29 (28.43%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eStreak artifact\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cem\u003eZ\u003c/em\u003e=-6.856\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.000\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e-2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0 (0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e-1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0 (0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e55 (53.92%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e47 (46.08%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0 (0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eImage sharpness\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cem\u003eZ\u003c/em\u003e=-8.702\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.000\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e-2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0 (0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e-1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0 (0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e10 (9.80%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e62 (60.78%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e30 (29.41%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eObjective (HU)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMean\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e971.69\u0026thinsp;\u0026plusmn;\u0026thinsp;14.01\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e964.12\u0026thinsp;\u0026plusmn;\u0026thinsp;13.32\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cem\u003et\u003c/em\u003e\u0026thinsp;=\u0026thinsp;12.743\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.000\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSD\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e52.96\u0026thinsp;\u0026plusmn;\u0026thinsp;14.49\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e25.20\u0026thinsp;\u0026plusmn;\u0026thinsp;7.26\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cem\u003et\u003c/em\u003e\u0026thinsp;=\u0026thinsp;31.638\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.000\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSNR\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e19.96\u0026thinsp;\u0026plusmn;\u0026thinsp;6.46\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e41.35\u0026thinsp;\u0026plusmn;\u0026thinsp;11.49\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cem\u003et\u003c/em\u003e =-30.088\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.000\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e summarizes subjective and objective image quality of UHRCT-DL in comparison with HRCT. On UHRCT-DL, the subjective assessment of image quality was rated as equivalent (score 0) or superior (scores\u0026thinsp;+\u0026thinsp;1 and +\u0026thinsp;2) compared to HRCT, achieving the highest cumulative scores of \u0026ge;\u0026thinsp;+\u0026thinsp;1 for noise magnitude (102/102, 100%). For image sharpness and streak artifact, 90.20% (92/102) and 46.08% (47/102) UHRCT-DL images were evaluated as definitely better than HRCT images respectively. As to objective image quality, UHRCT-DL algorithm significantly lowered the mean level of image noise and improved SNR (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001), mean CT value was slightly increased (8/964HU, 0.78%) at the same time (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001).\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this retrospective study, we investigated the application of deep learning-based UHRCT to assess pulmonary abnormalities associated with viral pneumonia. A total of 102 CT scans from 51 patients admitted into hospital with viral pneumonia were analyzed, comparing images acquired at admission and following discharge with those obtained from HRCT. Our findings reveal that UHRCT-DL demonstrates superior image quality and clarity for identifying various parenchymal and interstitial abnormalities compared to HRCT. The additional/different UHRCT-DL findings of crazy paving pattern and tree-in-bud opacities at admission indicated the inflammatory changes of interstitium, bronchioles and micro vessels during acute phase, whereas reticulations and bronchiolectasis after discharge characterized fibrosis changes during recovery\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. Thus, the improvement of image quality by UHRCT-DL may better reflect pathological change and benefit the clinical performance of viral pneumonia.