Development and Validation of an Interpretable Machine Learning Model for Early Prognosis Prediction in ICU Patients with Malignant Tumors and Hyperkalemia.

This study aims to develop and validate a machine learning (ML) predictive model for assessing mortality in patients with malignant tumors and hyperkalemia (MTH). We extracted data on patients with MTH from the Medical Information Mart for Intensive Care-IV, version 2.2 (MIMIC-IV v2.2) database. The dataset was split into a training set (75%) and a validation set (25%). We used the Least Absolute Shrinkage and Selection Operator (LASSO) regression to identify potential predictors, which included clinical laboratory indicators and vital signs. Pearson correlation analysis tested the correlation between predictors. In-hospital death was the prediction target. The Area Under the Curve (AUC) and accuracy of the training and validation sets of 7 ML algorithms were compared, and the optimal 1 was selected to develop the model. The calibration curve was used to evaluate the prediction accuracy of the model further. SHapley Additive exPlanations (SHAP) and Local Interpretable Model-agnostic Explanations (LIME) enhanced model interpretability. 496 patients with MTH in the Intensive Care Unit (ICU) were included. After screening, 17 clinical features were included in the construction of the ML model, and the Pearson correlation coefficient was <0.8, indicating that the correlation between the clinical features was small. eXtreme Gradient Boosting (XGBoost) outperformed other algorithms, achieving perfect scores in the training set (accuracy: 1.000, AUC: 1.000) and high scores in the validation set (accuracy: 0.734, AUC: 0.733). The calibration curves indicated good predictive calibration of the model. SHAP analysis identified the top 8 predictive factors: urine output, mean heart rate, maximum urea nitrogen, minimum oxygen saturation, minimum mean blood pressure, maximum total bilirubin, mean respiratory rate, and minimum pH. In addition, SHAP and LIME performed in-depth individual case analyses. This study demonstrates the effectiveness of ML methods in predicting mortality risk in ICU patients with MTH. It highlights the importance of predictors like urine output and mean heart rate. SHAP and LIME significantly enhanced the model's interpretability.

Exploring the therapeutic potential of "Xiaochaihu Decoction": a systematic review and meta-analysis on the clinical effectiveness and safety in managing cancer-related fever.

Objective: This study aimed to conduct the first meta-analysis to comprehensively evaluate the clinical effectiveness and safety of Xiaochaihu Decoction in treating Cancer-related Fever (CRF). Methods: Eight databases were systematically searched in September 2023. The risk of bias (ROB) 2.0 tool recommended by Cochrane Handbook was applied to evaluate the ROB of the included randomized controlled trials (RCTs). Additionally, the quality of evidence was assessed using the Grading of recommendations assessment, development and evaluation (GRADE) tool. Results: We included 18 RCTs involving 1,424 patients. Compared to Western medicine or Xinhuang Tablets, Xiaochaihu Decoction significantly improved clinical effectiveness in CRF patients (risk ratio [RR] = 1.24, 95% confidence interval [CI]: 1.17, 1.32) and expedited the normalization of body temperature (mean difference [MD] = -5.29, 95%CI: -5.59, -4.99). It also demonstrated a reduction in tumor necrosis factor-α (TNF-α) levels (MD = -0.63, 95%CI: -0.84, -0.41) and an increase in IL-2 levels (MD = 1.42, 95%CI: -1.09, 1.74). Analysis of Karnofsky Performance Status (KPS) scores showed that the use of Xiaochaihu Decoction improved the quality of life in CRF patients (RR = 1.57, 95%CI: 1.11, 2.22) and reduced the incidence of adverse events. However, it is important to note that the majority of included studies showed "some concerns" in risk of bias based on ROB 2.0, and the evidence quality assessed by GRADE method was rated as "low". Conclusion: While this study suggests the clinical effectiveness and safety of Xiaochaihu Decoction in treating patients with CRF, confirming these findings will necessitate additional high-quality, large-scale RCTs in future research. Systematic Review Registration: https://www.crd.york.ac.uk/prospero/, identifier CRD42023484068.

