November 19, 2024BMC Medical Informatics and Decision Making11 citationsOpen Access

Explainable predictions of a machine learning model to forecast the postoperative length of stay for severe patients: machine learning model development and evaluation

HCHa Na ChoUniversity of California, Irvine IAImjin AhnAsan Medical Center HGHansle GwonUlsan College

Structured PICO

Can an explainable machine learning model predict postoperative length of stay in severe surgical patients?

Population

Severe patients undergoing surgery

Intervention

Explainable machine learning model (XGBoost) using preoperative features from electronic medical records

Outcome

Postoperative length of stay

An explainable XGBoost machine learning model can predict postoperative length of stay using preoperative features, potentially aiding decision-making and operational facilitation.

Abstract

We successfully predicted the length of stay after surgery and provide explainable models with supporting analyses. In summary, we demonstrate the interpretation with the XGBoost model presenting insights on preoperative features and defining higher risk predictors to the length of stay outcome. Our development in explainable models supports the current in-depth knowledge for the future length of stay prediction on electronic medical records that aids the decision-making and facilitation of the operation department.

Ask AI

Helpful

Bookmark

View Full Paper