\u003c/p\u003e\u003cp\u003eIn recent years, UHRCT has been explored using advanced scanning techniques \u003csup\u003e\u003cspan additionalcitationids=\"CR20 CR21\" citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e. For example, Prayer F, et al found that photon-counting detector (PCD) CT revealed subtle lung abnormalities in symptomatic participants with persistent post- viral pneumonia symptoms\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. Yanagawa M, et al reported that super-high-resolution mode with 0.25-mm section thickness and a 2048 matrix have high diagnostic performance to predict invasiveness of lung adenocarcinoma\u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e. To some extent, both methods are not yet extensively facilitated in most hospitals and may involve increased radiation doses. Deep learning-based reconstruction techniques offer an alternative by enhancing image quality through post-processing without additional radiation exposure or specialized equipment. \u003csup\u003e\u003cspan additionalcitationids=\"CR25\" citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eSeveral studies have highlighted the potential of deep learning in improving the quality of CT imaging, especially in lung examination. Hata et al. proposed a method combining deep learning\u0026ndash;based denoising (DLD) and iterative reconstruction (IR) to improve the image quality of lung CT imaging, which utilize the ultra-low-dose CT to reconstruct the standard-does CT\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e. It can effectively reduce noise in low-dose CT images and improve spatial resolution. Nagayama et al. proposed a deep-learning-based reconstruction (SR-DLR) method to evaluate the image quality of coronary CT angiography (CCTA), which can effectively improve image clarity and noise characteristics while reducing halo artifacts caused by calcified plaques\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e. However, these approaches did not explore the comparative benefits of UHRCT over HRCT in clinical diagnosis. In contrast, our study focuses specifically on the advantages of UHRCT-DL for detecting pulmonary abnormalities in viral pneumonia, providing evidence of its clinical utility.\u003c/p\u003e\u003cp\u003eOur results find that UHRCT-DL achieved consistently higher subjective image quality scores, with all scans rated either as equivalent (score 0) or superior (scores\u0026thinsp;+\u0026thinsp;1 and +\u0026thinsp;2) to HRCT. Regarding objective image quality, the UHRCT-DL algorithm notably reduced mean image noise and improved the SNR, both with high statistical significance (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Additionally, UHRCT-DL demonstrated superior clarity for subtle CT findings, such as fine reticulations, tree-in-bud opacities, and bronchiolectasis, which were less distinguishable on HRCT. Additionally, in 28% CT scans, the detail of GGO were shown on UHRCT-DL as crazy paving pattern or reticulations which reflects interstitial changes of viral pneumonia not shown on routine HRCT. Furthermore, 2 cases (2%) of reticulation and 2 cases (3%) of bronchiolectasis were rated as \u0026ldquo;probably present\u0026rdquo; (score 1) on HRCT, suggesting that the elimination of ambiguous findings (score 1) on UHRCT-DL may enhance diagnostic confidence for clinicians.