Dried fruit intake can lower the risk of ulcerative colitis: evidence from a Mendelian randomization study.

BACKGROUND AND OBJECTIVES: This study aims to examine the causal relationship between dietary factors and ulcerative colitis (UC). METHODS AND STUDY DESIGN: The analysis utilized data from genome-wide association studies (GWAS). Dried fruit, vegetables, processed meat, fresh fruit, and cereal intake were examined as exposure factors. UC was considered the outcome. Two-sample Mendelian randomization (TSMR) analysis was performed using methods. Heterogeneity and horizontal pleiotropy assessments were conducted to ensure the robustness of our findings. Additionally, we applied False Discovery Rate (FDR) corrections for multiple tests. RESULTS: The analysis revealed a significant inverse causal relationship between dried fruit intake and UC risk (odds ratio [OR]: 0.488, 95% confidence interval [CI]: 0.261 to 0.915, p = 0.025). No significant association was observed between vegetable intake (OR: 1.742, 95% CI: 0.561 to 5.415, p = 0.337), processed meat intake (OR: 1.136, 95% CI: 0.552 to 2.339, p = 0.729), fresh fruit intake (OR: 0.977, 95% CI: 0.465 to 2.054, p = 0.952), cereal intake (OR: 1.195, 95% CI: 0.669 to 2.134, p = 0.547). The low heterogeneity observed across analyses and the confirmation of stability through leave-one-out analysis reinforce the reliability of these results. Moreover, after adjusting for multiple tests, none of the dietary factors reached a p-value below the conventional significance threshold of 0.05. CONCLUSIONS: This study provides evidence of a potential association between dried fruit intake and a reduced risk of UC. Further MR studies incorporating larger GWAS datasets are needed to confirm these findings.

Performance of ChatGPT on Chinese Master's Degree Entrance Examination in Clinical Medicine.

BACKGROUND: ChatGPT is a large language model designed to generate responses based on a contextual understanding of user queries and requests. This study utilised the entrance examination for the Master of Clinical Medicine in Traditional Chinese Medicine to assesses the reliability and practicality of ChatGPT within the domain of medical education. METHODS: We selected 330 single and multiple-choice questions from the 2021 and 2022 Chinese Master of Clinical Medicine comprehensive examinations, which did not include any images or tables. To ensure the test's accuracy and authenticity, we preserved the original format of the query and alternative test texts, without any modifications or explanations. RESULTS: Both ChatGPT3.5 and GPT-4 attained average scores surpassing the admission threshold. Noteworthy is that ChatGPT achieved the highest score in the Medical Humanities section, boasting a correct rate of 93.75%. However, it is worth noting that ChatGPT3.5 exhibited the lowest accuracy percentage of 37.5% in the Pathology division, while GPT-4 also displayed a relatively lower correctness percentage of 60.23% in the Biochemistry section. An analysis of sub-questions revealed that ChatGPT demonstrates superior performance in handling single-choice questions but performs poorly in multiple-choice questions. CONCLUSION: ChatGPT exhibits a degree of medical knowledge and the capacity to aid in diagnosing and treating diseases. Nevertheless, enhancements are warranted to address its accuracy and reliability limitations. Imperatively, rigorous evaluation and oversight must accompany its utilization, accompanied by proactive measures to surmount prevailing constraints.

Effectiveness and Safety of Chinese Herbal Injections Combined with SOX Chemotherapy Regimens for Advanced Gastric Cancer: a Bayesian Network Meta-Analysis.

