Comparative Analysis of Machine Learning Models for Prediction of Acute Liver Injury in Sepsis Patients.

Introduction: Acute liver injury (ALI) is a common complication of sepsis and is associated with adverse clinical outcomes. We aimed to develop a model to predict the risk of ALI in patients with sepsis after hospitalization. Methods: Medical records of 3196 septic patients treated at the Lishui Central Hospital in Zhejiang Province from January 2015 to May 2023 were selected. Cohort 1 was divided into ALI and non-ALI groups for model training and internal validation. The initial laboratory test results of the study subjects were used as features for machine learning (ML), and models built using nine different ML algorithms were compared to select the best algorithm and model. The predictive performance of model stacking methods was then explored. The best model was externally validated in Cohort 2. Results: In Cohort 1, LightGBM demonstrated good stability and predictive performance with an area under the curve (AUC) of 0.841. The top five most important variables in the model were diabetes, congestive heart failure, prothrombin time, heart rate, and platelet count. The LightGBM model showed stable and good ALI risk prediction ability in the external validation of Cohort 2 with an AUC of 0.815. Furthermore, an online prediction website was developed to assist healthcare professionals in applying this model more effectively. Conclusions: The Light GBM model can predict the risk of ALI in patients with sepsis after hospitalization.

Application of interpretable machine learning algorithms to predict acute kidney injury in patients with cerebral infarction in ICU.

BACKGROUND: Acute kidney injury (AKI) is not only a complication but also a serious threat to patients with cerebral infarction (CI). This study aimed to explore the application of interpretable machine learning algorithms in predicting AKI in patients with cerebral infarction. METHODS: The study included 3920 patients with CI admitted to the Intensive Care Unit and Emergency Medicine of the Central Hospital of Lishui City, Zhejiang Province. Nine machine learning techniques, including XGBoost, logistics, LightGBM, random forest (RF), AdaBoost, GaussianNB (GNB), Multi-Layer Perceptron (MLP), support vector machine (SVM), and k-nearest neighbors (KNN) classification, were used to develop a predictive model for AKI in these patients. SHapley Additive exPlanations (SHAP) analysis provided visual explanations for each patient. Finally, model effectiveness was assessed using metrics such as average precision (AP), sensitivity, specificity, accuracy, F1 score, precision-recall (PR) curve, calibration plot, and decision curve analysis (DCA). RESULTS: The XGBoost model performed better in the internal validation set and the external validation set, with an AUC of 0.940 and 0.887, respectively. The five most important variables in the model were, in order, glomerular filtration rate, low-density lipoprotein, total cholesterol, hemiplegia and serum kalium. CONCLUSION: This study demonstrates the potential of interpretable machine learning algorithms in predicting CI patients with AKI.

The establishment and evaluation of a swine model of deinagkistrodon acutus snakebite envenomation.

OBJECTIVE: To establish a preclinical large-animal model of Deinagkistrodon acutus snakebite envenomation and evaluate its feasibility. METHODS: The venom of D. acutus (0 mg/kg, 1 mg/kg, 2 mg/kg, 5 mg/kg, or 10 mg/kg) was injected into the left biceps femoris of 11 male pigs. Then, the circumferences of the limbs were regularly measured, and changes in muscle injury biomarkers, blood parameters, coagulation function, vital organ function and injury biomarkers were regularly detected. At 24 h after venom injection, the animals were euthanized, and the pathological damage to the vital organs mentioned above was evaluated. RESULTS: The two pigs receiving 10 mg/kg and 5 mg/kg snake venom died at 8 h and 12 h after injection, respectively. The remaining pigs were equally divided into 0 mg/kg, 1 mg/kg, and 2 mg/kg snake venom groups, and all of them survived to 24 h after injection. Compared with the pigs receiving 0 mg/kg snake venom, the pigs receiving 1 mg/kg or 2 mg/kg snake venom exhibited significant abnormities, including limb swelling; increased muscle injury biomarker creatine kinase (CK) and coagulation function indicators prothrombin time and D-dimer; and decreased blood routine indicator platelet and coagulation function indicator fibrinogen. Moreover, significant abnormalities in myocardial and cerebral function and injury biomarkers in the heart, brain, liver, kidney and intestine were also observed. In particular, the abnormalities mentioned above were significantly obvious in those pigs receiving 2 mg/kg snake venom. Pathological evaluation revealed that the morphology of muscle, heart, brain, liver, kidney, and intestine in those pigs receiving 0 mg/kg snake venom was normal; however, pathological damage was observed in those pigs receiving 1 mg/kg and 2 mg/kg snake venom. Similarly, the pathological damage was more severe in those pigs receiving 2 mg/kg snake venom. CONCLUSION: The intramuscular injection of 2 mg/kg D. acutus venom seems to be an optimal dose for examining the preclinical efficacy of existing and novel therapeutics for treating D. acutus envenomation in pigs.

Photoredox Catalysis Induced Bisindolylation of Ethers/Alcohols via Sequential C-H and C-O Bond Cleavage.

A visible-light-engaged 2-fold site-selective alkylation of indole derivatives with aliphatic ethers or alcohols has been accomplished for easy access to symmetric 3,3'-bisindolylmethane derivatives. The experimental data suggest a sequential photoredox catalysis induced radical addition and proton-mediated Friedel-Crafts alkylation mechanism.

Manganese-catalyzed synthesis of monofluoroalkenes via C-H activation and C-F cleavage.

The manganese-catalyzed α-fluoroalkenylation of arenes via C-H activation and C-F cleavage has been described. This protocol provides a very useful method for the synthesis of monofluoroalkenes with predominant unconventional E-isomer selectivity which complements the existing strategies for the access to these molecular architectures. In addition, the selectivity of ß-defluorination in the catalytic cycle not only determines the configurations of the products but also obviates the use of external oxidants, providing a good example in the exploitation of manganese-catalyzed redox-neutral C-H transformations.