\u003c/p\u003e\u003cp\u003eAt admission, the most common additional/different CT findings on UHRCT-DL included crazy paving patterns and tree-in-bud opacities. The crazy paving pattern, characterized by focal area of GGO with septal thickening and intralobular reticular lines, indicate alveolar septal inflammation and fibroblast proliferation in the early stage of viral pneumonia. The different CT findings of crazy paving patterns on UHRCT-DL reflects the inflammation of both parenchyma and interstitium\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. The tree-in-bud pattern is conventionally used to describe infection and inflammation of the bronchioles. Viral pneumonia may affect both respiratory and circulatory system\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e,\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e. Patel et al first introduced the vascular tree-in-bud pattern in severe COVID-19 as a characteristic pattern\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. Interestingly, this CT finding was found to be correlated with longer ventilation and hospitalization, highlighting its prognostic role\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. After patients\u0026rsquo; discharge, reticulations and bronchiolectasis were most observed additional/different CT findings. In areas that had been classified as GGO at HRCT, UHRCT-DL showed fine reticulations, which are typical manifestation of interstitial lung abnormalities in 15 cases. Traction bronchiolectasis, defined as dilatation of bronchioles within the central portions of the secondary pulmonary lobule, is typically associated with fibrosis during recovery. Therefore, the distinctive CT findings identified by deep learning-based UHRCT may more accurately reflect underlying pathological changes, enhancing clinical insight and management of viral pneumonia.\u003c/p\u003e\u003cp\u003eOur study has several limitations. First, the inclusion of only hospitalized patients with complete medical records limits the generalizability of findings to individuals with mild or long-term pulmonary abnormalities. Second, the single-center, retrospective design and the relatively small sample size underscore the need for larger, multicenter studies to confirm our findings across a broader patient population. Finally, images were acquired from various CT scanners using different reconstruction algorithms at conventional radiation doses. Future studies exploring UHRCT-DL with low-dose protocols may provide further insights into optimizing image quality while minimizing radiation exposure.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eDeep learning-based UHRCT significantly enhanced image quality, offering a more precise depiction of viral pneumonia related CT features. The unique findings observed on UHRCT-DL may more accurately reflect inflammatory changes during the acute phase and early fibrotic changes during recovery. These advancements could provide a methodological reference for future research on CT image reconstruction in viral pneumonia and other diffuse lung diseases.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003cp\u003e This study was performed in accordance with the ethical principles of Declaration of Helsinki.\u003c/p\u003e\u003c/p\u003e\u003cp\u003e\u003ch2\u003eConsent for publication\u003c/h2\u003e\u003cp\u003eNot applicable.\u003c/p\u003e\u003c/p\u003e\u003cp\u003e\u003ch2\u003eCompeting Interests\u003c/h2\u003e\u003cp\u003eAuthor B.P. and L.X. are stockholder of RadioDynamic Medical\u003c/p\u003e\u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e\u003cp\u003eThis study has received funding by Beijing Hospitals Authority Clinical Medicine Development of Special Funding Support (ZLRK202306), Natural Science Foundation of Xiamen, China (3502Z202373090) and National Natural Science Foundation of China (82000103).\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eYanli Gao: Writing \u0026ndash; original draft, Writing \u0026ndash; review \u0026amp; editing, Visualization, Validation, Resources, Project administration, Methodology, Investigation, Formal analysis, Conceptualization.Boyang Pan: Data curation, Writing \u0026ndash; review \u0026amp; editing, ValidationLin Niu: Formal analysis, InvestigationLibo Xu: Data curationZiheng Guo: Data curationWeican Liu: Data curationPenghui Sun: ResourcesYanyan Zhang: ResourcesXiaoli Xu: Funding acquisitionNanjie Gong: Methodology, Software, Funding acquisitionQi Yang: Supervision, Project administration, Conceptualization, Funding acquisition\u003c/p\u003e\u003ch2\u003eAcknowledgements\u003c/h2\u003e\u003cp\u003eNot applicable.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eBrixey AG, Oh AS, Alsamarraie A, Chung JH. Pictorial Review of Fibrotic Interstitial Lung Disease on High-Resolution CT Scan and Updated Classification. Chest 2024;165(4):908-923. DOI: 10.1016/j.chest.2023.11.037.\u003c/li\u003e\n\u003cli\u003eMarton N, Gyebnar J, Fritsch K, et al. Photon-counting computed tomography in the assessment of rheumatoid arthritis-associated interstitial lung disease: an initial experience. Diagn Interv Radiol 2023;29(2):291-299. DOI: 10.4274/dir.2023.221959.\u003c/li\u003e\n\u003cli\u003eOhno Y, Akino N, Fujisawa Y, et al. Comparison of lung CT number and airway dimension evaluation capabilities of ultra-high-resolution CT, using different scan modes and reconstruction methods including deep learning reconstruction, with those of multi-detector CT in a QIBA phantom study. Eur Radiol 2023;33(1):368-379. DOI: 10.1007/s00330-022-08983-1.\u003c/li\u003e\n\u003cli\u003eTatsugami F, Higaki T, Kawashita I, et al. Improvement of Spatial Resolution on Coronary CT Angiography by Using Super-Resolution Deep Learning Reconstruction. Acad Radiol 2023;30(11):2497-2504. DOI: 10.1016/j.acra.2022.12.044.\u003c/li\u003e\n\u003cli\u003eNagayama Y, Iwashita K, Maruyama N, et al. Deep learning-based reconstruction can improve the image quality of low radiation dose head CT. Eur Radiol 2023;33(5):3253-3265. DOI: 10.1007/s00330-023-09559-3.\u003c/li\u003e\n\u003cli\u003eSun Y, Pan B, Li Q, et al. Clinical ultra-high resolution CT scans enabled by using a generative adversarial network. Med Phys 2023;50(6):3612-3622. DOI: 10.1002/mp.16172.\u003c/li\u003e\n\u003cli\u003eGaillandre Y, Duhamel A, Flohr T, et al. Ultra-high resolution CT imaging of interstitial lung disease: impact of photon-counting CT in 112 patients. Eur Radiol 2023;33(8):5528-5539. DOI: 10.1007/s00330-023-09616-x.\u003c/li\u003e\n\u003cli\u003ePrayer F, Kienast P, Strassl A, et al. Detection of Post-COVID-19 Lung Abnormalities: Photon-counting CT versus Same-Day Energy-integrating Detector CT. Radiology 2023;307(1):e222087. DOI: 10.1148/radiol.222087.\u003c/li\u003e\n\u003cli\u003eKoo HJ, Lim S, Choe J, Choi S-H, Sung H, Do K-H. Radiographic and CT Features of Viral Pneumonia. RadioGraphics 2018;38(3):719-739. DOI: 10.1148/rg.2018170048.\u003c/li\u003e\n\u003cli\u003eKoo HJ, Choi S-H, Sung H, Choe J, Do K-H. RadioGraphics Update: Radiographic and CT Features of Viral Pneumonia. RadioGraphics 2020;40(4):E8-E15. DOI: 10.1148/rg.2020200097.\u003c/li\u003e\n\u003cli\u003ePan F, Ye T, Sun P, et al. Time Course of Lung Changes at Chest CT during Recovery from Coronavirus Disease 2019 (COVID-19). Radiology 2020;295(3):715-721. DOI: 10.1148/radiol.2020200370.\u003c/li\u003e\n\u003cli\u003eLee JH, Koh J, Jeon YK, Goo JM, Yoon SH. An Integrated Radiologic-Pathologic Understanding of COVID-19 Pneumonia. Radiology 2023;306(2):e222600. DOI: 10.1148/radiol.222600.\u003c/li\u003e\n\u003cli\u003eSandino CM, Cheng JY, Chen F, Mardani M, Pauly JM, Vasanawala SS. Compressed Sensing: From Research to Clinical Practice with Deep Neural Networks. IEEE Signal Process Mag 2020;37(1):111-127. DOI: 10.1109/MSP.2019.2950433.\u003c/li\u003e\n\u003cli\u003eWei Lu ZS, Jinghui Chu. A novel 3D medical image super-resolution method based on densely connected network. Biomedical Signal Processing and Control 2020;62: 102120.\u003c/li\u003e\n\u003cli\u003eD. Qiu YCaXW. Residual Dense Attention Networks for COVID-19 Computed Tomography Images Super Resolution. IEEE Transactions on Cognitive and Developmental Systems 2023;15(2):904-913.\u003c/li\u003e\n\u003cli\u003eMa C, Rao Y, Lu J, Zhou J. Structure-Preserving Image Super-Resolution. IEEE Trans Pattern Anal Mach Intell 2022;44(11):7898-7911. DOI: 10.1109/TPAMI.2021.3114428.\u003c/li\u003e\n\u003cli\u003eBankier AA, MacMahon H, Colby T, et al. Fleischner Society: Glossary of Terms for Thoracic Imaging. Radiology 2024;310(2):e232558. DOI: 10.1148/radiol.232558.