ABSTRACT
Objective:To assess the effectiveness of a model created using clinical features and preoperative chest CT imaging features in predicting the chronic obstructive pulmonary disease (COPD) among patients diagnosed with lung cancer.Methods:A retrospective analysis was conducted on clinical (age, gender, smoking history, smoking index, etc.) and imaging (lesion size, location, density, lobulation sign, etc.) data from 444 lung cancer patients confirmed by pathology at the Second Affiliated Hospital of Naval Medical University between June 2014 and March 2021. These patients were randomly divided into a training set (310 patients) and an internal test set (134 patients) using a 7∶3 ratio through the random function in Python. Based on the results of pulmonary function tests, the patients were further categorized into two groups: lung cancer combined with COPD and lung cancer non-COPD. Initially, univariate analysis was performed to identify statistically significant differences in clinical characteristics between the two groups. The variables showing significance were then included in the logistic regression analysis to determine the independent factors predicting lung cancer combined with COPD, thereby constructing the clinical model. The image features underwent a filtering process using the minimum absolute value convergence and selection operator. The reliability of these features was assessed through leave-P groups-out cross-validation repeated five times. Subsequently, a radiological model was developed. Finally, a combined model was established by combining the radiological signature with the clinical features. Receiver operating characteristic (ROC) curves and decision curve analysis (DCA) curves were plotted to evaluate the predictive capability and clinical applicability of the model. The area under the curve (AUC) for each model in predicting lung cancer combined with COPD was compared using the DeLong test.Results:In the training set, there were 182 cases in the lung cancer combined with COPD group and 128 cases in the lung cancer non-COPD group. The combined model demonstrated an AUC of 0.89 for predicting lung cancer combined with COPD, while the clinical model achieved an AUC of 0.82 and the radiological model had an AUC of 0.85. In the test set, there were 78 cases in the lung cancer combined with COPD group and 56 cases in the lung cancer non-COPD group. The combined model yielded an AUC of 0.85 for predicting lung cancer combined with COPD, compared to 0.77 for the clinical model and 0.83 for the radiological model. The difference in AUC between the radiological model and the clinical model was not statistically significant ( Z=1.40, P=0.163). However, there were statistically significant differences in the AUC values between the combined model and the clinical model ( Z=-4.01, P=0.010), as well as between the combined model and the radiological model ( Z=-2.57, P<0.001). DCA showed the maximum net benifit of the combined model. Conclusion:The developed synthetic diagnostic combined model, incorporating both radiological signature and clinical features, demonstrates the ability to predict COPD in patients with lung cancer.
ABSTRACT
Objective:To explore the differences of functional small airway and pulmonary vascular parameters in chronic obstructive pulmonary disease (COPD) of different imaging phenotypes.Methods:One hundred and thirty COPD patients underwent biphasic CT scanning in Shanghai Changzheng Hospital from August 2018 to August 2020 were analyzed retrospectively. The patients were classified into three phenotypes based on the presence of emphysema and bronchial wall thickening on CT images. Phenotype A: no emphysema or mild emphysema, with or without bronchial wall thickening; Phenotype E: obvious emphysema without bronchial wall thickening; phenotype M: significant emphysema and bronchial wall thickening were present. Parametric response map (PRM) and pulmonary vascular parameters were quantitatively measured at the whole lung level. PRM parameters included the volume of emphysema (PRMV Emphysema), the volume of functional small airway (PRMV fSAD), the volume of normal pulmonary parenchyma (PRMV Normal) and its volume percentage (%). Pulmonary vascular parameters included the number of vessels (N) and cross-sectional area vessels<5 mm 2 (N -CSA<5) at 6, 9, 12, 15, 18 21, 24 mm distance from the pleura. ANOVA or Kruskal-Wallis H tests were used to compare the differences for PRM and pulmonary vascular parameters among the three phenotypes, and LSD or Bonferroni tests were used for multiple comparisons. Results:There were significant differences among the three phenotypes for PRMV fSAD, PRMV Emphysema, PRMV fSAD%, PRMV Emphysema%, and PRMV Normal% at the whole lung level ( P<0.05). PRMV Emphysema, PRMV Emphysema%, PRMV Fsad, PRMV fSAD% of phenotype A were lower than those of phenotype E and M ( P<0.001), while there was no significant difference for PRMV Emphysema, PRMV Emphysema%, PRMV fSAD, PRMV fSAD% between phenotype E and phenotype M ( P>0.05). There were significant differences in N and N -CSA<5 that 6 mm distance from the pleura among the three groups( P<0.05). Among them, N and N -CSA<5 that 6 mm distance from pleura in phenotype M were significantly lower than those in phenotype A( P<0.001,0.002); No significant differences was found in N between phenotype M and phenotype E( P>0.05), while there was significant differences in N -CSA<5 between phenotype M and phenotype E( P=0.034). Conclusion:Biphasic quantitative CT analysis can reflect the heterogeneity of the functional small airways and pulmonary vascular abnormality in COPD with different phenotypes, and provide objective evidence for individualized diagnosis and treatment.
