Objective To investigate the value of computer-assisted quantification of pulmonary embolism volume(PEV)in identifying mild-to-high-risk acute pulmonary embolism(APE).Methods We retrospectively enrolled 143 patients with suspected APE confirmed by computed tomography pulmonary angiography(CTPA)at Yan'an University Affiliated Hospital from January 2017 to December 2020.According to the 2018 Chinese Guidelines for Diagnosis,Treatment and Prevention of Pulmonary Thromboembolism,all the patients were divided into low-risk group(n=88)and mild-to-high-risk group(n=55).We collected the patients'basic demographic data,clinical manifestations,and serum levels of N-terminal-B type natriuretic peptide precursor(NT-proBNP)and D-dimer.Based on CTPA images,the degree of pulmonary thromboembolism was artificially evaluated to obtain the pulmonary artery occlusion index(PAOI).The thrombus was segmented using the pulmonary embolism detection tool based on digital lung,and PEV was calculated.We compared the differences in clinical and laboratory indicators and PAOI and PEV between the two risk groups.We analyzed the value of PAOI and PEV in identifying mild-to-high-risk APE using receiver operating characteristic(ROC)curves,and used Logistic regression analysis to identify independent risk factors in predicting mild-to-high-risk APE.Different models were established.Results Compared with the low-risk group,APE patients in the mild-to-high-risk group were older(P＜0.05),had lower diastolic blood pressure(P＜0.05),higher levels of D-dimer and NT-proBNP(P＜0.05),lower levels of platelet count,arterial oxygen partial pressure and arterial carbon dioxide partial pressure(P＜0.05),and higher levels of PAOI and PEV(P＜0.001).ROC curve analysis showed that the area under the curve for PEV in identifying mild-to-high-risk APE was 0.809(95％CI:0.734-0.884),while that for PAOI was 0.753(95％CI:0.667-0.839).Logistic regression analysis showed that PEV and NT-proBNP were independent risk factors for mild-to-high-risk APE(P＜0.05).Conclusion PEV and NT-proBNP are independent risk factors for mild-to-high-risk APE.

【Objective】 To compare the value of CT pulmonary angiography (CTPA) and echocardiography in predicting the degree pulmonary hypertension (PH). 【Methods】 Fifty-four patients in our hospital who underwent right heart catheterization, CTPA and echocardiography for suspected or confirmed PH from November 2013 to April 2021 were retrospectively recruited. Pulmonary artery systolic pressure (PASP) and mean pulmonary artery pressure (PAMP) were measured by right heart catheterization. According to PAMP, the patients were divided into non-PH group and mild, moderate, and severe PH groups. The three-dimensional model of the pulmonary artery was obtained by computer-aided segmentation, and the best fitting circle diameter (Dfit), inscribed circle diameter (Dmin), circumscribed circle diameter (Dmax), hydraulic diameter (Dh), cross-sectional area (Area), circumference (Scf), and the largest area and largest short diameter of the right ventricle were measured. Echocardiography was used to estimate PASP. The differences in the above parameters between different PH groups were compared, the correlations of the above parameters with PASP and PAMP were analyzed, and statistically significant indicators were included to establish three predictive models of PAMP (Model 1: CTPA pulmonary artery; Model 2: CTPA pulmonary artery+CTPA right ventricle; Model 3: CTPA pulmonary artery+CTPA right ventricle+echocardiography), and receiver operating characteristic (ROC) curves were used to compare the diagnostic performance of the three models. 【Results】 Some parameters of main pulmonary artery (Dfit, Dmin, Dmax, Dh, Area, Scf), right pulmonary artery (Dmax, Dh, Area, Scf), left pulmonary artery (Dfit), and right ventricular short diameter increased with the increase of PAMP (P<0.05). Dfit, Dmin, Dh, and area of main pulmonary artery, right pulmonary artery and left pulmonary artery were positively correlated with PASP and PAMP (P<0.05). Right ventricular short diameter and right ventricular maximum area were positively correlated with PASP (P<0.05), and right ventricular short diameter was positively correlated with PAMP (P<0.05). The estimated value of pulmonary artery systolic blood pressure in echocardiography was positively correlated with PASP and PAMP (P<0.05). Model 1, Model 2 and Model 3 could all be used to identify mild PH and moderate PH, among which Model 3 had the best performance in identifying non-PH and mild PH, moderate PH and severe PH. 【Conclusion】 CTPA and echocardiography are helpful in assessing the degree of PH, and the combination of the two has better accuracy in distinguishing non-PH from mild PH, moderate and severe PH.

Objective:To investigate the feasibility of classifying imaging patterns of dermatomyositis/polymyositis-related interstitial lung disease (DM/PM-ILD) into subtypes based on chest CT radiomics features and a model was constructed by machine learning algorithms.Methods:From November 2011 to November 2020, 107 patients diagnosed with PM/DM-ILD at the First Affiliated Hospital of Xi′an Jiaotong University were retrospectively analyzed. A total of 315 cases with chest CT were collected. Doctors pre-classified image patterns, including 105 cases with non-specific interstitial pneumonia (NSIP), 90 cases with organizing pneumonia (OP), and 66 cases with non-specific interstitial pneumonia combined with organizing pneumonia (NSIP+OP), 35 cases with common interstitial pneumonia (UIP), and 19 cases with diffuse alveolar damage (DAD), ANOVA was used to test the difference of baseline clinical information among the imaging classification groups. All images were divided into the training set and the est set by stratified random sampling at a ratio of 4∶1. In each CT scan, 3D slicer was used to segment each lung lobe, and then reconstructed into 3 mm 3 of voxels, and Pyradiomics library was used to extract the radiomic features of the whole lung and each lobe. The multi-classification goal was achieved by constructing random forest base classifiers for each of the five groups and then voting as the final model. In the process of constructing the base classifier, firstly, the balance between sample groups was achieved by SMOTETomek comprehensive sampling, and the optimal feature set was selected by independent sample t test and L1 regularized least absolute shrinkage and selection operator (LASSO) regression. In this study, the Radiomics model was constructed based on chest CT radiomics features, and the Radiomics + model was constructed by introducing gender and age information. The base classifier and the integration model use the mean accuracy and the area under the receiver operator characteristics analysis curve (AUC) to evaluate the performance, respectively. Results:There was a statistically significant difference ( P<0.05) between the ages of the NSIP, OP, NSIP+OP, UIP, and DAD groups [(57±13),(53±8),(54±10),(44±11), and (46±8)years old, respectively], F=11.82, P<0.001. In the Radiomics model, for each group of NSIP, OP, NSIP+OP, UIP, and DAD, the AUCs of the training set were 0.87, 0.91, 0.91, 0.96, and 0.99, respectively, and the AUC of the test set were 0.81, 0.82, 0.79, 0.93, 0.89. In the final Radiomics + model, for each group of NSIP, OP, NSIP+OP, UIP, and DAD, the AUCs of the training set were 0.89, 0.91, 0.92, 0.97, and 0.99, respectively, and the AUCs of the test set were 0.84, 0.82, 0.78, 0.94, 0.90. Conclusion:Based on chest CT radiomics features and key clinical features (sex, age), the Radiomics + model constructed by machine learning has good classification performance for the imaging patterns of PM/DM-LD.

Objective:To explore the value of biphasic quantitative CT on small airway disease and emphysema injury in patients with smoking combined with chronic obstructive pulmonary disease (COPD).Methods:A total of 186 male physical examination subjects who underwent biphasic CT and pulmonary function (PFT) examinations in the Affiliated Hospital of Yan′an University from July 2018 to September 2020 were enrolled in this retrospective study. These subjects were divided into 121 smokers with COPD (COPD group), aged 34 to 84 (64±8) years old and 65 smokers without COPD (non-COPD group) aged 34 to 72 (61±5) years old. According to the guidelines of the COPD global initiative, patients in COPD group were divided into Global Initiative for Chronic Obstructive Pulmonary Disease (GOLD) Ⅰ-Ⅳ grades. The original DICOM data of CT were imported into the "Digital Lung" test and analysis platform. Quantitative parameters of functional small airway disease percentage (fSAD%) and emphysema area percentage (Emph%) of each lobe were calculated. The differences of CT quantitative parameters among non-COPD group and each grade in COPD group were analyzed by One-Way ANOVA or Kruskal-Wallis H test. The correlation between the smoking index and CT quantitative parameters was analyzed by Spearman correlation analysis. Results:There were significant differences in fSAD% and Emph% of each lobe among non-COPD group and COPD group GOLD Ⅰ-Ⅳ ( P<0.001). Except that the Emph% in right middle lobe of GOLD grade Ⅰ was higher than that of GOLD grade Ⅱ in COPD group, the fSAD% and Emph% in other lobes increased gradually with the increase of GOLD grade in COPD group. The fSAD% and Emph% were larger in the right middle lobe and both upper lobes of COPD group GOLD Ⅰ-Ⅳ. The comparison among each lobe showed that the differences were statistically significant ( P<0.01), except for the fSAD% and Emph% of GOLD Ⅳ ( P=0.395, 0.840). The smoking index was positively correlated with fSAD% and Emph% in each lung lobe. Among them, smoking index was highly correlated with fSAD% in the lower right lobe and Emph% in the lower left lobe ( r=0.474, 0.619, P<0.001). Conclusion:The biphasic quantitative CT can early and sensitively reflect the degree of small airway disease and emphysema injury in smoking combined with COPD, which is of great significance for the early diagnosis and evolution of COPD.

