Hospital Admission Prediction Using Pre-hospital Variables

With the rapid outstripping of healthcare resourcesby the demands on hospital care, it is important to findmore effective and efficient ways for managing care.This research is aimed at developing new admissionprediction models using various pre-hospital variablesto help hospital estimate the patients to be admitted.We developed a framework of hospital admissionprediction and proposed two novel approaches tocapture semantics of chief complaints to enhanceprediction. Our experiments on a hospital datasetdemonstrated that our proposed models outperformedseveral benchmark methods.