This study aims to introduce how to use the PROBAST (prediction model risk of bias assessment tool) to evaluate risk of bias and applicability of the study of diagnostic or prognostic predictive models, including the introduction of the background, the scope of application and use of the tool. This tool mainly involves the four areas of participants, predictors, outcomes and analyses. The risk of bias in the research is evaluated through the four areas, while the applicability is evaluated in the first three. PROBAST provides a standardized approach to evaluate the critical appraisal of the study of diagnostic or prognostic predictive models, which screens qualified literature for data analysis and helps to establish a scientific basis for clinical decision-making.
Objective To systematically review the predictive model of stroke-related pneumonia risk. Methods The CNKI, WanFang Data, CBM, PubMed, Web of Science, Embase, MEDLINE and Cochrane Library databases were electronically searched to collect studies on risk prediction models for stroke-associated pneumonia from inception to February 15, 2023. Two researchers independently screened the literature and extracted data. The risk of bias and applicability of the models were assessed using PROBAST. Results A total of 18 studies and 27 SAP risk prediction models were included. The AUC values for inclusion in the model ranged from 0.67 to 0.96, and the number of candidate predictors ranged from 4 to 25, with the most common predictors being age, NIHSS score, dysphagia, mRS score, and impaired consciousness (GCS score). Conclusion The overall predictive performance of SAP risk prediction models is good, but their predictive performance cannot be directly compared because of the differences in study type, study population, and SAP diagnostic criteria. Moreover, 72.3% of the models are not externally validated, and most of the studies have a high risk of bias.
In recent years, wearable devices have seen a booming development, and the integration of wearable devices with clinical settings is an important direction in the development of wearable devices. The purpose of this study is to establish a prediction model for postoperative pulmonary complications (PPCs) by continuously monitoring respiratory physiological parameters of cardiac valve surgery patients during the preoperative 6-Minute Walk Test (6MWT) with a wearable device. By enrolling 53 patients with cardiac valve diseases in the Department of Cardiovascular Surgery, West China Hospital, Sichuan University, the grouping was based on the presence or absence of PPCs in the postoperative period. The 6MWT continuous respiratory physiological parameters collected by the SensEcho wearable device were analyzed, and the group differences in respiratory parameters and oxygen saturation parameters were calculated, and a prediction model was constructed. The results showed that continuous monitoring of respiratory physiological parameters in 6MWT using a wearable device had a better predictive trend for PPCs in cardiac valve surgery patients, providing a novel reference model for integrating wearable devices with the clinic.
Because of the diversity and complexity of clinical indicators, it is difficult to establish a comprehensive and reliable prediction model for induction of labor (IOL) outcomes with existing methods. This study aims to analyze the clinical indicators related to IOL and to develop and evaluate a prediction model based on a small-sample of data. The study population consisted of a total of 90 pregnant women who underwent IOL between February 2023 and January 2024 at the Shanghai First Maternity and Infant Healthcare Hospital, and a total of 52 clinical indicators were recorded. Maximal information coefficient (MIC) was used to select features for clinical indicators to reduce the risk of overfitting caused by high-dimensional features. Then, based on the features selected by MIC, the support vector machine (SVM) model based on small samples was compared and analyzed with the fully connected neural network (FCNN) model based on large samples in deep learning, and the receiver operating characteristic (ROC) curve was given. By calculating the MIC score, the final feature dimension was reduced from 55 to 15, and the area under curve (AUC) of the SVM model was improved from 0.872 before feature selection to 0.923. Model comparison results showed that SVM had better prediction performance than FCNN. This study demonstrates that SVM successfully predicted IOL outcomes, and the MIC feature selection effectively improves the model’s generalization ability, making the prediction results more stable. This study provides a reliable method for predicting the outcome of induced labor with potential clinical applications.
Objectiv To provide a comprehensive overview of model performance and predictive efficacy of machine learning techniques to predict septic shock in children, in order to target and improve the quality and predictive power of models for future studies. MethodsTo systematically review all studies in four databases (PubMed, Embase, Web of Science, ScienceDirect) on machine learning prediction of septic shock in children before April 1, 2024. Two investigators independently conducted literature screening, literature data extraction and bias assessment, and conducted a systematic review of basic information, research data, study design and prediction models. Model discrimination, which area under the curve (AUC), was pooled using a random-effects model and meta-analysis was performed. Subgroup analyses were performed according to sample sizes, machine learning models, types of predictors, number of predictors, etc. And publication bias and sensitivity analyses were performed for the included literature. Results A total of 11 studies were included, of which 2 were at low risk of bias, 7 were at unknown risk of bias, and 2 were at high risk of bias. The data used in the included studies included both public and non-public electronic medical record databases, and the machine learning models used included logistic regression, random forest, support vector machine, and XGBoost, etc. The predictive models constructed based on different databases appeared to have different results in terms of the characteristic variables, so identifying the key variables of the predictive models requires further validation on other datasets. Meta-analysis showed the pooled AUC of 0.812 (95%CI 0.763 to 0.860, P<0.001), and further subgroup analyses showed that larger sample sizes (≥1 000) and predictor variable types significantly improved the predictive effect of the model, and the difference in AUC was statistically significant (95%CI not overlapping). The funnel plot showed that there was publication bias in the study, and when the extreme AUC values were excluded, the meta-analysis yielded a total AUC of 0.815 (95%CI 0.769 to 0.861, P<0.001), indicating that the extreme AUC values were insensitive. ConclusionMachine learning technology has shown some potential in predicting septic shock in children, but the quality of existing research needs to be strengthened, and future research work should improve the quality of research and improve the prediction effect of the model by expanding the sample size.