Negin Ashrafi

All Publications

• Enhanced prediction of ventilator-associated pneumonia in patients with traumatic brain injury using advanced machine learning techniques. Scientific reports Ashrafi, N., Abdollahi, A., Alaei, K., Pishgar, M. 2025; 15 (1): 11363

  Abstract

  Ventilator-associated pneumonia significantly increases morbidity, mortality, and healthcare costs among patients with traumatic brain injury. Accurately predicting risk can facilitate earlier interventions and improve patient outcomes. This study leveraged the MIMIC III database, identifying traumatic brain injury cases through standardized clinical criteria. A rigorous data preprocessing workflow included missing value imputation, correlation checks, and expert-driven feature selection, reducing an initial set of features to a subset of critical predictors encompassing demographics, comorbidities, laboratory values, and clinical interventions. To address class imbalance, the Synthetic Minority Oversampling Technique (SMOTE) was applied within a five-fold cross-validation framework, ensuring a balanced training set while maintaining an unbiased validation process. Six machine learning models, including Support Vector Machine, Logistic Regression, Random Forest, XGBoost, Artificial Neural Network, and AdaBoost, were trained using extensive hyperparameter tuning. Comprehensive evaluations were conducted based on multiple metrics, including Area Under the Curve (AUC), accuracy, F1 score, sensitivity, specificity, Positive Predictive Value, and Negative Predictive Value. XGBoost emerged as the top performing algorithm, achieving an AUC of 0.94 and an accuracy of 0.875 on the test set, marking substantial improvements over previously reported best results. An ablation study validated the necessity of each retained feature, indicating that any feature removal led to a decline in model performance. Furthermore, SHAP analysis underscored ICU length of stay, hospital length of stay, serum potassium, and blood urea nitrogen as key contributors to ventilator associated pneumonia risk. Overall, the results demonstrate that advanced ensemble learning, meticulous feature selection, and effective class imbalance handling can significantly enhance early detection in traumatic brain injury cases. These findings have meaningful clinical implications, offering a framework for more timely interventions, optimized resource allocation, and improved patient care in critical settings.

  View details for DOI 10.1038/s41598-025-95779-0

  View details for PubMedID 40175458

  View details for PubMedCentralID PMC11965472

• A machine learning-based prediction of hospital mortality in mechanically ventilated ICU patients. PloS one Li, H., Ashrafi, N., Kang, C., Zhao, G., Chen, Y., Pishgar, M. 2024; 19 (9): e0309383

  Abstract

  Mechanical ventilation (MV) is vital for critically ill ICU patients but carries significant mortality risks. This study aims to develop a predictive model to estimate hospital mortality among MV patients, utilizing comprehensive health data to assist ICU physicians with early-stage alerts.We developed a Machine Learning (ML) framework to predict hospital mortality in ICU patients receiving MV. Using the MIMIC-III database, we identified 25,202 eligible patients through ICD-9 codes. We employed backward elimination and the Lasso method, selecting 32 features based on clinical insights and literature. Data preprocessing included eliminating columns with over 90% missing data and using mean imputation for the remaining missing values. To address class imbalance, we used the Synthetic Minority Over-sampling Technique (SMOTE). We evaluated several ML models, including CatBoost, XGBoost, Decision Tree, Random Forest, Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and Logistic Regression, using a 70/30 train-test split. The CatBoost model was chosen for its superior performance in terms of accuracy, precision, recall, F1-score, AUROC metrics, and calibration plots.The study involved a cohort of 25,202 patients on MV. The CatBoost model attained an AUROC of 0.862, an increase from an initial AUROC of 0.821, which was the best reported in the literature. It also demonstrated an accuracy of 0.789, an F1-score of 0.747, and better calibration, outperforming other models. These improvements are due to systematic feature selection and the robust gradient boosting architecture of CatBoost.The preprocessing methodology significantly reduced the number of relevant features, simplifying computational processes, and identified critical features previously overlooked. Integrating these features and tuning the parameters, our model demonstrated strong generalization to unseen data. This highlights the potential of ML as a crucial tool in ICUs, enhancing resource allocation and providing more personalized interventions for MV patients.

  View details for DOI 10.1371/journal.pone.0309383

  View details for PubMedID 39231126

  View details for PubMedCentralID PMC11373795

• Prediction of sepsis mortality in ICU patients using machine learning methods. BMC medical informatics and decision making Gao, J., Lu, Y., Ashrafi, N., Domingo, I., Alaei, K., Pishgar, M. 2024; 24 (1): 228

  Abstract

  Sepsis, a life-threatening condition, accounts for the deaths of millions of people worldwide. Accurate prediction of sepsis outcomes is crucial for effective treatment and management. Previous studies have utilized machine learning for prognosis, but have limitations in feature sets and model interpretability.This study aims to develop a machine learning model that enhances prediction accuracy for sepsis outcomes using a reduced set of features, thereby addressing the limitations of previous studies and enhancing model interpretability.This study analyzes intensive care patient outcomes using the MIMIC-IV database, focusing on adult sepsis cases. Employing the latest data extraction tools, such as Google BigQuery, and following stringent selection criteria, we selected 38 features in this study. This selection is also informed by a comprehensive literature review and clinical expertise. Data preprocessing included handling missing values, regrouping categorical variables, and using the Synthetic Minority Over-sampling Technique (SMOTE) to balance the data. We evaluated several machine learning models: Decision Trees, Gradient Boosting, XGBoost, LightGBM, Multilayer Perceptrons (MLP), Support Vector Machines (SVM), and Random Forest. The Sequential Halving and Classification (SHAC) algorithm was used for hyperparameter tuning, and both train-test split and cross-validation methodologies were employed for performance and computational efficiency.The Random Forest model was the most effective, achieving an area under the receiver operating characteristic curve (AUROC) of 0.94 with a confidence interval of ±0.01. This significantly outperformed other models and set a new benchmark in the literature. The model also provided detailed insights into the importance of various clinical features, with the Sequential Organ Failure Assessment (SOFA) score and average urine output being highly predictive. SHAP (Shapley Additive Explanations) analysis further enhanced the model's interpretability, offering a clearer understanding of feature impacts.This study demonstrates significant improvements in predicting sepsis outcomes using a Random Forest model, supported by advanced machine learning techniques and thorough data preprocessing. Our approach provided detailed insights into the key clinical features impacting sepsis mortality, making the model both highly accurate and interpretable. By enhancing the model's practical utility in clinical settings, we offer a valuable tool for healthcare professionals to make data-driven decisions, ultimately aiming to minimize sepsis-induced fatalities.

  View details for DOI 10.1186/s12911-024-02630-z

  View details for PubMedID 39152423

  View details for PubMedCentralID 6429642