Dataset Information

Machine learning vs. conventional methods for prediction of 30-day readmission following percutaneous mitral edge-to-edge repair.

ABSTRACT:

Background

Identifying predictors of readmissions after mitral valve transcatheter edge-to-edge repair (MV-TEER) is essential for risk stratification and optimization of clinical outcomes.

Aims

We investigated the performance of machine learning [ML] algorithms vs. logistic regression in predicting readmissions after MV-TEER.

Methods

We utilized the National-Readmission-Database to identify patients who underwent MV-TEER between 2015 and 2018. The database was randomly split into training (70 %) and testing (30 %) sets. Lasso regression was used to remove non-informative variables and rank informative ones. The top 50 informative predictors were tested using 4 ML models: ML-logistic regression [LR], Naive Bayes [NB], random forest [RF], and artificial neural network [ANN]/For comparison, we used a traditional statistical method (principal component analysis logistic regression PCA-LR).

Results

A total of 9425 index hospitalizations for MV-TEER were included. Overall, the 30-day readmission rate was 14.6 %, and heart failure was the most common cause of readmission (32 %). The readmission cohort had a higher burden of comorbidities (median Elixhauser score 5 vs. 3) and frailty score (3.7 vs. 2.9), longer hospital stays (3 vs. 2 days), and higher rates of non-home discharges (17.4 % vs. 8.5 %). The traditional PCA-LR model yielded a modest predictive value (area under the curve [AUC] 0.615 [0.587-0.644]). Two ML algorithms demonstrated superior performance than the traditional PCA-LR model; ML-LR (AUC 0.692 [0.667-0.717]), and NB (AUC 0.724 [0.700-0.748]). RF (AUC 0.62 [0.592-0.677]) and ANN (0.65 [0.623-0.677]) had modest performance.

Conclusion

Machine learning algorithms may provide a useful tool for predicting readmissions after MV-TEER using administrative databases.

SUBMITTER: Sulaiman S

PROVIDER: S-EPMC10762683 | biostudies-literature | 2023 Nov

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Machine learning vs. conventional methods for prediction of 30-day readmission following percutaneous mitral edge-to-edge repair.

Sulaiman Samian S Kawsara Akram A El Sabbagh Abdallah A Mahayni Abdulah Amer AA Gulati Rajiv R Rihal Charanjit S CS Alkhouli Mohamad M

Cardiovascular revascularization medicine : including molecular interventions 20230518

<h4>Background</h4>Identifying predictors of readmissions after mitral valve transcatheter edge-to-edge repair (MV-TEER) is essential for risk stratification and optimization of clinical outcomes.<h4>Aims</h4>We investigated the performance of machine learning [ML] algorithms vs. logistic regression in predicting readmissions after MV-TEER.<h4>Methods</h4>We utilized the National-Readmission-Database to identify patients who underwent MV-TEER between 2015 and 2018. The database was randomly spli ...[more]

PMID: 37248108

Similar Datasets

Project description:IntroductionMitral valve repair has been established as the preferred treatment option in the management of degenerative mitral valve disease. Compared with other surgical treatment options, mitral valve repair is associated with increased survival and decreased rates of both complications and reoperations. However, among patients undergoing mitral valve repair, little is known about the predictors of postoperative outcomes. The purpose of this study is to identify preoperative patient risk factors associated with postoperative morbidity and mortality within 30 days of mitral valve repair.MethodsData was derived from the American College of Surgeons National Surgical Quality Improvement Program database to assess patients who underwent mitral valve repair from 2011 through 2017. Preoperative risk factors were analyzed to determine their association with a variety of postoperative 30-day outcome measures.ResultsOne thousand three hundred and sixty-six patients underwent mitral valve repair; 849 (62.2%) males and 517 (37.8%) females. Ages ranged from 18 to 90 years, with a mean age of 64 years. The overall 30-day mortality was 3.1% (43 patients). Among the 12 identified risk factors associated with increased mortality on univariate analysis, pre-operative hematocrit level was the only variable significantly correlated with mortality after undergoing multivariate analysis. 259 patients (19.1%) were discharged to a location other than home, an outcome associated with 22 identified risk factors. Among these risk factors, female gender, age, dialysis, pre-operative serum sodium, pre-operative serum albumin, and partial or full living dependency remained statistically significant following multivariate analysis. 126 patients (9.2%) experienced unplanned readmission. This outcome was associated with five risk factors, of which only dyspnea upon mild exertion was significant on multivariate analysis. Reoperation occurred in 105 patients (7.7%). Of the seven identified variables associated with reoperation, patient age, pre-operative platelet count, dyspnea upon mild exertion were independent predictors on multivariate analysis. 53 patients (3.9%) underwent reintubation, which was associated with 11 identified risk factors. Among them, patient age and pre-operative INR value were predictive of reintubation on multivariate analysis. 26 patients (1.9%) experienced stroke, of whom age was the only associated risk factor on both univariate and multivariate analysis. 31 patients (2.3%) experienced acute renal failure, which correlated with 11 risk factors on univariate analysis. Of these, only patient age and pre-operative hematocrit were identified as independent predictors on multivariate analysis.ConclusionsOutcomes are good following mitral valve repair. Although a substantial number of risk factors were found to be associated with adverse outcomes, only a small subset remained statistically significant following multivariate analysis. Identification of these risk factors may help guide clinical decision making with respect to which patients are the best candidates to undergo mitral valve repair.

Project description:Background and Purpose: Hospital readmissions impose a substantial burden on the healthcare system. Reducing readmissions after stroke could lead to improved quality of care especially since stroke is associated with a high rate of readmission. The goal of this study is to enhance our understanding of the predictors of 30-day readmission after ischemic stroke and develop models to identify high-risk individuals for targeted interventions. Methods: We used patient-level data from electronic health records (EHR), five machine learning algorithms (random forest, gradient boosting machine, extreme gradient boosting-XGBoost, support vector machine, and logistic regression-LR), data-driven feature selection strategy, and adaptive sampling to develop 15 models of 30-day readmission after ischemic stroke. We further identified important clinical variables. Results: We included 3,184 patients with ischemic stroke (mean age: 71 ± 13.90 years, men: 51.06%). Among the 61 clinical variables included in the model, the National Institutes of Health Stroke Scale score above 24, insert indwelling urinary catheter, hypercoagulable state, and percutaneous gastrostomy had the highest importance score. The Model's AUC (area under the curve) for predicting 30-day readmission was 0.74 (95%CI: 0.64-0.78) with PPV of 0.43 when the XGBoost algorithm was used with ROSE-sampling. The balance between specificity and sensitivity improved through the sampling strategy. The best sensitivity was achieved with LR when optimized with feature selection and ROSE-sampling (AUC: 0.64, sensitivity: 0.53, specificity: 0.69). Conclusions: Machine learning-based models can be designed to predict 30-day readmission after stroke using structured data from EHR. Among the algorithms analyzed, XGBoost with ROSE-sampling had the best performance in terms of AUC while LR with ROSE-sampling and feature selection had the best sensitivity. Clinical variables highly associated with 30-day readmission could be targeted for personalized interventions. Depending on healthcare systems' resources and criteria, models with optimized performance metrics can be implemented to improve outcomes.

Dataset Information

Machine learning vs. conventional methods for prediction of 30-day readmission following percutaneous mitral edge-to-edge repair.

Background

Aims

Methods

Results

Conclusion

Publications

Machine learning vs. conventional methods for prediction of 30-day readmission following percutaneous mitral edge-to-edge repair.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets