Project description:Computed tomography (CT) is the preferred imaging method for diagnosing 2019 novel coronavirus (COVID19) pneumonia. We aimed to construct a system based on deep learning for detecting COVID-19 pneumonia on high resolution CT. For model development and validation, 46,096 anonymous images from 106 admitted patients, including 51 patients of laboratory confirmed COVID-19 pneumonia and 55 control patients of other diseases in Renmin Hospital of Wuhan University were retrospectively collected. Twenty-seven prospective consecutive patients in Renmin Hospital of Wuhan University were collected to evaluate the efficiency of radiologists against 2019-CoV pneumonia with that of the model. An external test was conducted in Qianjiang Central Hospital to estimate the system's robustness. The model achieved a per-patient accuracy of 95.24% and a per-image accuracy of 98.85% in internal retrospective dataset. For 27 internal prospective patients, the system achieved a comparable performance to that of expert radiologist. In external dataset, it achieved an accuracy of 96%. With the assistance of the model, the reading time of radiologists was greatly decreased by 65%. The deep learning model showed a comparable performance with expert radiologist, and greatly improved the efficiency of radiologists in clinical practice.
Project description:Presently, coronavirus disease-19 (COVID-19) is spreading worldwide without an effective treatment method. For COVID-19, which is often asymptomatic, it is essential to adopt a method that does not cause aggravation, as well as a method to prevent infection. Whether aggravation can be predicted by analyzing the extent of lung damage on chest computed tomography (CT) scans was examined. The extent of lung damage on pre-intubation chest CT scans of 277 patients with COVID-19 was assessed. It was observed that aggravation occurred when the CT scan showed extensive damage associated with ground-glass opacification and/or consolidation (p < 0.0001). The extent of lung damage was similar across the upper, middle, and lower fields. Furthermore, upon comparing the extent of lung damage based on the number of days after onset, a significant difference was found between the severe pneumonia group (SPG) with intubation or those who died and non-severe pneumonia group (NSPG) ≥3 days after onset, with aggravation observed when ≥14.5% of the lungs exhibited damage at 3–5 days (sensitivity: 88.2%, specificity: 72.4%) and when ≥20.1% of the lungs exhibited damage at 6–8 days (sensitivity: 88.2%, specificity: 69.4%). Patients with aggravation suddenly developed hypoxemia after 7 days from the onset; however, chest CT scans obtained in the paucisymptomatic phase without hypoxemia indicated that subsequent aggravation could be predicted based on the degree of lung damage. Furthermore, in subjects aged ≥65 years, a significant difference between the SPG and NSPG was observed in the extent of lung damage early beginning from 3 days after onset, and it was found that the degree of lung damage could serve as a predictor of aggravation. Therefore, to predict and improve prognosis through rapid and appropriate management, evaluating patients with factors indicating poor prognosis using chest CT is essential.
Project description:On 31 December 2019, a severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) emerged in Wuhan, Hubei province, China, and caused the outbreak of the Coronavirus Disease 2019 (COVID-19). To date, computed tomography (CT) findings have been recommended as major evidence for the clinical diagnosis of COVID-19 in Hubei, China. This review focuses on the imaging characteristics and changes throughout the disease course in patients with COVID-19 in order to provide some help for clinicians. Typical CT findings included bilateral ground-glass opacity, pulmonary consolidation, and prominent distribution in the posterior and peripheral parts of the lungs. This review also provides a comparison between COVID-19 and other diseases that have similar CT findings. Since most patients with COVID-19 infection share typical imaging features, radiological examinations have an irreplaceable role in screening, diagnosis and monitoring treatment effects in clinical practice.
Project description:ObjectiveThe aim of our study was to assess the effect of a short-term treatment with low-moderate corticosteroid (CS) doses by both a quantitative and qualitative assessment of chest HRCT of COVID-19 pneumonia.MethodsCORTICOVID is a single-center, cross-sectional, retrospective study involving severe/critical COVID-19 patients with mild/moderate ARDS. Lung total severity score was obtained according to Chung and colleagues. Moreover, the relative percentages of lung total severity score by ground glass opacities, consolidations, crazy paving, and linear bands were computed. Chest HRCT scores, P/F ratio, and laboratory parameters were evaluated before (pre-CS) and 7-10 days after (post-CS) methylprednisolone of 0.5-0.8 mg/kg/day.FindingsA total of 34 severe/critical COVID-19 patients were included in the study, of which 17 received Standard of Care (SoC) and 17 CS therapy in add-on. CS treatment disclosed a significant decrease in HRCT total severity score [median = 6 (IQR: 5-7.5) versus 10 (IQR: 9-13) in SoC, p < 0.001], as well in single consolidations [median = 0.33 (IQR: 0-0.92) versus 6.73 (IQR: 2.49-8.03) in SoC, p < 0.001] and crazy paving scores [mean = 0.19 (SD = 0.53) versus 1.79 (SD = 2.71) in SoC, p = 0.010], along with a significant increase in linear bands [mean = 2.56 (SD = 1.65) versus 0.97 (SD = 1.30) in SoC, p = 0.006]. GGO score instead did not significantly differ at the end of treatment between the two groups. Most post-CS GGO, however, derived from previous consolidations and crazy paving [median = 1.5 (0.35-3.81) versus 2 (1.25-3.8) pre-CS; p = 0.579], while pre-CS GGO significantly decreased after methylprednisolone therapy [median = 0.66 (0.05-1.33) versus 1.5 (0.35-3.81) pre-CS; p = 0.004]. CS therapy further determined a significant improvement in P/F levels [median P/F = 310 (IQR: 235.5-370) versus 136 (IQR: 98.5-211.75) in SoC; p < 0.001], and a significant increase in white blood cells, lymphocytes, and neutrophils absolute values.ConclusionThe improvement of all chest HRCT findings further supports the role of CS adjunctive therapy in severe/critical COVID-19 pneumonia.
