Artificial intelligence for stepwise diagnosis and monitoring of COVID-19.

BACKGROUND: Main challenges for COVID-19 include the lack of a rapid diagnostic test, a suitable tool to monitor and predict a patient's clinical course and an efficient way for data sharing among multicenters. We thus developed a novel artificial intelligence system based on deep learning (DL) and federated learning (FL) for the diagnosis, monitoring, and prediction of a patient's clinical course. METHODS: CT imaging derived from 6 different multicenter cohorts were used for stepwise diagnostic algorithm to diagnose COVID-19, with or without clinical data. Patients with more than 3 consecutive CT images were trained for the monitoring algorithm. FL has been applied for decentralized refinement of independently built DL models. RESULTS: A total of 1,552,988 CT slices from 4804 patients were used. The model can diagnose COVID-19 based on CT alone with the AUC being 0.98 (95% CI 0.97-0.99), and outperforms the radiologist's assessment. We have also successfully tested the incorporation of the DL diagnostic model with the FL framework. Its auto-segmentation analyses co-related well with those by radiologists and achieved a high Dice's coefficient of 0.77. It can produce a predictive curve of a patient's clinical course if serial CT assessments are available. INTERPRETATION: The system has high consistency in diagnosing COVID-19 based on CT, with or without clinical data. Alternatively, it can be implemented on a FL platform, which would potentially encourage the data sharing in the future. It also can produce an objective predictive curve of a patient's clinical course for visualization. KEY POINTS: • CoviDet could diagnose COVID-19 based on chest CT with high consistency; this outperformed the radiologist's assessment. Its auto-segmentation analyses co-related well with those by radiologists and could potentially monitor and predict a patient's clinical course if serial CT assessments are available. It can be integrated into the federated learning framework. • CoviDet can be used as an adjunct to aid clinicians with the CT diagnosis of COVID-19 and can potentially be used for disease monitoring; federated learning can potentially open opportunities for global collaboration.

Revolution spectral CT for urinary stone with a single/mixed composition in vivo: a large sample analysis.

PURPOSE: To analyze various compositions of urinary stones using revolution spectral CT (rapid kV switching dual-energy CT) in vivo. METHODS: 202 patients with urinary stones underwent spectral CT before surgery. Zeff peak, overall scope and CT values were detected. Moreover, water/iodine attenuating material images were obtained. Removed stones were subjected to infrared spectroscopy after surgery. The results of infrared spectroscopy were compared with CT. RESULTS: 28 stones (14.08%) with single composition, 165 stones with two mixed compositions (81.68%), and 9 stones with three mixed compositions (4.46%) were observed. When Zeff peaks of stones with single/mixed compositions were summarized together, 146 peaks of calcium oxalate monohydrate, 119 peaks of calcium oxalate dihydrate, 55 peaks of carbapatite, 38 peaks of urate, 16 peaks of struvite, and 11 peaks of brushite were totally observed. 93.8% of calcium oxalate monohydrate had Zeff peaks between 13.3 and 14.0. 91.6% of calcium oxalate dihydrate had peaks between 12.0 and 13.3. For carbapatite, 90.9% of stones had peaks from 14.0 to 15.0. A total of 94.8% of urate had peaks between 7.0 and 11.0. 93.8% of struvite had peaks between 11.0 and 13.0, and 90.9% of brushite had peaks between 12.0 and 14.0. Moreover, densities of urate, struvite and brushite were low density in iodine-based images and high-density in water-based images. CONCLUSION: The in-vivo analysis of spectral CT in urinary stone revealed characteristics of different compositions, especially mixed compositions. An in-vivo predictive model may be constructed to distinguish stone compositions.

The development and validation of a radiomic nomogram for the preoperative prediction of lung adenocarcinoma.

BACKGROUND: Accurate diagnosis of early lung cancer from small pulmonary nodules (SPN) is challenging in clinical setting. We aimed to develop a radiomic nomogram to differentiate lung adenocarcinoma from benign SPN. METHODS: This retrospective study included a total of 210 pathologically confirmed SPN (≤ 10 mm) from 197 patients, which were randomly divided into a training dataset (n = 147; malignant nodules, n = 94) and a validation dataset (n = 63; malignant nodules, n = 39). Radiomic features were extracted from the cancerous volumes of interest on contrast-enhanced CT images. The least absolute shrinkage and selection operator (LASSO) regression was used for data dimension reduction, feature selection, and radiomic signature building. Using multivariable logistic regression analysis, a radiomic nomogram was developed incorporating the radiomic signature and the conventional CT signs observed by radiologists. Discrimination and calibration of the radiomic nomogram were evaluated. RESULTS: The radiomic signature consisting of five radiomic features achieved an AUC of 0.853 (95% confidence interval [CI]: 0.735-0.970), accuracy of 81.0%, sensitivity of 82.9%, and specificity of 77.3%. The two conventional CT signs achieved an AUC of 0.833 (95% CI: 0.707-0.958), accuracy of 65.1%, sensitivity of 53.7%, and specificity of 86.4%. The radiomic nomogram incorporating the radiomic signature and conventional CT signs showed an improved AUC of 0.857 (95% CI: 0.723-0.991), accuracy of 84.1%, sensitivity of 85.4%, and specificity of 81.8%. The radiomic nomogram had good calibration power. CONCLUSION: The radiomic nomogram might has the potential to be used as a non-invasive tool for individual prediction of SPN preoperatively. It might facilitate decision-making and improve the management of SPN in the clinical setting.

