Key result
Data pre-processing using MICE with one-hot encoding and standardization improved XGBoost model performance for predicting heart failure outcomes compared to the default model.
Why the study?
Heart failure causes high readmission and mortality rates, making accurate predictive models essential, but the impact of data pre-processing techniques on XGBoost model performance remains to be determined.
Does data pre-processing improve the performance of XGBoost models in predicting all-cause readmission and mortality in heart failure patients?
Does data pre-processing improve the performance of XGBoost models in predicting all-cause readmission and mortality in heart failure patients?
Data pre-processing, particularly MICE imputation with one-hot encoding and standardization, modestly improves XGBoost model performance for predicting readmission and mortality in heart failure patients.
Answers are grounded in this paper and cite their sources.
Pre-processing may enhance XGBoost HF readmission models; leaves open clinical adoption pending prospective validation.
Background: Heart failure poses a significant global health challenge, with high rates of readmission and mortality. Accurate models to predict these outcomes are essential for effective patient management. This study investigates the impact of data pre-processing techniques on XGBoost model performance in predicting all-cause readmission and mortality among heart failure patients. Methods: A dataset of 168 features from 2008 heart failure patients was used. Pre-processing included handling missing values, categorical encoding, and standardization. Four imputation techniques were compared: Mean, Multivariate Imputation by Chained Equations (MICEs), k-nearest Neighbors (kNNs), and Random Forest (RF). XGBoost models were evaluated using accuracy, recall, F1-score, and Area Under the Curve (AUC). Robustness was assessed through 10-fold cross-validation. Results: The XGBoost model with kNN imputation, one-hot encoding, and standardization outperformed others, with an accuracy of 0.614, recall of 0.551, and F1-score of 0.476. The MICE-based model achieved the highest AUC (0.647) and mean AUC (0.65 ± 0.04) in cross-validation. All pre-processed models outperformed the default XGBoost model (AUC: 0.60). Conclusions: Data pre-processing, especially MICE with one-hot encoding and standardization, improves XGBoost performance in heart failure prediction. However, moderate AUC scores suggest further steps are needed to enhance predictive accuracy.
No takes yet. Share an insight, caveat, or question.
Hidayaturrohman et al. (2024) studied Heart failure (n=2,008). Data pre-processing techniques (e.g., MICE, kNN, one-hot encoding, standardization) vs. Default XGBoost model was evaluated on Model performance for predicting all-cause readmission and mortality (AUC). Data pre-processing using MICE with one-hot encoding and standardization improved XGBoost model performance for predicting heart failure outcomes compared to the default model.
Synapse has enriched 10 related papers. Consider them for comparative context: