Guangzhou Chest Hospital

This study aimed to develop and validate stratified machine learning models for early prediction of anti-tuberculosis drug-induced liver injury (ATB-DILI) risk, targeting both a general tuberculosis (TB) treatment population and a high-risk subgroup with chronic hepatitis B (CHB) co-infection, based on real-world clinical data. A single-center retrospective cohort study was conducted using data from 11,361TB patients (3,787 ATB-DILI cases and 7,574 controls) and a CHB subgroup of 1,017 patients (339 cases and 678 controls) after propensity score matching. Ten machine learning algorithms, including Logistic Regression, Random Forest (RF), and XGBoost, were applied. Models were trained and validated using a 1:1 split and 10-fold cross-validation. Performance was evaluated using AUC, accuracy, sensitivity, specificity, precision, and F1-score. Model interpretability was enhanced using SHapley Additive exPlanations (SHAP). In the overall population, ensemble methods such as RF and XGBoost achieved AUCs of 0.960 and 0.954, respectively, on the validation set. In the CHB subgroup, RF and XGBoost performed even better, with AUCs of 0.994. Key predictors in the general population included ALT, eosinophil count, AST, and procalcitonin, while in the CHB subgroup, total bile acid, procalcitonin, ALP, and prealbumin were most influential. SHAP analysis revealed non-linear relationships between features and ATB-DILI risk, aligning with clinical knowledge. Stratified machine learning models, particularly ensemble methods, demonstrated excellent performance in predicting ATB-DILI risk and highlighted distinct injury mechanisms between general and CHB co-infected TB patients. This approach offers a clinically interpretable and accurate tool for early warning of ATB-DILI, supporting personalized risk assessment and management.

Bronchopleural fistula (BPF) is a common and severe complication in thoracic surgery, characterized by its complex nature and high mortality rate. Bronchial occluder is one of the effective methods for the interventional treatment of BPF under bronchoscopy. This work was to evaluate the adoption and efficacy of individualized bronchial occluder designed with the assistance of 3D printing technology in the treatment of BPF. This single-center, retrospective study included 60 patients (35 males and 25 females, aged 40-78 years, without severe dysfunction of major organs or history of mental disorders) diagnosed with BPF between January 2020 and December 2023, who received different treatment and were divided into the Control Group (CG, receiving thoracotomy treatment), Therapy group-I (TG-I, receiving septal occluder treatment), and Therapy group-II (TG-II, receiving treatment with 3D-printed individualized bronchial occluder). Serum levels of high-sensitivity C-reactive protein (hs-CRP) and procalcitonin (PCT) were measured by enzyme-linked immunosorbent assay preoperatively and at 1, 4, and 12 weeks postoperatively. Daytime cough scores were assessed according to guidelines, pain scores were evaluated using a visual analog scale, clinical efficacy was determined based on consensus criteria, and complications were recorded. Quantitative variables were compared using one-way ANOVA with Bonferroni correction, while categorical variables were analyzed by chi-square or Fisher's exact test. the overall clinical efficacy was higher in TG-I and TG-II versus CG, with TG-II showing a greatly higher efficacy after twelve weeks of treatment (P < 0.05). Relative to CG, both TG-I and TG-II demonstrated lower serum levels of hs-CRP and PCT, markedly reduced cough and pain symptom scores, and markedly lower complication rates (P < 0.05). Additionally, TG-II resulted in notably lower serum inflammatory markers, cough and pain symptom scores, and complication rates versus TG-I (P < 0.05). The 3D-printed personalized occluder (TG-II) significantly reduced postoperative inflammatory responses and alleviated cough-associated pain, demonstrating superior short-term efficacy and safety compared to conventional treatment (CG) and commercially available occluders (TG-I). This approach offers distinct advantages for the personalized precision treatment of BPF. However, further clinical studies are required to validate its potential as an alternative to surgical repair.

Clinically significant pulmonary abnormalities, like tuberculosis, nodules, and chronic obstructive pulmonary disease (COPD), hinder lung function. As a non-invasive diagnostic tool, radiology can identify various pulmonary conditions. Artificial intelligence (AI) advancements, particularly deep learning-based applications, have improved the efficiency and accuracy of radiological examinations. However, creating accurate AI models is still challenging due to the requirement for extensive manual annotation. To address this, we propose Smart Imagery Framing and Truthing (SIFT) to assist annotators in labeling and delineating pulmonary abnormalities and corresponding boundaries on CT images. Integrating Mask-RCNN, the system is developed on a training dataset of 9,078 CT images with 166,724 slices and 18,367 regions of interest (ROIs) corresponding to 47 different abnormalities/diseases. The proposed system processes an independent testing set with 2,199 CT images containing 40,493 slices and 4,280 ROIs by predicting both abnormality/disease types and corresponding ROI boundary locations. Two senior radiologists (&ge; 30 years of radiological experience) delineated each image of training and testing sets in a double-blinded way, and their intersection area functioned as the ground truth. Evaluation metrics include intersection over union (IOU) value calculated based on the ROI area and the ground truth, the area under the ROC curve (AUC), and sensitivity on the slice-level of abnormality label category. For ROI segmentation, the results showed that 91.5% and 36.2% abnormalities had an IOU over 0.6 and 0.7, respectively. Regarding the metric of AUC of classification, there were 97.9%, 80.9%, and 42.6% of abnormalities whose AUC values were above 0.7, 0.8 and 0.9, respectively. For the sensitivity metric, all 47 label categories exceeded 0.8, and 68.1% (32/47) exceeded 0.9. SIFT demonstrates high performance in determining abnormality/disease types with corresponding boundary locations for the ROIs, which can be used to predict training and testing sets to develop AI technologies.

Although paracoccidioidomycosis is one of the most prevalent endemic mycoses in Latin American countries, where at least 10 million people are infected, the prevalence of paracoccidioidomycosis in China remains unknown because no related case has been reported, and its diagnosis is extremely challenging for local clinicians because of the complexity of disease progression and lack of specific evidence. Here, we report the first case of PCM in a male patient with a long-term latent infection in China. The results of special staining, immunohistochemistry, lymph node biopsy pathology, and metagenomic second-generation sequencing indicated that the patient was infected with Paracoccidioides brasiliensis. In this case, the patient was administered voriconazole 200 mg twice daily. After continuous treatment for 6 months, the patient's symptoms improved significantly, and the medication was discontinued. The outpatient follow-up revealed no discomfort. This case is of great value for the early diagnosis, treatment, and prevention of the spread of this emerging infectious disease in China.

Pulmonary sarcomatoid carcinoma (PSC) is a rare non-small-cell lung cancer with sarcomatous components or sarcomatoid differentiation, high degree of malignancy, and insensitivity to chemotherapy or radiotherapy. The management of PSC coexisting with tuberculosis (TB) poses a greater challenge, as it necessitates concurrent administration of both anti-TB and anti-neoplastic therapies. The efficacy of anti-PD-1 immunotherapy in non-small-cell lung cancer is promising, but its safety in patients with co-existent TB remains uncertain. Here, we describe a case of advanced PSC coexisting with TB, which experienced progression-free survival (PFS) of over 36 months after receiving anti-TB and anti-neoplastic therapy composed of chemotherapy, vascular targeting therapy, and PD-1 inhibitors simultaneously. The patient is still being followed up with a satisfactory quality of life. This paper is focused on the characteristics and treatment of PSC and discuss the clinical strategies of lung cancer coexisting with TB.