Search

Purpose Despite the importance of radial endobronchial ultrasound (rEBUS) in transbronchial biopsy, researchers have yet to apply artificial intelligence to the analysis of rEBUS images.Materials and methods This study developed a convolutional neural network (CNN) to differentiate between malignant and benign tumours in rEBUS images. This study retrospectively collected rEBUS images from medical centres in Taiwan, including 769 from National Taiwan University Hospital Hsin-Chu Branch, Hsinchu Hospital for model training (615 images) and internal validation (154 images) as well as 300 from National Taiwan University Hospital (NTUH-TPE) and 92 images were obtained from National Taiwan University Hospital Hsin-Chu Branch, Biomedical Park Hospital (NTUH-BIO) for external validation. Further assessments of the model were performed using image augmentation in the training phase and test-time augmentation (TTA).Results Using the internal validation dataset, the results were as follows: area under the curve (AUC) (0.88 (95% CI 0.83 to 0.92)), sensitivity (0.80 (95% CI 0.73 to 0.88)), specificity (0.75 (95% CI 0.66 to 0.83)). Using the NTUH-TPE external validation dataset, the results were as follows: AUC (0.76 (95% CI 0.71 to 0.80)), sensitivity (0.58 (95% CI 0.50 to 0.65)), specificity (0.92 (95% CI 0.88 to 0.97)). Using the NTUH-BIO external validation dataset, the results were as follows: AUC (0.72 (95% CI 0.64 to 0.82)), sensitivity (0.71 (95% CI 0.55 to 0.86)), specificity (0.76 (95% CI 0.64 to 0.87)). After fine-tuning, the AUC values for the external validation cohorts were as follows: NTUH-TPE (0.78) and NTUH-BIO (0.82). Our findings also demonstrated the feasibility of the model in differentiating between lung cancer subtypes, as indicated by the following AUC values: adenocarcinoma (0.70; 95% CI 0.64 to 0.76), squamous cell carcinoma (0.64; 95% CI 0.54 to 0.74) and small cell lung cancer (0.52; 95% CI 0.32 to 0.72).Conclusions Our results demonstrate the feasibility of the proposed CNN-based algorithm in differentiating between malignant and benign lesions in rEBUS images.

Abstract Background Timely differentiating between pulmonary tuberculosis (TB) and nontuberculous mycobacterial lung disease (NTM-LD), which are radiographically similar, is important because infectiousness and treatment differ. This study aimed to evaluate whether artificial intelligence could distinguish between TB or NTM-LD patients by chest X-rays (CXRs) from suspects of mycobacterial lung disease. Methods A total of 1500 CXRs, including 500 each from patients with pulmonary TB, NTM-LD, and patients with clinical suspicion but negative mycobacterial culture (Imitator) from two hospitals, were retrospectively collected and evaluated in this study. We developed a deep neural network (DNN) and evaluated model performance using the area under the receiver operating characteristic curves (AUC) in both internal and external test sets. Furthermore, we conducted a reader study and tested our model under three scenarios of different mycobacteria prevalence. Results Among the internal and external test sets, the AUCs of our DNN model were 0.83 ± 0.005 and 0.76 ± 0.006 for pulmonary TB, 0.86 ± 0.006 and 0.64 ± 0.017 for NTM-LD, and 0.77 ± 0.007 and 0.74 ± 0.005 for Imitator. The DNN model showed higher performance on the internal test set in classification accuracy (66.5 ± 2.5%) than senior (50.8 ± 3.0%, p

Introduction: Nontuberculous mycobacteria (NTM) may be present in the respiratory tract of patients with lung cancer. We investigated the association of pulmonary NTM with the clinical features and outcomes of patients with lung cancer. Methods: Between 2015 and 2019, the data of patients diagnosed with lung cancer at a medical center in northern Taiwan were analyzed. Patients whose respiratory specimens were culture-positive for NTM were identified (NTM group). For each patient in the NTM group, a matched control was selected (control group). The survival of the two groups was compared using the Kaplan–Meier method and Cox proportional hazards regression analysis. Results: Among 8718 patients with lung cancer, 5418 (62.1%) underwent a sputum mycobacterial culture. At least one NTM species was isolated from 138 (2.5%) patients. The median age was 72 years (range: 64–80). In the NTM group, 19.8% fulfilled both the microbiological and radiographic criteria for the diagnosis of NTM lung disease. Compared with the control group, the NTM group exhibited a lower body mass index (22.4 vs. 23.6, p = 0.025) and a higher prevalence of structural lung disease (38.9% vs. 22.2%, p = 0.004). The two-year survival was not significantly different between the two groups (hazard ratio [HR]: 1.110; 95% confidence interval [CI]: 0.702–1.754, p = 0.656). In patients receiving chemotherapy, pulmonary NTM was associated with worse survival (HR: 2.497, 95% CI: 1.262–4.943, p = 0.009). Conclusions: Except in patients receiving chemotherapy, pulmonary NTM may not be clinically relevant in patients with lung cancer.