\u003c/li\u003e\n\u003cli\u003ePatel BV, Arachchillage DJ, Ridge CA, et al. Pulmonary Angiopathy in Severe COVID-19: Physiologic, Imaging, and Hematologic Observations. Am J Respir Crit Care Med 2020;202(5):690-699. DOI: 10.1164/rccm.202004-1412OC.\u003c/li\u003e\n\u003cli\u003eInoue A, Johnson TF, Walkoff LA, et al. Lung Cancer Screening Using Clinical Photon-Counting Detector Computed Tomography and Energy-Integrating-Detector Computed Tomography: A Prospective Patient Study. J Comput Assist Tomogr 2023;47(2):229-235. DOI: 10.1097/RCT.0000000000001419.\u003c/li\u003e\n\u003cli\u003eDunning CAS, Marsh JF, Jr., Winfree T, et al. Accuracy of Nodule Volume and Airway Wall Thickness Measurement Using Low-Dose Chest CT on a Photon-Counting Detector CT Scanner. Invest Radiol 2023;58(4):283-292. DOI: 10.1097/RLI.0000000000000933.\u003c/li\u003e\n\u003cli\u003eChamberlin JH, Smith CD, Maisuria D, et al. Ultra-high-resolution photon-counting detector computed tomography of the lungs: Phantom and clinical assessment of radiation dose and image quality. Clin Imaging 2023;104:110008. DOI: 10.1016/j.clinimag.2023.110008.\u003c/li\u003e\n\u003cli\u003eHata A, Yanagawa M, Ninomiya K, et al. Photon-Counting Detector CT Radiological-Histological Correlation in Cadaveric Human Lung Nodules and Airways. Invest Radiol 2025;60(2):151-160. DOI: 10.1097/RLI.0000000000001117.\u003c/li\u003e\n\u003cli\u003eYanagawa M, Tsubamoto M, Satoh Y, et al. Lung Adenocarcinoma at CT with 0.25-mm Section Thickness and a 2048 Matrix: High-Spatial-Resolution Imaging for Predicting Invasiveness. Radiology 2020;297(2):462-471. DOI: 10.1148/radiol.2020201911.\u003c/li\u003e\n\u003cli\u003eSakai Y, Hida T, Matsuura Y, et al. Impact of a new deep-learning-based reconstruction algorithm on image quality in ultra-high-resolution CT: clinical observational and phantom studies. Br J Radiol 2023;96(1141):20220731. DOI: 10.1259/bjr.20220731.\u003c/li\u003e\n\u003cli\u003eNakamura Y, Narita K, Higaki T, Akagi M, Honda Y, Awai K. Diagnostic value of deep learning reconstruction for radiation dose reduction at abdominal ultra-high-resolution CT. Eur Radiol 2021;31(7):4700-4709. DOI: 10.1007/s00330-020-07566-2.\u003c/li\u003e\n\u003cli\u003eWiebe M, Haston C, Lamey M, Narayan A, Rajapakshe R. The effect of spatial resolution on deep learning classification of lung cancer histopathology. BJR Open 2023;5(1):20230008. DOI: 10.1259/bjro.20230008.\u003c/li\u003e\n\u003cli\u003eHata A, Yanagawa M, Yoshida Y, et al. Combination of Deep Learning-Based Denoising and Iterative Reconstruction for Ultra-Low-Dose CT of the Chest: Image Quality and Lung-RADS Evaluation. AJR Am J Roentgenol 2020;215(6):1321-1328. DOI: 10.2214/AJR.19.22680.\u003c/li\u003e\n\u003cli\u003eNagayama Y, Emoto T, Kato Y, et al. Improving image quality with super-resolution deep-learning-based reconstruction in coronary CT angiography. Eur Radiol 2023;33(12):8488-8500. DOI: 10.1007/s00330-023-09888-3.\u003c/li\u003e\n\u003cli\u003eChen X, Long Z, Lei Y, et al. CT Differentiation and Prognostic Modeling in COVID-19 and Influenza A Pneumonia. Acad Radiol. 2025 Jul;32(7):4289-4300. doi: 10.1016/j.acra.2025.02.004. \u003c/li\u003e\n\u003cli\u003eJin L, Chen J, Wu L, et al. Central artery pulse pressure, not central arterial stiffness impact on all-cause mortality in patients with viral pneumonia infection. BMC Infectious Diseases 2024;24(1). DOI: 10.1186/s12879-024-10091-y.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-medical-imaging","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bmim","sideBox":"Learn more about [BMC Medical Imaging](http://bmcmedimaging.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bmim/default.aspx","title":"BMC Medical Imaging","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Multidetector Computed Tomography, Deep Learning, Pneumonia, Viral","lastPublishedDoi":"10.21203/rs.3.rs-7446633/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7446633/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e\u003cp\u003eAdvancement in deep learning has introduced significant potential for enhancing CT image quality without increasing patient radiation exposure. In this study, we sought to compare deep learning‑based ultra-high-resolution CT (UHRCT-DL) findings of viral pneumonia at admission and after discharge with that of HRCT images.