Machine-learning based prediction model for acute kidney injury induced by multiple wasp stings

Area of Science:

• Nephrology
• Toxicology
• Machine Learning in Medicine

Background:

• Acute kidney injury (AKI) is a severe complication of multiple wasp stings.
• Predictive models for wasp sting-related AKI are currently limited.
• Early identification of at-risk individuals is crucial for timely intervention.

Purpose of the Study:

• To develop and validate a machine learning-based clinical prediction model for AKI in patients with wasp stings.
• To identify key prognostic variables associated with AKI development after wasp envenomation.
• To create a user-friendly nomogram for clinical application.

Main Methods:

• Retrospective cohort study of 214 patients with wasp sting injuries.
• Utilized least absolute shrinkage and selection operator (LASSO) and multivariate logistic regression to identify prognostic factors.
• Constructed and validated a nomogram using internal validation, cross-validation, NRI, IDI, and DCA.

Main Results:

• 34.6% of patients developed AKI.
• Key predictors identified: number of stings, gross hematuria, SIRI, and platelet count.
• The nomogram demonstrated good predictive accuracy (AUC: 0.757) and discrimination.

Conclusions:

• The developed nomogram is a reliable tool for predicting AKI risk in wasp sting patients.
• This model facilitates early risk stratification and management.
• Further validation in diverse populations may enhance its generalizability.