The First Hospital of Kunming

Fetal hypoxia is a leading cause of neonatal morbidity and mortality. Cardiotocography (CTG) is widely used to predict fetal hypoxia during labor, but its interpretation remains suboptimal. Artificial intelligence (AI) models have been developed for CTG interpretation, but their clinical utility is limited by two major challenges: demonstrating superiority over human experts and ensuring explainability in real-world settings. A large dataset containing CTG traces from three tertiary hospitals between January 2014 and May 2022 was built for model development. Deep learning architectures, named Cardiotocography Artificial-intelligence Predictors (CAPs), were trained to predict fetal hypoxia from CTG traces based on CNN (CAP-C), Transformer (CAP-T), LSTM (CAP-L), and CfC (CAP-CfC) algorithms. The outcome was fetal hypoxia, determined by either low Apgar score (≤ 7 at 1 or 5 min) or umbilical artery acidemia (grade 1: pH of umbilical artery (pHa) < 7.20; grade 2: pHa < 7.15; grade 3: pHa < 7.10). Model performance was determined by area under the receiver operating characteristic curve (AUROC), evaluated through nationwide AI-human comparison and validated on the CTU-UHB dataset. Gradient-weighted class activation mapping (Grad-CAM) was applied to highlight the CTG regions that contributed most to the model's predictions. A total of 20,780 CTG traces were obtained for model development, and 467 cases were held out for the nationwide AI-human comparison. Among all models, CAP-L achieved highest AUROC in predicting fetal hypoxia (grade 1: 0.758, 95% CI: 0.754-0.761; grade 2: 0.770, 95% CI: 0.764-0.776; grade 3: 0.716, 95% CI: 0.700-0.732). In comparison with 10,571 expert responses, all CAP models achieved higher AUROC (0.757-0.789 vs. 0.715, P values in Delong test < 0.05). On the public CTU-UHB dataset, CAP-L achieved AUROC of 0.709, 0.727, and 0.730 in predicting fetal hypoxia with grade 1, 2, and 3 acidemia. Grad-CAM analysis showed that the CAP models leveraged variable and prolonged decelerations to predict fetal hypoxia, verified by perturbation-based faithfulness test. The CAP algorithms developed in this study showed superior performance in detecting fetal hypoxia from CTG traces compared to human experts, and demonstrated promising explainability, supporting clinical CTG interpretation. Clinical trial registration number: ChiCTR2100045316, ChiCTR2100052695, ChiCTR2400085338.

Low-fiber diets are a known risk factor for colorectal cancer (CRC). However, the burden of CRC associated with low-fiber intake across regions and age groups remains unclear. This study assesses the global, regional, and national burden of CRC due to low-fiber diets from 1990 to 2021. Using data from the Global Burden of Disease Study 2021, we applied comparative risk models to estimate CRC mortality and disability-adjusted life years associated with low-fiber diets across regions, age groups, and countries. The analysis showed significant disparities in CRC burden due to low-fiber diets. Low- and middle-income regions, particularly populations over 50 years old, bore the highest burden. The global burden has increased over time, particularly in regions undergoing dietary transitions. Targeted dietary interventions to increase fiber intake are essential to reduce the global CRC burden. Policymakers should focus on high-risk regions and populations to mitigate this preventable health issue.

Endoscopic retrograde appendicitis therapy (ERAT) and its visually enhanced technique (Endoscopic direct appendicitis therapy, EDAT) represent a novel minimally invasive approach, offering advantages such as high-definition real-time visualization and precise intervention. This article reports a case of a 49-year-old male with acute appendicitis accompanied by an early abscess. CT findings suggested appendicitis with abscess formation, while EDAT exploration confirmed perforation at the appendiceal tip along with adjacent fecalith impaction, further verified by contrast imaging. During the procedure, the fecalith was extracted via EDAT, followed by irrigation of the purulent cavity and placement of a modified 7 Fr × 7 cm pancreatic stent for drainage. The patient experienced immediate symptom relief postoperatively, with serial follow-up CT scans on the procedure day, 3rd, and 7th postoperative days demonstrating gradual abscess resolution and normalized inflammatory markers, leading to discharge on day 3. A subsequent 3-month telephone follow-up revealed no recurrence of abdominal pain. This case illustrates that EDAT/ERAT can provide integrated diagnosis and definitive management for complicated appendicitis, avoiding the trauma associated with surgical intervention and offering valuable clinical insights.

BackgroundChronic hepatitis B (CHB) with positive HBeAg status constitutes a significant contributor to the development of liver cirrhosis and hepatocellular carcinoma. Pegylated interferon-alpha (pegIFNα) is a common treatment, but its response rate remains limited, and the underlying mechanisms are not fully understood.MethodsEighty-two HBeAg-positive CHB patients were enrolled. miR-194-5p expression, HBeAg, and HBV DNA levels were detected using qRT-PCR, ELISA, and quantitative PCR, respectively. ROC and logistic regression analyses were performed. Cellular experiments, including dual-luciferase reporter and rescue assays, along with Western blot analysis of JAK-STAT pathway proteins, were conducted to verify targeting and function.ResultsComplete response (CR) patients had significantly lower baseline HBV DNA than suboptimal response (SR) patients. After 48 weeks of pegIFNα therapy, miR-194-5p expression decreased notably in the CR group and correlated positively with HBV DNA and HBeAg levels. miR-194-5p predicted treatment response with an AUC of 0.831 and was an independent predictor. Mechanistically, miR-194-5p targeted SOCS2. Functional studies demonstrated that miR-194-5p overexpression enhanced, while SOCS2 supplementation attenuated, pegIFNα-induced phosphorylation of STAT1/STAT2, thereby influencing cell viability and inflammatory factor expression (TNF-α, IL-6, IL-1β).ConclusionmiR-194-5p may predict pegIFNα response in HBeAg-positive CHB. It regulates interferon signaling by targeting SOCS2 and modulating the JAK-STAT pathway activation, suggesting the miR-194-5p/SOCS2 axis as a potential therapeutic target.

Pterygium syndrome is a common eye disease that often leads to vision loss and even blindness. There is increasing evidence that miRNAs play a key role in the progression of pterygium, but the function of miR-381-3p in pterygium has not been studied. Therefore, this study aimed to investigate the effect of miR-381-3p on the progression of pterygium and to elucidate its potential molecular mechanisms. Human pterygium fibroblasts (HPFs) were isolated from clinical pterygium tissues. The expression of key genes and proteins was detected via RT-qPCR and western blotting. Cell proliferation was detected by CCK-8 and scratch assay, while cell invasion was examined by Transwell assay. Protein interactions were investigated by coimmunoprecipitation. First, we found that the expression level of miR-381-3p was significantly reduced in pterygium tissues. Second, we found that the overexpression of miR-381-3p in HPFs inhibited the proliferation, migration, and invasion abilities of HPFs while inducing cell apoptosis. In addition, in pterygium tissue, the expression of MCPIP1 was downregulated, and the expression of HACE1 and TRIP12 was upregulated. Importantly, MCPIP1 interference partially attenuated the positive effects of miR-381-3p overexpression described above, and miR-381-3p could target HACE1, while HACE1 could bind to TRIP12. Mechanistic studies revealed that miR-381-3p inhibited the binding of HACE1 to TRIP12 through the inhibition of HACE1 expression, thereby inhibiting the ubiquitination and degradation of MCPIP1 and improving the progression of pterygium. Our study highlights the powerful potential of miR-381-3p in improving the progression of pterygium, laying the foundation for the development of new intervention targets for related diseases.

Myocardial infarction (MI) is the most common cardiovascular disease that has a serious impact on human health and is one of the most common causes of death in the world. Apoptosis and myocardial fibrosis after MI are the key pathological features of poor myocardial remodelling, which further lead to heart failure, and are also the main reason for the high mortality from MI. Exosomes (Exos) from GATA4-overexpressing bone marrow mesenchymal stem cells (BMSCs) were extracted and identified using transmission electron microscopy (TEM), nanoscale tracking analysis (NTA), and marker detection. Tandem mass tags (TMT) quantitative proteomics, Agilent miRNA microarray, and GO/KEGG function analysis were used to obtain differentially expressed proteins/miRNAs. The expression levels using the Wald test and RT-qPCR to identify and validate. Their related mRNA-miRNA-circRNA regulatory networks were constructed. Hypoxic mouse cardiomyocytes were cultured, and MI mice were used for subsequent experiments. Cell differentiation, apoptosis, and marker gene expression were detected using RT-qPCR, IF, flow cytometry, and Western blot. Four key apoptosis proteins, CXCL12, CLU, CD44, and IGF1, and 20 key differentially expressed miRNAs. Among them, CLU and IGF1 were upregulated, but CXCL12 and CD44 were downregulated in Exos from the GATA4-overexpressing BMSCs group. The expression of mmu-miR-467g and mmu-miR-5127 was downregulated in Exos from the GATA4-overexpressing BMSCs group, whereas the other 18 key miRNAs were upregulated. Exos from GATA4-overexpressing BMSCs inhibit apoptosis, activate the LXR/RXR pathway, and improve MI. Knockout of CLU reversed this effect. Our research further discovered that GATA4-overexpressing BMSCs-derived Exos improved MI through downregulation of CLU. GATA4-overexpressing BMSCs-derived Exos may regulate MI via the aforementioned four key proteins and may be related to the LXR/RXR signalling pathway.