Computed Tomography-Based Radiomic Features for Diagnosis of Indeterminate Small Pulmonary Nodules.

OBJECTIVE: This study aimed to determine the potential of radiomic features extracted from preoperative computed tomography to discriminate malignant from benign indeterminate small (≤10 mm) pulmonary nodules. METHODS: A total of 197 patients with 210 nodules who underwent surgical resections between January 2011 and March 2017 were analyzed. Three hundred eighty-five radiomic features were extracted from the computed tomographic images. Feature selection and data dimension reduction were performed using the Kruskal-Wallis test, Spearman correlation analysis, and principal component analysis. The random forest was used for radiomic signature building. The receiver operating characteristic curve analysis was used to evaluate the model performance. RESULTS: Fifteen principal component features were selected for modeling. The area under the curve, sensitivity, specificity, and accuracy of the prediction model were 0.877 (95% confidence interval [CI], 0.795-0.959), 81.8% (95% CI, 72.0%-90.9%), 77.4% (95% CI, 63.9%-89.3%), and 80.0% (95% CI, 72.0%-86.7%) in the validation cohort, respectively. CONCLUSIONS: Computed tomography-based radiomic features showed good discriminative power for benign and malignant indeterminate small pulmonary nodules.

[Value of anti-neutrophil cytoplasmic antibody in assessing the severity of bronchiolitis obliterans in children].

OBJECTIVE: To study the value of anti-neutrophil cytoplasmic antibody (ANCA) in assessing the severity of bronchiolitis obliterans (BO) in children. METHODS: A prospective analysis was performed on 59 children who were diagnosed with BO from June 2009 to October 2014. ELISA was used to measure the concentrations of myeloperoxidase (MPO)-ANCA and proteinase 3 (PR3)-ANCA in serum. According to the results of ELISA, the children were divided into three groups: double-negative ANCA (n=22), single-positive ANCA (n=17), and double-positive ANCA (n=20). The three groups were compared in terms of the scores of BO risk factors, clinical symptoms, chest high-resolution computed tomography (HRCT), and lung pathology on admission, as well as the changes in the expression level of ANCA and the scores of clinical symptoms and chest HRCT over time. RESULTS: Compared with the double-negative ANCA group, the double-positive ANCA group had a significantly higher score of BO risk factors (P<0.05), and the single-positive ANCA group and the double-positive ANCA group had significantly higher scores of clinical symptoms, chest HRCT, and lung pathology (P<0.05). The children were followed up for 6 months after discharge, and there were significant reductions in MPO-ANCA and PR3-ANCA titers from admission and discharge to the end of follow-up (P<0.05), as well as a significant reduction in the score of clinical symptoms from admission to the end of follow-up (P<0.05), while there was no significant change in the score of chest HRCT from admission to the end of follow-up (P>0.05). The single-positive ANCA and double-positive ANCA groups still had a significantly higher score of clinical symptoms than the double-negative ANCA group (P<0.05). CONCLUSIONS: The expression level of ANCA is correlated with the severity of BO in children and thus has certain clinical significance in disease evaluation.

Preoperative localisation of pulmonary ground-glass opacity using medical adhesive before thoracoscopic resection.

OBJECTIVES: To evaluate the safety and efficiency of computed tomography (CT)-guided medical adhesive, α-cyanoacrylate, for preoperative localisation of pulmonary ground-glass opacity (GGO) used for guiding the video-assisted thoracoscopic surgical (VATS) excision METHODS: The procedure was performed on 188 consecutive patients with solitary GGO (pure GGO = 90 cases; mixed GGO = 98 cases) prior to the thoracoscopic procedure. The complications and efficacy of this method were analysed. The resected GGO was analysed pathologically. RESULTS: The mean duration of the procedure was 16.3 ± 5.2 min. The preoperative localisation was 100% successful. All GGOs were successfully resected by VATS. Asymptomatic pneumothorax was developed in 16/188 patients (8.5%) and mild pulmonary haemorrhage occurred in 15 cases (7.9%) post-localisation. None of the patients required any further treatment for the complications. CONCLUSION: Preoperative localisation using CT-guided medical adhesive, α-cyanoacrylate, is a safe and short-duration procedure, with high accuracy and success rates with respect to VATS resection of GGO. KEY POINTS: • Preoperative localisation is crucial for successful resection of GGO by VATS. • Preoperative adhesive localisation provides an up to 100% successful localisation rate with few complications. • Preoperative adhesive localisation enabled VATS resection in 100% of the GGO. • Preoperative adhesive localisation is safe and effective for VATS resection of GGO.

Quantitative Low-Dose Computed Tomography of the Lung Parenchyma and Airways for the Differentiation between Chronic Obstructive Pulmonary Disease and Asthma Patients.

BACKGROUND: It is difficult to differentiate between chronic obstructive pulmonary disease (COPD) and asthma in clinics; therefore, for diagnostic purposes, imaging-based measurements could be beneficial to differentiate between the two diseases. OBJECTIVES: We aim to analyze quantitative measurements of the lung and bronchial parameters that are provided by low-dose computed tomography (CT) to differentiate COPD and asthma from an imaging perspective. MATERIALS AND METHODS: 69 COPD patients, 52 asthma patients, and 20 healthy subjects were recruited to participate in CT imaging and pulmonary function tests (PFTs). Comparative analysis was performed to identify differences between COPD and asthma in CT measurements. PFT measurements enabled validation of the differentiation between COPD and asthma patients. RESULTS: There were significant differences among the COPD, asthma, and healthy control groups. The differences were more significant among the following: inspiratory emphysema index (EI)-950 (%), expiratory lung volume, expiratory mean lung density (MLD), and expiratory EI-950 (%) and EI-850 (%). The COPD group had a significantly higher EI-950 (%) than the asthma group (p = 0.008). There were significant differences among the three groups in lumen area (LA), wall area (WA), total area, and Pi10WA. The asthma group had significantly higher WA%/WV% than both the COPD (p = 0.002) and the control group (p = 0.012). There was high sensitivity in EI-950 (%), EI-850 (%) and expiratory MLD in the parenchyma and high sensitivity in LA and Pi10WA in small airways in the differential diagnosis of COPD and asthma. CONCLUSION: To aid the diagnosis, CT can provide quantitative measurements to differentiate between COPD and asthma patients.

Magnetic Resonance Imaging and Pathologic Findings of 26 Cases With Uterine Adenomatoid Tumors.

OBJECTIVE: To analyze the magnetic resonance imaging (MRI) features and pathologic findings of uterine adenomatoid tumors (ATs) for improved diagnostic accuracy and facilitating differential diagnosis of the tumors. METHODS: We investigated retrospectively 26 patients with uterine ATs confirmed by pathology. Before operation, all patients accepted multiple MRI scans, including T1-weighted image (T1WI), T2-weighted image (T2WI), T2WI/spectrally adiabatic inversion recovery, and T1-weighted enhanced imaging. Two radiologists reviewed all the MRI sequences on PACS workstations for all patients to evaluate the location, shape, size, margin, intensity, and enhancement of ATs. RESULTS: All uterine ATs exhibited either single round solid (n = 24) or predominantly cystic (n = 2) masses with either well-defined (n = 23) or ill-defined margin (n = 3). The diameter range of the tumors was 1.0 to 7.0 cm (mean, 3.8 cm). Solid masses were isointensive on T1WI and hypointensive on T2WI with moderate enhancement. The degree of enhancement in solid tumors was either lower than (18/24 [75%]) or equal to (6/24 [25%]) that of the myometrium. Predominantly cystic masses presented as cystic lesions with a little irregular solid nodule inside. The cystic parts were hypointensive on T1WI and hyperintensive on T2WI without enhancement, whereas the solid nodules were isointensive on both T1WI and T2WI with moderate enhancement. A large part of the uterine ATs (69.2% [18/26]) coexisted with other uterine diseases. On pathology, uterine ATs were characterized as gland-like structures with irregular expansion of tubular cavities, which might be correlated with low enhancement of tumors. The tumors were lined with flat or cuboidal mesothelial cells and residue of smooth muscle component, which might contribute to their hypointensive appearance on T2WI. CONCLUSIONS: Small solid uterine masses with homogeneous hypointensity on T2WI and lower enhancement or cystic lesions with inner irregular solid nodule may indicate the diagnosis of uterine ATs, and final diagnosis can be determined with pathology.

[Effect of lung strain on pulmonary edema in dogs with acute respiratory distress syndrome].