Objective:To investigate the development of intrapulmonary vascular volume (IPVV) in normal preschool children based on quantitative measurement on chest CT.Methods:The CT data of 407 normal preschool children (236 males and 171 females, aged 1-72 months, with a median of 36 months) who underwent chest CT examination from January 2014 to May 2017 in the "Digital Lung" imaging database were retrospectively collected. The pulmonary vessels were segmented by the "Digital Lung" automatic detection tool, and the IPVV of the whole lung, the right lung, the left lung and each lobe were obtained, and the IPVV upper/lower and IPVV left/right were calculated. According to the age, the subjects were divided into infant period (0-12 months), early childhood period (13-36 months) and preschool period (37-72 months), with 30 cases (17 males and 13 females), 175 cases (95 males and 80 females) and 202 cases (124 males and 78 females) respectively. Spearman correlation analysis was used to evaluate the correlation between IPVV and month age. One-way analysis of variance or Kruskal-Wallis H test was used to compare the differences of IPVV, IPVV upper/lower and IPVV left/right between different months of age. Independent sample t test or Mann-Whitney U test was used to compare the differences of IPVV between different genders, and the normal reference range of IPVV in normal preschool children of different months of age was established. Results:The IPVV of the whole lung, right lung, left lung and each lung lobe were positively correlated with age, the correlation coefficient was 0. 638-0.820 in males and 0. 683-0.791 in females (all P<0.001). There was no significant difference in IPVV of whole lung, right lung, left lung and each lobe between male and female from 0 to 12 months (all P>0.05), but there was significant difference in IPVV of whole lung, right lung, left lung and each lobe between male and female from 13 to 36 months (all P<0.05). There were significant differences in IPVV of the whole lung, right lung, left lung and upper lobe of both lungs between boys and girls from 37 to 72 months (all P<0.05). IPVV upper/lower in the right lung (χ 2=14.00, 12.87, P=0.001, 0.002) and IPVV upper/lower in the left lung (χ 2=6.65, 22.84, P=0.036,<0.001) were significantly different in both boys and girls among 3 months of age. And with the increase of age, it showed a decreasing trend. There was no significant difference in IPVV upper/lower between boys and girls at the same age (all P>0.05).There was no significant difference in IPVV left/right among different months and between different sexes (all P>0.05). Finally, the normal reference value range of IPVV of different genders in infancy, early childhood and preschool age was calculated. Conclusions:The increase of pulmonary vessels in normal preschool children is positively correlated with age. There is no significant difference in IPVV between boys and girls in infant period, but IPVV in boys is larger than that in girls in early childhood period and preschool period. IPVV in the lower lung increased faster than that in the upper lung, but there was no significant difference between the left and right lungs.

Chronic obstructive pulmonary disease (COPD) refers to a group of highly heterogeneous diseases with high prevalence rate and fatality rate, which has become a worldwide public health challenge. Quantitative CT can effectively evaluate the characteristic changes of pulmonary parenchyma, airway and pulmonary vessels, having great significances in early diagnosis, monitoring disease progression and individualized treatment of COPD.

To examine the feasibility of using a computer tool for stratifying the severity of Coronavirus Disease 2019 (COVID-19) based on computed tomography (CT) images. We retrospectively examined 44 confirmed COVID-19 cases. All cases were evaluated separately by radiologists (visually) and through an in-house computer software. The degree of lesions was visually scored by the radiologist, as follows, for each of the 5 lung lobes:0, no lesion present;1,<1/3 involvement;2,>1/3 and<2/3 involvement;and 3,>2/3 involvement. Lesion density was assessed based on the proportion of ground-glass opacity (GGO), consolidation and fibrosis of the lesions. The parameters obtained using the computer tool included lung volume (mL), lesion volume (mL), lesion percentage (％), and mean lesion density (HU) of the whole lung, right lung, left lung, and each lobe. The scores obtained by the radiologists and quantitative results generated by the computer software were tested for correlation. A Chi-square test was used to test the consistency of radiologist- and computer-derived lesion percentage in the right/left lung, upper/lower lobe, and each of the 5 lobes. The results showed a strong to moderate correlation between lesion percentage scores obtained by radiologists and the computer software (r ranged from 0.7679 to 0.8373, P < 0.05), and a moderate correlation between the proportion of GGO and mean lesion density (r=-0.5894, P<0.05), and proportion of consolidation and mean lesion density (r=0.6282, P<0.05). Computer-aided quantification showed a statistical significant higher lesion percentage for lower lobes than that assessed by the radiologists (x2 = 8.160, P = 0.004). Our experiments demonstrated that the computer tool could reliably and accurately assess the severity and distribution of pneumonia on CT scans.

