Objective:To explore the value of bedside chest radiograph in the diagnosis and follow-up of severe and critical COVID-19.Methods:Twenty-nine patients with severe or critical COVID-19 were collected from January 23 to February 23, 2020，from four COVID-19 designated hospitals in Guangdong Province. Bedside radiography was taken in all the 29 patients, ranged from 1 to 16 times for each patient. Twenty-seven patients underwent follow-up, and the number of re-examination ranged 1 to 15 times, and the interval of review is 1 to 8 days.The imaging findings of bedside chest radiography and the imaging changes on follow-up chest radiography were analyzed retrospectively.Results:Twenty-nine patients were collected. The radiography showed the lesions involved all more than 3 lung fields. The films showed consolidation shadow in 19 cases, multiple patches of shadow in 23 cases, reticular pattern in 12 cases, strips shadow in 14 cases, interlobar fissure thickening in 18 cases, and "white lung" in 4 cases.The complications included pleural effusion in 4 cases, pneumothorax in 2 cases, mediastinal and subcutaneous emphysema in 1 case. The radiography showed the lesions progressed in 15 cases, with expanded involvement of the lung.The increase of lesion density was found in 6 cases, new lesions were noted in 5 cases, while both of them were found in 4 cases. Nine cases showed improvement, with reduced range and decreased density. Patchy or consolidation shadow turned to strips shadow or articular pattern shadow in 8 cases.There was no significant change in 3 cases with large consolidation shadow.Conclusions:Bedside chest radiography has a good value in the follow-up of severely and critically ill patients with COVID-19, and can provide great help for clinicians to evaluate their condition.

Objective To explore the predictive value of CT radiomics and morphological features for the prognosis and survival in non-small cell lung cancer(NSCLC)patients.Methods The clinic data of 300 NSCLC patients(300 lesions)were downloaded from the Cancer Imaging Archive,with 210 randomly selected as the training set and 90 as the test set.According to the prognosis and survival,the patients were divided into two groups with survival period≤3 and＞3 years.3D Slicer software was used to delineate the regions of interest layer by layer in CT images,and the radiomics features were extracted from each region of interest.Both t-test and least absolute shrinkage and selection operator were utilized for radiomics feature screening.Three types of prediction models,namely radiomics model,morphological model and combined model,were constructed with Logistic regression,whose performances were evaluated using the receiver operating characteristic(ROC)curve.Results The differences in radiomics labels and mediastinal lymph node metastasis between the training set and the test set were statistically significant.For radiomics model,morphological model and combined model,the area under the ROC curve was 0.784(95%CI:0.722-0.847),0.734(95%CI:0.664-0.804)and 0.748(95%CI:0.680-0.815)in the training set,and 0.737(95%CI:0.630-0.844),0.665(95%CI:0.554-0.777)and 0.687(95%CI:0.578-0.797)in the test set,which demonstrated that radiomics model had the best diagnostic performance.Conclusion The CT radiomics model can effectively predict the prognosis and survival in NSCLC patients.

Objective To develop deep learning-based auxiliary diagnostic models for diverse pulmonary diffuse cystic diseases,and subsequently evaluate their classification performance to identify the optimal model for clinical diagnosis.Methods A total of 288 patients diagnosed with idiopathic pulmonary fibrosis(IPF),pulmonary lymphangioleiomyomatosis(PLAM),and pulmonary Langerhans cell histiocytosis(PLCH)were prospectively enrolled from the First Affiliated Hospital of Guangzhou Medical University between January 2010 and October 2022,comprising 76 cases of IPF,179 cases of PLAM,and 33 cases of PLCH.A total of 877 CT cases were collected,comprising 232 cases of IPF,557 cases of PLAM,and 88 cases of pulmonary PLCH.Based on the cutoff date of December 31,2019,the CT scans were divided into two datasets:dataset A consisted of 500 CT scans including 185 IPF cases,265 PLAM cases,and 50 PLCH cases;while dataset B comprised 377 CT scans with a distribution of 47 IPFcases,292 PLAMcases,and 38 PLCH cases.The Dataset A was randomly partitioned into training set,validation set,and test set in a ratio of 7∶1∶2.Subsequently,six distinct deep learning neural networks were employed for training after preprocessing and data augmentation.Receiver operating characteristic curves were generated to assess the model performance using metrics such as area under the curve(AUC),accuracy,sensitivity,specificity,and F1 score in order to identify the optimal model.Furthermore,a test set B comprising 30 randomly selected cases from dataset B for each disease type was utilized to evaluate the trained optimal model by employing the same aforementioned metrics.Results In test A,six well-established diagnostic models demonstrated superior classification performance for IPF and LAM,with an AUC greater than 0.9.For LCH,EfficientNet exhibited low classification efficiency with an AUC between 0.6 and 0.7,while Vgg11 showed an AUC between 0.8 and 0.9;the other four models displayed excellent classification efficiency with an AUC greater than 0.9.Except for Inception V3,the remaining five diagnostic models performed poorly in identifying and classifying LCH lesions.Considering multiple indicators,the InceptionV3 model showcased optimal comprehensive performance among the six models,achieving high evaluation parameters such as overall accuracy(94.90%),precision(93.49%),recall(90.84%),and specificity(96.91%).TestB was conducted using the trained InceptionV3 model resulting in an accuracy of 81%,precision of 82%,recall of 81%,and specificity of 90%.Conclusions Six recognition and classification models,developed using deep learning technology in conjunction with pulmonary CT images,demonstrate effective discrimination between LAM,LCH,and IPF.Notably,the model constructed utilizing the InceptionV3 neural network exhibits superior efficiency in accurately recognizing and classifying IPF and LAM.

Objective The evaluation of lymph node properties before lung cancer surgery has a great impact with the choice of surgical methods.Although there are various examination methods,many methods have invasive or accuracy problems.In order to improve the accuracy of diagnosis,we mainly discuss the value of wide-body spectral CT in the differential diagnosis of mediastinal metastatic lymph nodes,non-metastatic lymph nodes in lung cancer patients and reactive hyperplastic lymph nodes.Methods The clinical and imaging data of 64 patients with lung cancer and 28 patients with pulmonary inflammatory lesions were retrospectively analyzed.All patients underwent plain scan and enhanced dual-phase spectral CT scan.The size,density,three-phase IC,NIC,and λHU of lymph nodes in metastatic,non-metastatic and inflammatory reactive hyperplasia groups were measured on 70 keV single-energy images and iodine-based images,respectively.The single-factor variance and Kruskal-Wallis H rank sum test were used to analyze and compare the differences.Results The short diameter of metastatic lymph nodes was larger than that of non-metastatic lymph nodes and reactive hyperplastic lymph nodes(P<0.001).The plain scan density of reactive hyperplastic lymph nodes was higher than that of metastatic lymph nodes(P<0.001),but there was no significant difference between non-metastatic lymph nodes(P=0.325).The CT values of reactive hyperplastic lymph nodes in arterial phase and venous phase were higher than those of metastatic and non-metastatic lymph nodes(P<0.05).Except for NIC in arterial phase,IC,NIC and λHU in plain scan,IC and λHU in arterial phase,IC,NIC and λHU in venous phase of reactive hyperplastic lymph nodes and metastatic lymph nodes were statistically significant(all P<0.05).There was no significant difference in IC,NIC and λHU between reactive hyperplastic lymph nodes and non-metastatic lymph nodes in plain scan,arterial phase and venous phase(all P>0.05).Conclusion The quantitative and spectral curve slope of iodine in mediastinal metastatic lymph nodes of lung cancer were basically lower than those in reactive hyperplastic lymph nodes.The quantitative parameters of spectral CT had certain diagnostic efficacy in differentiating metastatic lymph nodes and reactive hyperplastic lymph nodes,while the spectral parameters of non-metastatic lymph nodes and reactive hyperplastic lymph nodes were not statistically significant.