OBJECTIVE: To explore the effect of lung strain on pulmonary edema in dogs with acute respiratory distress syndrome (ARDS) preliminarily. METHODS: Twenty dogs were recruited to induct ARDS models by venous injection of oleic acid and grouped into Group S1.0, S1.5, S2.0, S2.5 randomly, the lung strain was previously designated as 1.0, 1.5, 2.0 or 2.5 for each group. The ventilation was used to support the dogs for 24 hours with tidal volume calculated from lung strain, functional residual capacity and the volume recruitment by positive end expiratory pressure. Extra vascular lung water index (EVLWI) and pulmonary vascular permeability index (PVPI) were tested by thermodilution technique and wet to dry ratio of lung tissue was evaluated by weighing method. RESULTS: The tidal volume of Group S1.0, S1.5, S2.0, S2.5 was (7.5 ± 1.7), (13.9 ± 5.8), (17.9 ± 4.7) and (20.3 ± 7.9) ml/kg, respectively. After ventilation for 4 h, EVLWI in Group S2.0 and Group S2.5 were higher than them in Group S1.0 and Group S1.5 ((38.9 ± 6.5), (40.1 ± 7.7) vs (26.6 ± 5.2), (28.8 ± 5.4) ml/kg; F = 5.627, P = 0.009). At the end of study, PVPI of Group S2.5 increased significantly versus Group S1.0, S1.5 and S2.0 (16.7 ± 5.4 vs 7.0 ± 2.3, 9.2 ± 2.2, 11.5 ± 6.0; F = 4.329, P = 0.022). Wet to dry ratios of lung tissue in Group S2.0 and Group S2.5 were significantly higher than Group S1.0 and Group S1.5 (10.4 ± 1.2, 10.8 ± 1.5 vs 8.3 ± 1.2, 8.0 ± 2.2; F = 5.627, P = 0.009) after ventilation. CONCLUSION: Lung strain above 2.0 aggravates pulmonary edema dramatically and the effect might be slight if lung strain was smaller than 1.5.

Correlation of pulmonary function indexes determined by low-dose MDCT with spirometric pulmonary function tests in patients with chronic obstructive pulmonary disease.

OBJECTIVE: The objective of our study was to evaluate the correlation between pulmonary function indexes determined by low-dose MDCT and those obtained from routine spirometric pulmonary function tests (PFTs) in patients with chronic obstructive pulmonary disease (COPD). MATERIALS AND METHODS: Lung function of patients with COPD stages 0-III was evaluated by both MDCT and spirometric PFTs. Scanning was performed at maximum end-inspiration and maximum end-expiration. RESULTS: A very strong correlation was found between extrapolated expiratory lung volume (LVex) and COPD stage (r = 0.802, p < 0.001) and between extrapolated LVex and the ratio of forced expiratory volume in 1 second and percentage forced vital capacity (FEV1/FVC%) (r = -0.831, p < 0.001). Moreover, strong positive correlations were found between inspiratory lung volume (LVin) and total lung capacity (TLC) (r = 0.658, p < 0.001), LVex and residual volume (RV) (r = 0.683, p < 0.001), extrapolated LVex and RV (r = 0.640, p < 0.001), LVex and RV/TLC (r = 0.602, p < 0.001), LVex/LVin and RV/TLC (r = 0.622, p < 0.001), extrapolated LVex and RV/TLC (r = 0.663, p < 0.001), and LVex and COPD stage (r = 0.697, p < 0.001). CONCLUSION: Low-dose MDCT lung function indexes correlate well with spirometric PFT results, and the highest correlation is at end-expiration. Low-dose MDCT may be useful for evaluating lung function in patients with COPD.

[The clinical and CT two-phase imaging features of bronchial-pulmonary arterial fistula].

OBJECTIVE: To study the clinical features of bronchial-pulmonary arterial fistula, and to analyze its imaging features. METHODS: In continuous five months, 502 patients for pulmonary angiography were analyzed by pulmonary/aortic arterial two-phase scanning. The 128-slice MSCT (Siemens Definition AS 128) was used with the following parameters: the speed of 0.5 s/weeks, the collimator width of 64 × 0.6 mm, the pitch of 0.9, the tube voltage of 120 kV, the contrast agent of 300 mg/ml (1.2 ml/kg) , and the flow rate of 4.3 ml/s. Automatic trigger technology was used, while the threshold of the main pulmonary artery trunk was set to 80 HU. After 4 s delay, the pulmonary-arterial phase was scanned for 3-5 s. Then, the aortic-arterial phase was taken after 12 s. Finally, the clinical features and CTA two-phase images were analyzed by two radiologists, respectively. The diagnozied criteria of CTA images for bronchial-pulmonary arterial fistula were as following.In pulmonary/aortic arterial two-phase scanning, pulmonary artery or aortic artery could be displayed, respectively. The filling defect of fistula's pulmonary artery was observed in pulmonary arterial phase. However, the filling defect of fistula's pulmonary artery had significant filling in aortic arterial phase, with the similar density intensity of aortic artery.In addition, the thicken bronchial artery were observed in the fistula area. CLINICAL FEATURES: In all 502 patients, 65 positive cases of the bronchial-pulmonary arterial fistula included 37 male cases and 28 female cases with ages from 45 to 83 years (69 ± 11). The clinical symptoms included hemoptysis (32%), anhelation (69%), hypoxia (66%), the raise of D2 dimer (70%), and pulmonary hypertension (64%). CTA two-phase images features: In the pulmonary-arterial phase, the intensity difference of pulmonary/aortic was [322 ± 122 (100-751)] HU. The local filling defect in the proximal pulmonary artery (12%) and the filling defect in the whole pulmonary artery (88%) were observed in 65 positive cases.In the aortic-arterial phase, the intensity difference of pulmonary/aortic was [251 ± 89 (85-428)] HU. The local enhancement in the proximal pulmonary artery (24%) and the enhancement in the whole pulmonary artery (76%) were observed in 58 positive cases. The visible thicken bronchial artery were observed in the fistula area of all cases.In 65 cases of bronchial-pulmonary arterial fistula, the fistula lesions contained 56 cases of lung lesions (including 35 cases of honeycomb lung, 16 cases of atelectasis, and 3 cases of chronic mass-like pneumonia) and 9 cases of vascular lesions (including 4 cases of chronic pulmonary artery embolism, 3 cases of congenital vascular malformation, 1 case of pulmonary arthritis, and 1 case of pulmonary artery aneurysm). 437 cases of non bronchial-pulmonary arterial fistula had 4 cases of of vascular lesions and 76 cases of lung lesions. There were significant statistic difference between the fistula and vascular lesions or lung lesions (the value of χ(2): 37.51 or 165.11, all values of P < 0.001). CONCLUSION: The disease of bronchial-pulmonary arterial fistula usually occurred in the chronic pneumonia and the pulmonary vascular lesions. The CT pulmonary/aortic arterial two-phase scanning could detect the homodynamic changes to diagnosis this disease correctly. The pulmonary embolism need be differentiated.