Project description:BackgroundAccurately differentiating pneumocystis from cytomegalovirus pneumonia is crucial for correct therapy selection in AIDS patients. Hence, the goal of this study was to compare the computerized tomography (CT) features of pneumocystis pneumonia and cytomegalovirus pneumonia in AIDS patients and identify clinical hallmarks to accurately distinguish these two pathologies.MethodsA total of 112 AIDS patients (78 with pneumocystis pneumonia and 34 cytomegalovirus pneumonia) at Beijing Ditan Hospital from January 2017 to May 2019 were included in this study. Two experienced chest radiologists retrospectively reviewed CT images for 17 features including ground-glass opacity, consolidation, nodules, and halo sign. Binary logistic regression analyses were conducted to identify the significant parameters that distinguished pneumocystis pneumonia from cytomegalovirus pneumonia. Correlations were analyzed by Pearson or Spearman correlation analyses. Result were considered significant if P < 0.05.ResultsThe presence of consolidation, halo signs, and nodules (all P < 0.05) were significantly more frequent in patients with cytomegalovirus pneumonia than in those with pneumocystis pneumonia. Small nodules (32.5% in cytomegalovirus pneumonia, 6.41% in pneumocystis pneumonia, P < 0.001) without perilymphatic distribution were particularly common in patients with cytomegalovirus pneumonia. Large nodules were not found in any of patients with cytomegalovirus pneumonia. The presence of ground-glass opacity, reticulation, and bronchial wall thickening (all P > 0.05) were common in both groups.ConclusionsAnalysis of consolidation, nodules, and halo signs may contribute to the differential diagnosis of pneumocystis pneumonia or cytomegalovirus pneumonia. However, some CT features considered typical in one or other diseases appear with similar frequency in both cohorts of AIDS patients. CT features are potentially useful for the differential diagnosis of pneumocystis pneumonia and cytomegalovirus pneumonia in AIDS patients.
Project description:PurposeThe purpose of our study was to determine the usability of lung ultrasonography (LUS) in the diagnosis of COVID-19, and to match the morphological features of lesions detected on computed tomography (CT) with the findings observed on LUS.MethodsSixty patients with COVID-19 were included in this prospective study. Patients were examined by radiology and anesthesia clinic specialists for a visual CT score. A LUS 12-zone ultrasonography protocol was applied by the investigator blinded to the CT and PCR test results. The characteristics of abnormal findings and the relationship of lesions to the pleura and the distance to the pleura were investigated.ResultsForty-five males and 25 females evaluated within the scope of the study had an average age of 61.2 ± 15.3 years. The total CT score was calculated as 14.3 ± 5.3, and the LUS score was found to be 19.9 ± 7.6. There was a statistically significant positive correlation between the measured LUS and CT scores (r = 0.857, p < 0.001). The mean distance of these lesions to the pleura was 5.2 ± 1.76 cm. LUS findings in 51 areas corresponded to non-pleural lesions on CT. There was a negative correlation between the measured distance to the pleura and the LUS scores (p < 0.001, r = - 0.708).ConclusionThe results of this study showed that the correlation between CT and LUS findings may be used in the diagnosis of COVID-19 pneumonia, although there are some limitations. ClinicalTrials.gov identifier: NCT04719234.
Project description:BackgroundClinical manifestation and neonatal outcomes of pregnant women with coronavirus disease 2019 (COVID-19) were unclear in Wuhan, China.MethodsWe retrospectively analyzed clinical characteristics of pregnant and nonpregnant women with COVID-19 aged from 20 to 40, admitted between January 15 and March 15, 2020 at Union Hospital, Wuhan, and symptoms of pregnant women with COVID-19 and compared the clinical characteristics and symptoms to historic data previously reported for H1N1.ResultsAmong 64 patients, 34 (53.13%) were pregnant, with higher proportion of exposure history (29.41% vs 6.67%) and more pulmonary infiltration on computed tomography test (50% vs 10%) compared to nonpregnant women. Of pregnant patients, 27 (79.41%) completed pregnancy, 5 (14.71%) had natural delivery, 18 (52.94%) had cesarean section, and 4 (11.76%) had abortion; 5 (14.71%) patients were asymptomatic. All 23 newborns had negative reverse-transcription polymerase chain results, and an average 1-minute Apgar score was 8-9 points. Pregnant and nonpregnant patients show differences in symptoms such as fever, expectoration, and fatigue and on laboratory tests such as neurophils, fibrinogen, D-dimer, and erythrocyte sedimentation rate. Pregnant patients with COVID-19 tend to have more milder symptoms than those with H1N1.ConclusionsClinical characteristics of pregnant patients with COVID-19 are less serious than nonpregnant. No evidence indicated that pregnant women may have fetal infection through vertical transmission of COVID-19. Pregnant patients with H1N1 had more serious condition than those with COVID-19.