Predictors of length of stay in the coronary care unit in patient with acute coronary syndrome based on data mining methods

Predictors of prolonged hospital stay and in-hospital mortality in female patients with acute myocardial infarction with specific reference to diabetes

This study aimed to determine predictors of prolonged length of stay (LOS) and mortality in female STEMI patients with diabetes in comparison to female STEMI patients without diabetes. This retrospective single-center study was conducted between 2015 and 2020 as part of the STEMI registry in our center and included all STEMI patients. RESULTS: Out of 3081 STEMI patients, 16% (N = 498) were female, and 64% (n = 318) of them had diabetes. Diabetic patients were less commonly presented with anterior wall myocardial infarction (AWMI) (47% vs 65%, p = 0.001), but tended to have a higher prevalence of left main (LM) significant disease compared to non-diabetic patients (4% vs 1%, p = 0.06). Of the patients, 36% had prolonged LOS (≥5 days), and they were more obese with a higher incidence of hemoglobin drop ≥ 3 g/dL, higher admission troponin, and peak creatinine. 22% of patients with prolonged LOS came to the hospital after receiving lytic therapy and showed a higher prevalence of multivessel stenosis. Prolonged LOS patients had a higher incidence of pulmonary edema and cardiac arrest, which necessitated ventilation, although they did not show a significant difference in mortality from the group with LOS < 5 days. Our study found a significant negative correlation between glycosylated hemoglobin (HbA1c) and left ventricular ejection fraction (LVEF) (p = 0.02), but a positive correlation was noticed between HbA1c and LOS (p < 0.001). Regression analysis showed that diabetes mellitus, hemoglobin drop, creatinine level, and LVEF were significant predictors of prolonged LOS among females, although HbA1c, BMI ≥ 30, STEMI type, and peak creatinine level were independent predictors of prolonged LOS among female diabetic patients. Age ≥ 65 years and LVEF were significant independent predictors of mortality (p = 0.04, 0.02 respectively) in STEMI female patients. CONCLUSIONS: Diabetes mellitus, bleeding, renal impairment, and LVEF were found to be significant independent predictors of prolonged LOS among STEMI females. Age ≥ 65 years and LVEF were significant predictors of mortality among STEMI female patients.

Predicting the risk of hypertension using machine learning algorithms: A cross sectional study in Ethiopia

BACKGROUND AND OBJECTIVES

Hypertension (HTN), a major global health concern, is a leading cause of cardiovascular disease, premature death and disability, worldwide. It is important to develop an automated system to diagnose HTN at an early stage. Therefore, this study devised a machine learning (ML) system for predicting patients with the risk of developing HTN in Ethiopia.

MATERIALS AND METHODS

The HTN data was taken from Ethiopia, which included 612 respondents with 27 factors. We employed Boruta-based feature selection method to identify the important risk factors of HTN. The four well-known models [logistics regression, artificial neural network, random forest, and extreme gradient boosting (XGB)] were developed to predict HTN patients on the training set using the selected risk factors. The performances of the models were evaluated by accuracy, precision, recall, F1-score, and area under the curve (AUC) on the testing set. Additionally, the SHapley Additive exPlanations (SHAP) method is one of the explainable artificial intelligences (XAI) methods, was used to investigate the associated predictive risk factors of HTN.

RESULTS

The overall prevalence of HTN patients is 21.2%. This study showed that XGB-based model was the most appropriate model for predicting patients with the risk of HTN and achieved the accuracy of 88.81%, precision of 89.62%, recall of 97.04%, F1-score of 93.18%, and AUC of 0. 894. The XBG with SHAP analysis reveal that age, weight, fat, income, body mass index, diabetes mulitas, salt, history of HTN, drinking, and smoking were the associated risk factors of developing HTN.

CONCLUSIONS

The proposed framework provides an effective tool for accurately predicting individuals in Ethiopia who are at risk for developing HTN at an early stage and may help with early prevention and individualized treatment.

Building a predictive model to assist in the diagnosis of cervical cancer

Aim: Cervical cancer is still one of the most common gynecologic cancers in the world. Since cervical cancer is a potentially preventive cancer, earlier detection is the most effective technique for decreasing the worldwide incidence of the illness. Materials and methods: This research presents a novel ensemble technique for predicting cervical cancer risk. Specifically, the authors introduce a voting classifier that aggregates prediction probabilities from multiple machine-learning models: logistic regression, K-nearest neighbor, decision tree, XGBoost and multilayer perceptron. Results: The average accuracy, precision, recall and f1-score of the voting classifier were 96.6, 97.4, 95.9 and 96.6, respectively. Furthermore, the voting algorithm gains average high values for all evaluation metrics (accuracy, precision, recall and f1-score). The f1-score of the algorithm is 96%, which demonstrates the robustness of the model. Conclusion: The findings suggest that the probability of having cervical cancer can be accurately predicted utilizing the voting technique.