ABSTRACT
Objective:To explore the predictive value of random forest regression model for pulmonary function test.Methods:From August 2018 to December 2019, 615 subjects who underwent screening for three major chest diseases in Shanghai Changzheng Hospital were analyzed retrospectively. According to the ratio of forced expiratory volume in the first second to forced vital capacity (FEV 1/FVC) and the percentage of forced expiratory volume in the first second to the predicted value (FEV 1%), the subjects were divided into normal group, high risk group and chronic obstructive pulmonary disease (COPD) group. The CT quantitative parameter of small airway was parameter response mapping (PRM) parameters, including lung volume, the volume of functional small airways disease (PRMV fSAD), the volume of emphysema (PRMV Emph), the volume of normal lung tissue (PRMV Normal), the volume of uncategorized lung tissue (PRMV Uncategorized) and the percentage of the latter four volumes to the whole lung (%). ANOVA or Kruskal Wallis H was used to test the differences of basic clinical characteristics (age, sex, height, body mass), pulmonary function parameters and small airway CT quantitative parameters among the three groups; Spearman test was used to evaluate the correlation between PRM parameters and pulmonary function parameters. Finally, a random forest regression model based on PRM combined with four basic clinical characteristics was constructed to predict lung function. Results:There were significant differences in the parameters of whole lung PRM among the three groups ( P<0.001). Quantitative CT parameters PRMV Emph, PRMV Emph%, and PRMV Normal% showed a moderate correlation with FEV 1/FVC ( P<0.001). Whole lung volume, PRMV Normal,PRMV Uncategorized and PRMV Uncategorized% were strongly or moderately positively correlated with FVC ( P<0.001), other PRM parameters were weakly or very weakly correlated with pulmonary function parameters. Based on the above parameters, a random forest model for predicting FEV 1/FVC and a random forest model for predicting FEV 1% were established. The random forest model for predicting FEV 1/FVC predicted FEV 1/FVC and actual value was R 2=0.864 in the training set and R 2=0.749 in the validation set. The random forest model for predicting FEV 1% predicted FEV 1% and the actual value in the training set was R 2=0.888, and the validation set was R 2=0.792. The sensitivity, specificity and accuracy of predicting FEV 1% random forest model for the classification of normal group from high-risk group were 0.85(34/40), 0.90(65/72) and 0.88(99/112), respectively; and the sensitivity, specificity and accuracy of predicting FEV 1/FVC random forest model for differentiating non COPD group from COPD group were 0.89(8/9), 1.00 (112/112) and 0.99(120/121), respectively. While the accuracy of two models combination for subclassification of COPD [global initiative for chronic obstructive lung disease (GOLD) Ⅰ, GOLDⅡ and GOLD Ⅲ+Ⅳ] was only 0.44. Conclusions:Small airway CT quantitative parameter PRM can distinguish the normal population, high-risk and COPD population. The comprehensive regression prediction model combined with clinical characteristics based on PRM parameter show good performance differentiating normal group from high risk group, and differentiating non-COPD group from COPD group. Therefore, one-stop CT scan can evaluate the functional small airway and PFT simultaneously.
ABSTRACT
Objective:To investigate the value of CT features in predicting visceral pleural invasion (VPI) in clinical stage ⅠA peripheral lung adenocarcinoma under the pleura.Methods:The CT signs of 274 patients with clinical stage ⅠA peripheral lung adenocarcinoma under the pleura diagnosed in Changzheng Hospital of Naval Medical University from January 2015 to November 2021 were retrospectively analyzed. According to the ratio of 6∶4, 164 patients collected from January 2015 to August 2019 were used as the training group, and 110 patients collected from August 2019 to November 2021 were used as the validation group. The maximum diameter of the tumor (T), the maximum diameter of the consolidation part (C), and the minimum distance between the lesion and the pleura (DLP) were quantitatively measured, and the proportion of the consolidation part was calculated (C/T ratio, CTR). The CT signs of the tumor were analyzed, such as the relationship between the tumor and the pleura classification, the presence of a bridge tag sign, the location of the lesion, density type, shape, margin, boundary and so on. Variables with significant difference in the univariate analysis were entered into multivariate logistic regression analysis to explore predictors for VPI, and a binary logistic regression model was established. The predictive performance of the model was analyzed by receiver operating characteristic curve in the training and validation group.Results:There were 121 cases with VPI and 153 cases without VPI among the 274 patients with lung adenocarcinoma. There were 79 cases with VPI and 85 cases without VPI in the training group. Univariate analysis found that the maximum diameter of the consolidation part, CTR, density type, spiculation sign, vascular cluster sign, relationship of tumor and pleura and bridge tag sign between patients with VPI and those without VPI were significantly different in the training group( P<0.05). Multivariate logistic regression analysis found the relationship between tumor and pleura [taking type Ⅰ as reference, type Ⅱ (OR=6.662, 95%CI 2.364-18.571, P<0.001), type Ⅲ (OR=34.488, 95%CI 8.923-133.294, P<0.001)] and vascular cluster sign (OR=4.257, 95%CI 1.334-13.581, P=0.014) were independent risk factors for VPI in the training group. The sensitivity, specifcity, and area under curve (AUC) for the logistic model in the training group were 62.03%, 89.41% and 0.826, respectively, using the optimal cutoff value of 0.504. The validation group obtained an sensitivity, specifcity, and AUC of 92.86%, 47.06%, and 0.713, respectively, using the optimal cutoff value of 0.449. Conclusion:The relationship between the tumor and the pleura and the vascular cluster sign in the CT features can help to predict visceral pleural invasion in the clinical stage ⅠA peripheral lung adenocarcinoma under the pleura.