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e\u003cp\u003eA total of 51 inpatients (mean age 66.78 years; 33 males) of viral pneumonia underwent 102 CT scans at admission and after discharge. A deep learning-based super-resolution model, incorporating a dual-branch architecture for super-resolution and gradient guidance, was used to generate UHRCT-DL. UHRCT-DL and HRCT images were systematically reviewed for viral pneumonia CT findings, including ground-glass opacity (GGO), reticulation, tree-in-bud opacities, consolidation, linear bands, bronchiectasis, and bronchiectasis. Subjective and objective CT image quality was evaluated using a five-point Likert scale (\u0026minus;\u0026thinsp;2 to 2) and lung signal-to-noise ratios (SNRs).\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e\u003cp\u003eThe score of clarity of CT findings was significantly higher on UHRCT-DL for all CT findings at admission and after discharge. Compared with HRCT as reference image, the most frequently observed additional/different CT findings on UHRCT-DL at admission were crazy paving pattern (14/51, 27%) and tree-in-bud opacities (8/31, 26%), whereas reticulations (15/51, 29%) and bronchiolectasis (12/44, 27%) were most observed additional/different CT findings after discharge. The subjective and objective image quality of UHRCT-DL was superior to that of HRCT. UHRCT-DL algorithm significantly lowered the level of image noise and improved SNR (19.96\u0026thinsp;\u0026plusmn;\u0026thinsp;6.46 vs 41.35\u0026thinsp;\u0026plusmn;\u0026thinsp;11.49, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001).\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e\u003cp\u003eThe deep learning‑based UHRCT provided a more precise depiction of CT features of viral pneumonia, that better reflects the inflammatory changes during acute phase and early fibrotic changes during recovery.\u003c/p\u003e","manuscriptTitle":"Deep Learning‑based Ultra-high-resolution CT imaging of Viral Pneumonia at Admission and after Discharge","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-08 06:46:48","doi":"10.21203/rs.3.rs-7446633/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-03-06T05:18:46+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-05T11:23:30+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"188397687617138216467079688554418543975","date":"2026-02-17T08:01:18+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-11-08T16:26:16+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"177188850916400584264546225478630438027","date":"2025-11-02T11:56:16+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-09-23T03:55:50+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-09-18T01:40:10+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-09-08T10:29:58+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-09-06T02:19:26+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Medical Imaging","date":"2025-09-06T02:16:30+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-medical-imaging","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bmim","sideBox":"Learn more about [BMC Medical Imaging](http://bmcmedimaging.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bmim/default.aspx","title":"BMC Medical Imaging","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"9897ae30-9030-4ea0-96db-8ba01700abc7","owner":[],"postedDate":"October 8th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-04-07T16:01:09+00:00","versionOfRecord":{"articleIdentity":"rs-7446633","link":"https://doi.org/10.1186/s12880-026-02320-4","journal":{"identity":"bmc-medical-imaging","isVorOnly":false,"title":"BMC Medical Imaging"},"publishedOn":"2026-04-04 15:57:56","publishedOnDateReadable":"April 4th, 2026"},"versionCreatedAt":"2025-10-08 06:46:48","video":"","vorDoi":"10.1186/s12880-026-02320-4","vorDoiUrl":"https://doi.org/10.1186/s12880-026-02320-4","workflowStages":[]},"version":"v1","identity":"rs-7446633","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7446633","identity":"rs-7446633","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}