Objective    To summarize the individualized selection of surgical treatment strategies and the key points of perioperative management for patients with heart valve disease complicated with severe chronic heart failure. Methods    The clinical characteristics of 5 male patients with valvular heart disease complicated with severe chronic heart failure (CHF) were analyzed retrospectively from June 2017 to October 2018 in Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, with an average age of 60.21 years. Results    Five patients were given angiotensin receptor and neprilysin inhibitor (ARNI)-based anti-heart failure treatment after admission. The operation mode of these patients was decided to be valve replacement under cardiopulmonary bypass after individualized evaluation of patients’ improving symptoms. Three patients were treated with intra-aortic balloon pump (IABP) and continuous renal replacement therapy (CRRT) early after operation to assist patients in improving cardiac function. Five patients recovered oral anti-heart failure after awakening. All patients were discharged smoothly 2 weeks after operation. Conclusion    Individualized evaluation is needed for the choice of operation timing and mode, standardized preoperative treatment for heart failure, shortening the aortic blocking time during cardiopulmonary bypass, and early application of left ventricular adjuvant drugs or instruments are all important measures to help patients recover smoothly.

Objective To compare the predictive capability of multiple linear regression (MLR)and neural network model (NNM) for pulmonary artery obstruction index (PAOI) in pulmonary embolism.Methods One hundred and forty-seven APE patients (79 male,68 female) were collected from March 2015 to July 2016 in our hospital and randomly divided into training group and testing group with the ratio of 3 ∶ 1.Four indexes,including total volume (V),total length (L),total degree of embolism (D) and total number of clots (N) were calculated by computer assisted detection.Qanadli index (Q) as CT PAOI was calculated manually.With SPSS 14.2 modeler,the predictive value of Qanadli index ((Q)) was calculated by MLR and NNM respectively,with Qanadli index as dependent variable and V,L,D,N as independent variables.SPSS 22.0 Spearman test was used to analyze the correlation between (Q) and Q.Mean absolute error (MAE),mean relative error (MRE),root mean square error (RMSE) were used to quantify the accuracies of two methods.Results MLR equation was (Q)=10.98+ 1.37×V+0.06×L,model fitting was 0.764.NNM included one hidden layer and two neurons with accuracy of 0.868.In training group,the correlation between (Q) and Q in NNM (r=0.932,P＜0.01) was higher than MLR (r=0.879,P＜0.01);in testing group,the correlation between (Q) and Q in NNM (r=0.875,P＜0.01) was higher than MLR (r=0.868,P＜0.01).In training group,MAE,MRE and RMSE of NNM (5.144,0.274,6.957) were significantly lower (t=3.402,P=0.002) than MLR (6.784,0.282,8.700);in testing group,MAE,MRE and RMSE of NNM (6.643,0.312,9.195) were significantly lower (t=3.383,P=0.002) than MLR (8.505,0.334,10.361).Conclusion NNM is a better model in predicting CT pulmonary artery obstruction index of APE patients.

Objective To establish and validate a voxel-based method for the quantitative detection of air trapping (AT),and to explore its diagnostic value by preliminarily apply this method in chronic obstructive pulmonary disease (COPD) patients.Methods From March 2015 to February 2016,fifty healthy young volunteers and eighteen COPD patients who underwent both end-inspiratory and end-expiratory CT were included from the Digital Lung Multi-center Study.The quantitative parameters of AT and emphysema were measured by both the voxel-based quantitative method and the conventional threshold method,respectively.All subjects underwent pulmonary function examination within 3 days after CT examination.For healthy volunteers,paired sample rank-sum test was used to compare the difference of quantitative parameters between voxel-based method and threshold method,Spearman rank correlation analysis was used to investigate the correlation between quantitative parameters of the two methods and pulmonary function.For COPD patients,the distribution and extent of AT and emphysema in patients with similar degree of pulmonary function (PFT) injury were observed.Results There were varying degrees of AT in the asymptomatic youth,with a median value of 5.70％ for the voxel-based method and with a median value of 7.96％ for the conventional threshold method,there was significant difference(Z=-4.015,P＜0.001).The correlation between AT and emphysema parameters of the voxel-based method and PFT parameters (r=-0.399 and-0.494,-0.335 and-0.439 separately,P＜0.05) were higher than that of the conventional threshold method,respectively (r=-0.357 and-0.453,-0.284 and-0.391,respectively;all P＜0.05).Furthermore,the voxel-based method can classify COPD patients with similar degree of pulmonary function injury into three subtypes:AT-dominant,emphysema-dominant,and mixed.Conclusions The voxel-based AT quantitative measurement method not only has high sensitivity and accuracy,but also provides imaging phenotype for the diagnosis of COPD and provides assistant decision-making for clinical management.