ABSTRACT
Objective To evaluate the effectiveness of deep learning methods to detect subsolid nodules from chest X-ray images.Methods The building,training,and testing of the deep learning model were performed using the research platform developed by Infervision,China.The training dataset consisted of 1 965 chest X-ray images, which contained 85 labeled subsolid nodules and 1 880 solid nodules. Eighty-five subsolid nodules were confirmed by corresponding CT exams. We labeled each X-ray image using the corresponding reconstructed coronal slice from the CT exam as the gold standard,and trained the deep learning model using alternate training.After the training,the model was tested on a different dataset containing 56 subsolid nodules,which were also confirmed by corresponding coronal slices from CT exams. The model results were compared with an experienced radiologist in terms of sensitivity,specificity,and test time. Results Out of the testing dataset that contained 56 subsolid nodules, the deep learning model marked 72 nodules,which consisted of 39 true positives(TP)and 33 false positives(FP).The model took 17 seconds.The human radiologist marked 39 nodules,with 31 TP and 8 FP.The radiologist took 50 minutes and 24 seconds. Conclusions Subsolid nodules are prone to mis-diagnosis by human radiologists. The proposed deep learning model was able to effectively identify subsolid nodules from X-ray images.
ABSTRACT
Objective To develop and validate the radiomics nomogram on the discrimination of lung invasive adenocarcinoma from'non-invasive'lesion manifesting as ground glass nodule(GGN)and compare it with morphological features and quantitative imaging. Methods One hundred and sixty pathologically confirmed lung adenocarcinomas from November 2011 to December 2014 were included as primary cohort. Seventy-six lung adenocarcinomas from November 2014 to December 2015 were set as an independent validation cohort. Lasso regression analysis was used for feature selection and radiomics signature building. Radiomics score was calculated by the linear fusion of selected features. Multivariable logistic regression analysis was performed to develop models. The prediction performances were evaluated with ROC analysis and AUC,and the different prediction performance between different models and mean CT value were compared with Delong test. The generalization ability was evaluated with the leave-one-out cross-validation method. The performance of the nomogram was evaluated in terms of its calibration. The Hosmer-Lemeshow test was used to evaluate the significance between the predictive and observe values.Results Four hundred and eighty-five 3D features were extracted and reduced to 2 features as the most important discriminators to build the radiomics signatures. The individualized prediction model was developed with age, radiomics signature, spiculation and pleural indentation, which had the best discrimination performance(AUC=0.934)in comparison with other models and mean CT value(P<0.05)and showed better performance compared with the clinical model(AUC=0.743,P<0.001).The radiomics-based nomogram demonstrated good calibration in the primary and validation cohort, and showed improved differential diagnosis performance with an AUC of 0.956 in the independent validation cohort. Conclusion Individualized prediction model incorporating with age, radiomics signature, spiculation and pleural indentation, presenting with radiomics nomogram, could differentiate IAC from'non-invasive'lesion manifesting as GGN with the best performance in comparison with morphological features and quantitative imaging.
ABSTRACT
Objective To know the stability of 3 kinds of traditional Chinese drug injection with different solvent under different temperature and different storage time,and then reference to safty of clinical durgs. Methods Use particle analyzer,UV SpectropHotometer,pH Determination of three traditional Chinese medicine injection,at different temperatures and different times,the nuanber of particles,pH value,the value of UV absorbance Observe changes. Results The records of experimental data by repeated measures analysis of statistics: Ciwujia Injection,Aidi injection combined with normal saline solution after the particles of ≥ 10μm excess pharmacopoeia standards; ≥ 2μm number of particles is considerable,in the 32~35℃,Ciwujia at T=O min Aidi at T=I80 min ,the number of particles are smaller and with statistical significance,its pH value of the standard range. Xuesaitong injection with glucose injection ≥ 10μm particulate mixture at a higher temperature in excess of pH annacopoeia standards,number of ≥ 2 μm particles in the 4~8℃ and 20~23℃,T=30 min and 60 min time less with statistical significance,and its range of pH value less than pHarmacopoeia. Three kinds of Chinese medicine in the Department's largest UV absorbance peak value and appearance almost unchanged. Conclusions Different drugs in their relative Suggestions of temperature,time and place under intravenous drug use,or better terminal filter,in order to improve the safety of clinical medication.