Prediction of Severe Acute Pancreatitis at a Very Early Stage of the Disease Using Artificial Intelligence Techniques, Without Laboratory Data or Imaging Tests: The PANCREATIA Study.

OBJECTIVE: To evaluate machine learning (ML) models' performance in predicting acute pancreatitis (AP) severity using early-stage variables while excluding laboratory and imaging tests. BACKGROUND: Severe acute pancreatitis (SAP) affects ∼20% of patients with AP and is associated with high mortality rates. Accurate early prediction of SAP and in-hospital mortality is crucial for effective management. Traditional scores such as Acute Physiology and Chronic Health Disease Classification System II and Bedside Index of Severity in Acute Pancreatitis are complex and require laboratory tests, while early predictive models are lacking. ML has shown promising results in predictive modeling, potentially outperforming traditional methods. METHODS: We analyzed data from a prospective database of patients with AP admitted to Vall d'Hebron Hospital from November 2015 to January 2022. Inclusion criteria were adults diagnosed with AP according to the 2012 Atlanta classification. Data included basal characteristics, current medication, and vital signs. We developed ML models to predict SAP, in-hospital mortality, and intensive care unit (ICU) admission. The modeling process included 2 stages: (1) stage 0, which used basal characteristics and medication, and (2) stage 1, which included data from stage 0 and vital signs. RESULTS: Out of 634 cases, 594 were analysed. The stage 0 model showed an area under the curve values of 0.698 for mortality, 0.721 for ICU admission, and 0.707 for persistent organ failure. The stage 1 model improved performance with area under the curve values of 0.849 for mortality, 0.786 for ICU admission, and 0.783 for persistent organ failure. The models demonstrated comparable or superior performance to Acute Physiology and Chronic Health Disease Classification System II and Bedside Index of Severity in Acute Pancreatitis scores. CONCLUSIONS: The ML models showed good predictive capacity for SAP, ICU admission, and mortality using early-stage data without laboratory or imaging tests. This approach could revolutionize initial triage and management of patients with AP, providing a personalized prediction method based on early clinical data.