Contrast-enhanced CT and PET-CT characteristics of primary tracheal lymphoepithelioma-like carcinoma: case series.

Background: Primary tracheal lymphoepithelioma-like carcinoma (LELC) is extremely rare, with only a few cases reported so far, and few studies have focused on the radiological features. This study aimed to investigate contrast-enhanced computed tomography (CECT) and positron emission tomography-computed tomography (PET-CT) presentations of primary tracheal LELC to improve diagnosis. Methods: A retrospective analysis was conducted on the clinical and imaging data of 13 patients with confirmed primary tracheal LELC between December 2013 and August 2022. We analyzed the radiological profiles of lesions on the CECT and PET-CT images. Results: In 92.3% (12/13) of the cases, primary tracheal LELC lesions predominantly occurred in the thoracic segment. They manifested as singular, wide-based, eccentric, irregular nodules, or exhibited mass-like thickening of the tracheal wall with invasive growth both internally and externally along the wall. The thickest dimension of the lesion ranged from 9 to 28 mm, affecting a length of 30.8±13.5 mm. Luminal stenosis was evident in all patients, with the narrowest point reaching a stenosis rate of 85%. Lesion margins were clear in 69.2% (9/13), indistinct in 23.1% (3/13), and unclear in 7.7% (1/13) of all cases. Among the patients, 92.3% (12/13) exhibited a relatively uniform density on CT plain scans, with a CT value of 44.5±7.8 Hounsfield units (HU). Enhancement scans revealed moderate to marked enhancement in 75% (9/12) of cases. In 2 cases undergoing PET-CT examination, lesion standardized uptake values (SUVs) were 4.4 and 5.1, whereas enlarged lymph node SUVs were 7.7 and 6.3, respectively. Mediastinal lymph node enlargement was observed in 8 patients (61.5%, 8/13), with a maximum short axis of 11.1±5.5 mm. After treatment, 9 out of 12 patients (75%) showed no evidence of distant metastasis upon CT re-examination. Conclusions: Early detection of primary tracheal LELC allows for curative resection and may lead to a favorable prognosis. It presents with characteristic CT findings, and the utilization of PET-CT improves diagnosis and staging.

Lung stress and strain during mechanical ventilation in animals with and without pulmonary acute respiratory distress syndrome.

BACKGROUND: Mechanical ventilation is a life-saving therapy for acute respiratory distress syndrome (ARDS). However, unphysiological lung stress (transpulmonary pressure) and strain (the ratio between inflated volume and functional residual capacity) can induce lung injury. Unfortunately, both stress and strain are not measured directly because of technical limitations but predicted from airway plateau pressure (Pplat) and tidal volume (Vt). Recently, some literatures indicated that Pplat and Vt cannot be good surrogates without distinguishing pulmonary ARDS patients (ARDSp) from extrapulmonary ARDS patients. Analyzing them together might distort the truth. Thus, we established animal models of ARDSp to explore whether lung stress and strain can be surrogated precisely by Pplat and Vt. METHODS: Fifteen Beagle dogs were recruited to establish ARDSp models by injection of oleic acids. Esophageal manometry was performed to estimate pleural pressure and lung stress. Functional residual capacity was obtained through computed tomography. In the first stage, Vt was set at 10mL/kg body weight in healthy and injured lungs and then adjusted to achieve lung strain as 1.0, 1.5, and 2.0 in sequence. RESULTS: There was a good linear relationship between lung stress and Pplat in healthy and ARDS lungs (P<0.001). For a given Vt (10mL/kg body weight), the global lung strains varied from 0.197 to 0.416 and 0.467 to 0.715 in healthy individuals and different ARDS patients, respectively. On the contrary, Vt varied remarkably for a given lung strain. CONCLUSIONS: Pplat is an adequate surrogate for lung stress, but Vt cannot represent lung strain sufficiently.

