Objective To develop and validate a deep learning model based on chest CT images to accurately distinguish between drug-resistant(DR-TB)and-sensitive tuberculosis(DS-TB).Methods A retrospective study was conducted on 722 cases of confirmed secondary tuberculosis admitted in our center from January 2019 to December 2022.According to the results of antimicrobial susceptibility test,they were divided into 357 DS-TB cases and 365 DR-TB cases.Pre-existing U-Net segmentation model was employed to segment the lung parenchyma regions in CT images.The dataset was randomly partitioned into a training set and a testing set in an 8:2 ratio.Six 3D deep learning architectures(3D Swin Transformer,3D ShuffleNet v2,3D ViT,3D MobileNet v2,3D DenseNet,and 3D ResNet18)were employed to evaluate the discriminative efficiency between DS-TB and DR-TB.Hyperparameters were optimized by five-fold cross-validation on the training set to construct the optimal model.The performance of the constructed model was assessed using area under the curve(AUC),sensitivity,specificity,positive predictive value(PPV),negative predictive value(NPV),and F1-score.Six radiologists independently evaluated DR-TB identification on the test set,and their performance was compared with the best-performing deep learning model.Results The AUC value in DR-TB prediction was 0.583,0.704,0.698,0.758,0.736,and 0.841,respectively,for 3D Swin Transformer,3D ShuffleNet v2,3D ViT,3D MobileNet v2,3D DenseNet,and 3D ResNet18.The 3D ResNet18 model demonstrated optimal performance,achieving a sensitivity of 0.935(95%CI:0.880～0.987),a specificity of 0.642(95%CI:0.492～0.757),a PPV of 0.750(95%CI:0.663～0.835),an NPV of 0.896(95%CI:0.809～0.976),an AUC value of 0.841,and a F1-score of 0.832.The radiologists got a F1-score of 0.571,0.450,0.675,0.623,0.617 and 0.635,respectively,and the F1-score of the 3D ResNet18 model is all higher than that of the radiologists.The highest-performing radiologist achieved sensitivity,specificity,PPV and NPV of 0.701(95%CI:0.605～0.802),0.567(95%CI:0.447～0.684),0.651(95%CI:0.549～0.757),and 0.623(95%CI:0.500～0.754),with all these values lower than those of the 3D ResNet18 model(P<0.05).Class activation mapping showed that the 3D ResNet18 model could focus on key lesion areas.The class activation mapping demonstrated that the 3D ResNet18 model could effectively focus on critical lesion regions.Conclusion Our 3D ResNet18 model shows the best predictive performance in identifying DR-TB,and is expected to assist clinical diagnosis for DR-TB.

Objective: To investigate the features of chest CT imaging and dynamic changes of severe coronavirus disease 2019 (COVID-19). Methods: The clinical and computed tomography (CT) data of 17 patients diagnosed with severe COVID-19 admitted to Chongqing Public Health Medical Center from January 24 to February 6, 2020 were collected. The first chest CT manifestations and the dynamic changes of imaging during treatment were retrospectively analyzed. Results: The first chest CT manifestations of the 17 patients showed that 16 cases presented with peripheral and subpleural distributions, and 2 cases presented with 3 lobes involved, one case with 4 lobes involved and 14 cases with 5 lobes involved, and 17 cases presented with ground-glass opacities, ten cases with consolidation, seven cases with subpleural line, nine cases with air bronchogram, 3 cases with thickened lobular septum, two cases with bronchiectasis, two cases with pleural effusion, two cases with lymphadenopathy with the short diameter of 1.0-1.2cm. Among 16 patients who underwent repeated CT examination, the lesions of 8 patients showed continuous improvement, and those of the other 8 patients showed fluctuating changes. Conclusions The CT findings of severe COVID-19 patients are mainly ground-glass opacities and consolidation, with the peripheral distribution. The range of lesions is wide, with 5-lobe involvement mostly. Lymphadenopathy or pleural effusion is rare. Chest CT is useful for the evaluation for the therapeutic effects.

Objective:To investigate the features of chest computed tomography (CT) imaging and dynamic changes of severe corona virus disease 2019 (COVID-19).Methods:The clinical and CT data of 17 patients diagnosed with severe COVID-19 admitted to Chongqing Public Health Medical Center from January 24 to February 6, 2020 were collected. The first chest CT manifestations and the dynamic changes of imaging during treatment were retrospectively analyzed.Results:The first chest CT manifestations of the 17 patients showed that 16 cases presented with peripheral and subpleural distributions, and two cases presented with three lobes involved, one case with four lobes involved and 14 cases with five lobes involved, and 17 cases presented with ground-glass opacities, ten cases with consolidation, seven cases with subpleural line, nine cases with air bronchogram, three cases with thickened lobular septum, two cases with bronchiectasis, two cases with pleural effusion, three cases with lymphadenopathy with the short diameter of 1.0-1.2 cm.Among 16 patients who underwent repeated CT examination, the lesions of eight patients showed continuous improvement, and those of the other eight patients showed fluctuating changes.Conclusions:The CT findings of severe COVID-19 patients are mainly ground-glass opacities and consolidation, with the peripheral distribution. The range of lesions is wide, with five-lobe involvement mostly. Lymphadenopathy or pleural effusion is rare. Pynamic monitoring chest CT is useful for the evaluation for the therapeutic effects.