Skip to main content
Fig. 1 | World Journal of Emergency Surgery

Fig. 1

From: Machine-learning based prediction of appendicitis for patients presenting with acute abdominal pain at the emergency department

Fig. 1

Receiver operating characteristic (ROC) plots for the HIVE model (A) and the HIVE-LAB model (B) in predicting appendicitis versus other causes of AAP using the validation population. The top 10 individual parameter contributions to the models are represented by SHapley Additive exPlanations (SHAP) values scaled and plotted as percentage contributions to the prediction (C, D). Parameters outside the top 10 contribute a combined total of 17.5% to the HIVE model and 32.5% to the HIVE-LAB model. HIVE, intake, medical HIstory, Vital signs, physical Examination; HIVE-LAB, intake, medical HIstory, Vital signs, physical Examination, Laboratory testing; AAP, acute abdominal pain; Temp, Temperature; MAP, Mean Arterial Pressure; MH, Medical History; PE, Physical Examination

Back to article page

World Journal of Emergency Surgery

ISSN: 1749-7922