Semi-Supervised Segmentation of Interstitial Lung Disease Patterns from CT Images via Self-Training with Selective Re-Training.

Accurate segmentation of interstitial lung disease (ILD) patterns from computed tomography (CT) images is an essential prerequisite to treatment and follow-up. However, it is highly time-consuming for radiologists to pixel-by-pixel segment ILD patterns from CT scans with hundreds of slices. Consequently, it is hard to obtain large amounts of well-annotated data, which poses a huge challenge for data-driven deep learning-based methods. To alleviate this problem, we propose an end-to-end semi-supervised learning framework for the segmentation of ILD patterns (ESSegILD) from CT images via self-training with selective re-training. The proposed ESSegILD model is trained using a large CT dataset with slice-wise sparse annotations, i.e., only labeling a few slices in each CT volume with ILD patterns. Specifically, we adopt a popular semi-supervised framework, i.e., Mean-Teacher, that consists of a teacher model and a student model and uses consistency regularization to encourage consistent outputs from the two models under different perturbations. Furthermore, we propose introducing the latest self-training technique with a selective re-training strategy to select reliable pseudo-labels generated by the teacher model, which are used to expand training samples to promote the student model during iterative training. By leveraging consistency regularization and self-training with selective re-training, our proposed ESSegILD can effectively utilize unlabeled data from a partially annotated dataset to progressively improve the segmentation performance. Experiments are conducted on a dataset of 67 pneumonia patients with incomplete annotations containing over 11,000 CT images with eight different lung patterns of ILDs, with the results indicating that our proposed method is superior to the state-of-the-art methods.

Accurate classification of pulmonary nodules by a combined model of clinical, imaging, and cell-free DNA methylation biomarkers: a model development and external validation study.

BACKGROUND: There is an unmet clinical need for accurate non-invasive tests to facilitate the early diagnosis of lung cancer. We propose a combined model of clinical, imaging, and cell-free DNA methylation biomarkers that aims to improve the classification of pulmonary nodules. METHODS: We conducted a prospective specimen collection and retrospective masked evaluation study. We recruited participants with a solitary pulmonary nodule sized 5-30 mm from 24 hospitals across 20 cities in China. Participants who were aged 18 years or older and had been referred with 5-30 mm non-calcified and solitary pulmonary nodules, including solid nodules, part solid nodules, and pure ground-glass nodules, were included. We developed a combined clinical and imaging biomarkers (CIBM) model by machine learning for the classification of malignant and benign pulmonary nodules in a cohort (n=839) and validated it in two cohorts (n=258 in the first cohort and n=283 in the second cohort). We then integrated the CIBM model with our previously established circulating tumour DNA methylation model (PulmoSeek) to create a new combined model, PulmoSeek Plus (n=258), and verified it in an independent cohort (n=283). The clinical utility of the models was evaluated using decision curve analysis. A low cutoff (0·65) for high sensitivity and a high cutoff (0·89) for high specificity were applied simultaneously to stratify pulmonary nodules into low-risk, medium-risk, and high-risk groups. The primary outcome was the diagnostic performance of the CIBM, PulmoSeek, and PulmoSeek Plus models. Participants in this study were drawn from two prospective clinical studies that were registered (NCT03181490 and NCT03651986), the first of which was completed, and the second of which is ongoing because 25% of participants have not yet finished the required 3-year follow-up. FINDINGS: We recruited a total of 1380 participants. 1097 participants were enrolled from July 7, 2017, to Feb 12, 2019; 839 participants were used for the CIBM model training set, and the rest (n=258) for the first CIBM validation set and the PulmoSeek Plus training set. 283 participants were enrolled from Oct 26, 2018, to March 20, 2020, as an independent validation set for the PulmoSeek Plus model and the second validation set for the CIBM model. The CIBM model validation cohorts had area under the curves (AUCs) of 0·85 (95% CI 0·80-0·89) and 0·85 (0·81-0·89). The PulmoSeek Plus model had better discrimination capacity compared with the CIBM and PulmoSeek models with an increase of 0·05 in AUC (PulmoSeek Plus vs CIBM, 95% CI 0·022-0·087, p=0·001; and PulmoSeek Plus vs PulmoSeek, 0·018-0·083, p=0·002). The overall sensitivity of the PulmoSeek Plus model was 0·98 (0·97-0·99) at a fixed specificity of 0·50 for ruling out lung cancer. A high sensitivity of 0·98 (0·96-0·99) was maintained in early-stage lung cancer (stages 0 and I) and 0·99 (0·96-1·00) in 5-10 mm nodules. The decision curve showed that if an invasive intervention, such as surgical resection or biopsy, was deemed necessary at more than the risk threshold score of 0·54, the PulmoSeek Plus model would provide a standardised net benefit of 82·38% (76·06-86·79%), equivalent to correctly identifying approximately 83 of 100 people with lung cancer. Using the PulmoSeek Plus model to classify pulmonary nodules with two cutoffs (0·65 and 0·89) would have reduced 89% (105/118) of unnecessary surgeries and 73% (308/423) of delayed treatments. INTERPRETATION: The PulmoSeek Plus Model combining clinical, imaging, and cell-free DNA methylation biomarkers aids the early diagnosis of pulmonary nodules, with potential application in clinical decision making for the management of pulmonary nodules. FUNDING: The China National Science Foundation, the Key Project of Guangzhou Scientific Research Project, the High-Level University Construction Project of Guangzhou Medical University, the National Key Research & Development Programme, the Guangdong High Level Hospital Construction "Reaching Peak" Plan, the Guangdong Basic and Applied Basic Research Foundation, the National Natural Science Foundation of China, The Leading Projects of Guangzhou Municipal Health Sciences Foundation, the Key Research and Development Plan of Shaanxi Province of China, the Scheme of Guangzhou Economic and Technological Development District for Leading Talents in Innovation and Entrepreneurship, the Scheme of Guangzhou for Leading Talents in Innovation and Entrepreneurship, the Scheme of Guangzhou for Leading Team in Innovation, the Guangzhou Development Zone International Science and Technology Cooperation Project, and the Science and Technology Planning Project of Guangzhou.