Lung cancer remains the leading cause of cancer-related incidence and mortality worldwide. Its tendency for postoperative distant metastasis significantly compromises long-term prognosis and survival. Accurately predicting the metastatic potential in a timely manner is crucial for formulating optimal treatment strategies. This study aimed to comprehensively compare the predictive performance of nine machine learning (ML) models and to enhance interpretability through SHAP (Shapley Additive Explanations), with the goal of developing a practical and transparent risk stratification tool for postoperative lung cancer management. Clinical data from 3,120 patients with stage I-III lung cancer who underwent radical surgery were retrospectively collected and randomly divided into training and testing cohorts. A total of 52 clinical, pathological, imaging, and laboratory variables were analyzed. Nine ML models-including eXtreme Gradient Boosting (XGBoost), Random Forest (RF), Light Gradient Boosting Machine (LightGBM), Adaptive Boosting (AdaBoost), Decision Tree (DT), Gradient Boosting Decision Tree (GBDT), Gaussian Naive Bayes (GNB), Complement Naive Bayes (CNB), and Multilayer Perceptron classifier (MLP)-were developed and evaluated. Model performance was assessed using accuracy, precision, recall, F1 score, ROC-AUC, PR-AUC, calibration, and decision curve analysis (DCA). All models were evaluated using nested cross-validation (outer stratified 70/30 splits repeated 10 times; inner fivefold tuning), with decision thresholds prespecified in the inner loop and applied unchanged to held-out tests. Given the approximately 4:1 class imbalance, cost-sensitive learning was primarily adopted, and PR-AUC was reported in addition to ROC-AUC. Among the nine models, GBDT demonstrated the highest predictive performance, achieving an AUC of 0.810 (95% CI: 0.748-0.872), accuracy of 0.766, sensitivity of 0.698, and specificity of 0.786 in the test set. SHAP analysis revealed that adjuvant chemotherapy, adjuvant radiotherapy, pathological N stage, age, body mass index (BMI), and preoperative neutrophil count (Pre-ANC) were the most influential predictors of distant metastasis. The combination of model performance and interpretability supported the model's potential for integration into clinical workflows to assist in real-time decision-making. In this work, we carried out a systematic comparison of nine machine learning algorithms in a large postoperative cohort under a coherent and interpretable framework. By jointly considering discrimination, calibration, clinical benefit (via decision curve analysis), and SHAP-based explanations, we constructed a practical prognostic tool to guide personalized treatment strategies and follow-up care. This methodology offers a data-driven basis for precision management. Ultimately, our findings provide an internally validated reference framework that warrants external and multicenter validation prior to clinical deployment.

Multidrug-resistant (MDR) Chryseobacterium indologenes is an emerging pathogen causing challenging central nervous system (CNS) infections, for which there are no standardized treatment guidelines. A systematic review of the literature indicates that such infections are extremely rare. Although trimethoprim-sulfamethoxazole (TMP-SMX) is the most frequently reported first-line therapy, its clinical efficacy is not universal, posing a significant therapeutic challenge. We present the case of a 25-year-old postpartum woman who developed MDR C. indologenes meningitis and a brain abscess after undergoing a neurosurgical procedure. The infection did not respond to initial treatment with meropenem or a subsequent course of the first-line agent TMP-SMX. Based on antimicrobial susceptibility testing, therapy was switched to rifampin monotherapy, which led to rapid clinical and microbiological recovery. The patient completed a 24-day course of rifampin and remained recurrence-free during a 20-month follow-up. This is the first report of successful rifampin monotherapy for an MDR C. indologenes CNS infection in a patient unresponsive to TMP-SMX. When considered alongside existing literature, these findings highlight the essential role of susceptibility testing and establish rifampin as an important salvage therapy for this life-threatening infection, particularly when recommended first-line treatments fail.