Background: Randomized controlled trials (RCTs) have demonstrated that combining Chinese herbal injections (CHIs) with oxaliplatin plus tegafur (SOX) chemotherapy regimens improves clinical effectiveness and reduces adverse reactions in patients with advanced gastric cancer (AGC). These RCTs highlight the potential applications of CHIs and their impact on AGC patient prognosis. However, there is insufficient comparative evidence on the clinical effectiveness and safety of different CHIs when combined with SOX. Therefore, we performed a network meta-analysis to rank the clinical effectiveness and safety of different CHIs when combined with SOX chemotherapy regimens. This study aimed to provide evidence for selecting appropriate CHIs in the treatment of patients with AGC. Methods: We searched eight databases from their inception until March 2023. Surface Under the Cumulative Ranking Curve (SUCRA) probability values were used to rank the treatment measures, and the Confidence in Network Meta-Analysis (CINeMA) software assessed the grading of evidence. Results: A total of 51 RCTs involving 3,703 AGC patients were identified. Huachansu injections + SOX demonstrated the highest clinical effectiveness (SUCRA: 78.17%), significantly reducing the incidence of leukopenia (93.35%), thrombocytopenia (80.19%), and nausea and vomiting (95.15%). Shenfu injections + SOX improved Karnofsky's Performance Status (75.59%) and showed a significant reduction in peripheral neurotoxicity incidence (88.26%). Aidi injections + SOX were most effective in reducing the incidence of liver function damage (75.16%). According to CINeMA, most confidence rating results were classified as "low". Conclusion: The combination of CHIs and SOX shows promising effects in the treatment of AGC compared to SOX alone. Huachansu and Shenfu injections offer the greatest overall advantage among the CHIs, while Aidi injections are optimal for reducing the incidence of liver damage. However, further rigorous RCTs with larger sample sizes and additional pharmacological studies are necessary to reinforce these findings.

Comparative effectiveness and safety of Chinese medicine belly button application for childhood diarrhea: a Bayesian network meta-analysis of randomized controlled trials.

Background: Chinese medicine belly button application (CMBBA) has been used to treat childhood diarrhea (CD) in several randomized controlled trials (RCTs), but its effectiveness and combination strategy still need to be clarified. Therefore, we aimed to evaluate the effectiveness, safety, and the optimal combination strategy of CMBBA in treating CD. Methods: Up until January 2023, we searched for studies that met our inclusion criteria in six databases, including PubMed, the Cochrane Library, Chinese SinoMed, CNKI, VIP, and Wanfang. Heterogeneity was quantified using I2 statistics. A methodological evaluation was performed using the Cochrane Risk Bias Tool 2.0. The Confidence in Network Meta-Analysis online software was employed to evaluate evidence grading. A minimally contextualized framework was used to provide a comprehensive conclusion for the network meta-analysis. This study protocol was registered with PROSPERO. Results: We analyzed data from 33 RCTs that included 4,490 children with diarrhea. In terms of clinical effectiveness, CMBBA plus montmorillonite powder plus anti-infectives may be the most effective treatment option for children with diarrhea and concurrent infection according to a minimally contextualized framework. Either exclusive use of CMBBA or CMBBA in combination with modern medicine was beneficial in reducing the time to diarrhea disappearance (MD = -1.33 days, 95% CI: -1.59 to -1.08, Z = -10.103, p < 0.001) compared to modern medicine exclusively, and the difference was statistically significant. The combined usage of CMBBA could shorten the recovery time of dehydration by an average of 0.74 days (MD = -0.74 days, 95% CI: -1.10 to -0.37, Z = -3.931.103, p < 0.001). While some studies have reported mild allergic reactions and mild abdominal pain after CMBBA use, these symptoms can be cured in a relatively short period of time. Conclusions: The combination of CMBBA, montmorillonite powder, and anti-infectives may provide superior clinical effectiveness for children with diarrhea and concurrent infection. To treat CD, CMBBA can be used effectively and safely. However, the findings must be interpreted with cautiously due to the limited number of clinical trials and the low quality of the studies. In addition, the choice of treatment plan should also be based on the specific conditions of each patient. Systematic Review Registration: https://www.crd.york.ac.uk/prospero/, identifier: CRD42022380694.