Computed tomographic and pathological findings of dermatofibrosarcoma protuberans.

OBJECTIVE: The aim of the study was to investigate the computed tomographic (CT) and pathological findings of dermatofibrosarcoma protuberans to improve the awareness and understanding of the tumors from aspect of CT images. METHODS: The CT findings of 16 cases (17 tumors) with dermatofibrosarcoma protuberans confirmed by pathological findings were retrospectively selected. Fourteen cases were primary dermatofibrosarcoma protuberans, 2 cases were recurrent tumors. Thirteen patients had CT plain and enhanced scans, 1 patient had direct enhanced CT scan, 2 patients had only unenhanced scan. Images of the tumors were analyzed and compared with pathological results. RESULTS: Of the 16 cases (17 tumors total), 9 cases were on the trunk, 7 cases were on the head and the neck; 15 cases appeared as solitary isohypodense, ovoid, or round mass at the cutaneous and subcutaneous tissue, 1 case demonstrated 2 isodense masses on unenhanced CT images. The mean diameter of tumors was 4.0 cm, and the depth was 1.7 cm. The margin was well defined (n = 15 [88.2%]) or ill defined (n = 2 [11.8%]). Fifteen tumors revealed moderate or marked homogeneous (n = 12 [80%], smaller lesion, diameter <5 cm) or heterogeneous (n = 3 [20%], larger lesion, diameter ≥5 cm) enhancement on enhanced CT with intratumoral nonenhancement areas, which indicated intratumoral necrotic and cystic degeneration areas. No calcifications and metastasis were found. The histological examinations revealed large amounts of uniform spindle cells, which were arranged in "storiform" pattern. Immunohistochemical analysis revealed samples positive for CD34 and vimentin. CONCLUSION: The common imaging findings of dermatofibrosarcoma protuberans include a solitary, superficial, subcutaneous solid mass, various homogenous or heterogeneous enhancements due to degenerative areas. Computed tomographic scan is helpful to detect the size, location, depth and range of this tumor.

Retrospective analysis of 76 immunocompetent patients with primary pulmonary cryptococcosis.

BACKGROUND: Pulmonary cryptococcosis typically occurs in immunocompromised patients, but it can also occur in immunocompetent patients. Our objective was to describe the clinical manifestations, diagnosis, and management of primary pulmonary cryptococcosis in immunocompetent patients. METHODS: We retrospectively reviewed the clinical data of 76 patients with primary pulmonary cryptococcosis who were admitted to our hospital from 1995 to 2010. RESULTS: Pulmonary cryptococcosis was pathologically proven in all patients. Mean patient age was 42.5 years and 55 patients (72%) were male. The major clinical manifestations were cough (47 pts, 62%), expectoration (29 pts, 38%), fever (16 pts, 21%), chest pain (15 pts, 20%), dyspnea (17 pts, 22%), and emaciation (10 pts, 13%). Eighteen patients (24%) were asymptomatic. Most patients were admitted due to shadows on chest X-rays. Lesions were more common in the lower lung (60 pts, 78.9%) than in the upper lung (25 pts, 32.9%). More lesions (28 pts, 37%) were characterized by patchy consolidations. Pulmonary cryptococcosis was confirmed histologically among all patients. Surgical removal of lesions or treatment with fluconazole and other antifungal agents for complete courses led to favorable outcomes for most patients. CONCLUSIONS: Primary pulmonary cryptococcosis was found mainly in immunocompetent patients aged <50 years without preexisting lung disease. Shadow on the chest X-ray is the predominant feature. Treatment with a complete course of fluconazole and/or other antifungal agents can achieve favorable outcome.

Association of FAS and FAS ligand polymorphisms with the susceptibility and severity of lumbar disc degeneration in Chinese Han population.

CONTEXT: Apoptosis is involved in the mechanism of lumbar disc degeneration (LDD). OBJECTIVE: We aim to determine whether the polymorphisms of FAS and FASL are associated with the presence and severity of LDD. METHODS: A total of 348 patients with LDD and 215 healthy controls were genotyped. RESULTS: Patients with LDD showed higher frequency of-1377GA and AA, as well as-844CT and TT genotypes than normal controls. These genotypes were found to be associated with the risk of higher grades of LDD. CONCLUSION: The polymorphisms of FAS and FASL may be associated with the presence and severity of LDD.

An Exploratory Study on the Stable Radiomics Features of Metastatic Small Pulmonary Nodules in Colorectal Cancer Patients.

OBJECTIVES: To identify the relatively invariable radiomics features as essential characteristics during the growth process of metastatic pulmonary nodules with a diameter of 1 cm or smaller from colorectal cancer (CRC). METHODS: Three hundred and twenty lung nodules were enrolled in this study (200 CRC metastatic nodules in the training cohort, 60 benign nodules in the verification cohort 1, 60 CRC metastatic nodules in the verification cohort 2). All the nodules were divided into four groups according to the maximum diameter: 0 to 0.25 cm, 0.26 to 0.50 cm, 0.51 to 0.75 cm, 0.76 to 1.0 cm. These pulmonary nodules were manually outlined in computed tomography (CT) images with ITK-SNAP software, and 1724 radiomics features were extracted. Kruskal-Wallis test was performed to compare the four different levels of nodules. Cross-validation was used to verify the results. The Spearman rank correlation coefficient is calculated to evaluate the correlation between features. RESULTS: In training cohort, 90 features remained stable during the growth process of metastasis nodules. In verification cohort 1, 293 features remained stable during the growth process of benign nodules. In verification cohort 2, 118 features remained stable during the growth process of metastasis nodules. It is concluded that 20 features remained stable in metastatic nodules (training cohort and verification cohort 2) but not stable in benign nodules (verification cohort 1). Through the cross-validation (n=100), 11 features remained stable more than 90 times. CONCLUSIONS: This study suggests that a small number of radiomics features from CRC metastatic pulmonary nodules remain relatively stable from small to large, and they do not remain stable in benign nodules. These stable features may reflect the essential characteristics of metastatic nodules and become a valuable point for identifying metastatic pulmonary nodules from benign nodules.

Retrospective comparison of high-resolution computed tomography of eosinophilic granulomatosis with polyangiitis with severe asthma.

BACKGROUND: Eosinophilic granulomatosis with polyangiitis (EGPA) is often misdiagnosed as severe asthma due to their similar clinical presentations. We compared the pulmonary radiologic features of EGPA to those of severe asthma by high-resolution computed tomography (HRCT) in order to early diagnose EGPA. METHODS: We retrospectively reviewed clinical records and HRCT findings of 96 patients with EGPA and 82 patients with severe asthma who were seen at our hospital from 2011 to 2017. We used a semi-quantitative grading system to evaluate radiological findings. A radiological only and a clinical-radiological model were used to differentiate EGPA from severe asthma. RESULTS: Bronchial wall thickening, air trapping, tree-in-bud opacities, bronchial mucus plugging, bronchiectasis, diffuse ground-glass opacities (GGOs), consolidation, and increased small vascular markings were more common in EGPA patients than in severe asthmatics (P<0.05). The gradings of GGO (grade 2 vs. grade 1) and tree-in-bud opacities (grade 2 vs. grade 0) were higher in EGPA patients than in severe asthmatics. The total image score of EGPA patients was significantly higher than that of severe asthmatics (P<0.05). In the radiological only and the clinical-radiological model, the area under the receiver operating characteristic (ROC) curves (AUCs) for the identification of EGPA and severe asthma were 0.904 [95% confidence interval (CI): 0.860 to 0.948] and 0.974 (95% CI: 0.955 to 0.993), respectively. CONCLUSIONS: Lung HRCT scan is useful in differentiating EGPA from severe asthma. In patients with difficult-to-treat asthma, an HRCT scan of the thorax should be performed should there be features that raise the suspicion of EGPA.