<ul><li>The study focuses on predicting Length of Stay (LOS) in ICU for neurological disease patients using various machine learning algorithms.</li><li>Different ML models such as K-Nearest Neighbors, Random Forest, XGBoost, CatBoost, LSTM, BERT, and Temporal Fusion Transformer were evaluated.</li><li>LOS was categorized into three groups: less than two days, less than a week, and a week or more for classification purposes.</li><li>Random Forest model performed best on static data with an accuracy, precision, recall, and F1-score of 0.68, while BERT model outperformed LSTM on time-series data with corresponding scores of 0.80.</li></ul>

Predicting Length of Stay in Neurological ICU Patients Using Classical Machine Learning and Neural Network Models: A Benchmark Study on MIMIC-IV

Discover more