Abstract Background To investigate the risk factors of pancreatitis after endoscopic retrograde cholangiopancreatography (ERCP) in patients with biliary tract diseases. Methods We retrospectively analyzed the clinical data of 480 patients who underwent ERCP for biliary tract diseases at the Affiliated Zhongshan Hospital of Dalian University from October 2011 to October 2016. The patients were divided into a study group (n = 75, with PEP) and a control group (n = 405, without PEP) based on whether they developed post-ERCP pancreatitis (PEP), and their clinical baseline data and intraoperative conditions were retrieved and compared. Then, factors associated with PEP were analyzed using logistic regression model, based on which a nomogram prediction model was constructed. The receiver operating characteristic (ROC) curve and calibration curve were used to evaluate the performance of the prediction model. Results Significant differences in age, sex, history of pancreatitis, history of choledocholithiasis, pancreatic duct imaging, pancreatic sphincterotomy, difficult cannulation, multiple cannulation attempts and juxtapapillary duodenal diverticula were observed between the two groups. Multivariate logistic regression analysis showed that age less than 60 years (OR, 0.477; 95% CI, 0.26–0.855), female sex (OR, 2.162; 95% CI, 1.220–3.831), history of pancreatitis (OR, 2.567; 95% CI, 1.218–5.410), history of choledocholithiasis (OR, 2.062; 95% CI, 1.162–3.658), pancreatic sphincterotomy (OR, 2.387; 95% CI, 1.298–4.390), pancreatic duct imaging (OR, 4.429; 95% CI, 1.481–13.242), multiple cannulation attempts (OR, 2.327; 95% CI, 1.205–4.493), difficult cannulation (OR, 2.421; 95% CI, 1.143–5.128), and JPD (OR, 2.002; 95% CI, 1.125–3.564) were independent risk factors for PEP. The nomogram for predicting the occurrence of PEP demonstrated an area under the ROC curve (AUC) of 0.787, and the calibration curves of the model showed good consistency between the predicted and actual probability of PEP. Conclusion Our results showed that age less than 60 years, female sex, history of pancreatitis, history of choledocholithiasis, pancreatic sphincterotomy, pancreatic duct imaging, multiple cannulation attempts, difficult cannulation and juxtapapillary duodenal diverticula were independent risk factors for PEP. In addition, the established nomogram demonstrated promising clinical efficacy in predicting PEP risk in patients who underwent ERCP for biliary tract diseases.

AIM: To investigate the risk and protective factors associated with the retinal nerve fiber layer defect (RNFLD) in a Chinese adult population. METHODS: This study was a cross-sectional population-based investigation including employees and retirees of a coal mining company in Kailuan City, Hebei Province. All the study participants underwent a comprehensive systemic and ophthalmic examination. RNFLD was diagnosed on fundus photographs. Binary logistic regression was used to investigate the risk and protective factors associated with the RNFLD. RESULTS: The community-based study included 14 440 participants. There were 10 473 participants in our study, including 7120 males (68.0%) and 3353 females (32.0%). The age range was 45-108y, averaging 59.56±8.66y. Totally 568 participants had RNFLD and the prevalence rate was 5.42%. A higher prevalence of RNFLD was associated with older age [P

The accumulation of the intermediate zeaxanthin and canthaxanthin in the astaxanthin biosynthesis pathway catalyzed by β-carotene hydroxylase (crtZ) and β-carotene ketolase (crtW) decreases the content of the astaxanthin. Here, we exploited directed evolution of the fusion of crtZ and crtW for improving astaxanthin biosynthesis in Saccharomyces cerevisiae. The results demonstrated that the fusion enzyme of crtZ-crtW with 2 X GGGGS peptides linker can effectively reduce the accumulation of intermediates and improves the content of astaxanthin. Compared with the control strain, the fusion enzyme of ketase and hydroxylase reduced zeaxanthin and canthaxanthin by 7 and 14 times and increased astaxanthin by 1.6 times, respectively. Moreover, 9 variant fusion mutants with improved astaxanthin production were generated through directed evolution. Combining these dominant mutants generated a variant, L95S + I206L, which increased the astaxanthin content of 3.8 times than the control strain. The AlphaFold2 assisted structural analysis indicated that these two mutations alter the interaction between the substrate and the enzymes pocket. Our research provided an efficient idea to reduce the accumulation of the intermediate products in complex biosynthesis pathway.

Background and purpose: Urinary tract infections (UTIs) are the most common bacterial infection in young children. This study aimed to formulate nomogram plots for clinicians to predict UTIs in children aged

Corneal transplantation is the most effective clinical treatment for corneal defects, but it requires precise size of donor corneas, surgical sutures, and overcoming other technical challenges. Postoperative patients may suffer graft rejection and complications caused by sutures. Ophthalmic glues that can long-term integrate with the corneal tissue and effectively repair the focal corneal damage are highly desirable. Herein, a hybrid hydrogel consisting of porcine decellularized corneal stroma matrix (pDCSM) and methacrylated hyaluronic acid (HAMA) was developed through a non-competitive dual-crosslinking process. It can be directly filled into corneal defects with various shapes. More importantly, through formation of interpenetrating network and stable amide bonds between the hydrogel and adjacent tissue, the hydrogel manifested excellent adhesion properties to achieve suture-free repair. Meanwhile, the hybrid hydrogel not only preserved bioactive components from pDCSM, but also exhibited cornea-matching transparency, low swelling ratio, slow degradation, and enhanced mechanical properties, which was capable of withstanding superhigh intraocular pressure. The combinatorial hydrogel greatly improved the poor cell adhesion performance of HAMA, supported the viability, proliferation of corneal cells, and preservation of keratocyte phenotype. In a rabbit corneal stromal defect model, the experimental eyes treated with the hybrid hydrogel remained transparent and adhered intimately to the stroma bed with long-term retention, accelerated corneal re-epithelialization and wound healing. Giving the advantages of high bioactivity, low-cost, and good practicality, the dual-crosslinked hybrid hydrogel served effectively for long-term suture-free treatment and tissue regeneration after corneal defect.