AI-Assisted Blood Gas Interpretation: A Comparative Study With an Emergency Physician

Abstract

Background: Blood gas interpretation is critical in emergency settings. Large language models like ChatGPT are increasingly used in clinical contexts, but their accuracy in interpreting arterial blood gases (ABGs) requires further validation.

Objective: To evaluate ChatGPT's interpretive concordance with an emergency physician across 25 theoretical ABG scenarios.

Methods: ABG cases covering respiratory and metabolic emergencies (e.g., COPD, DKA, AKI, sepsis, poisoning) were analyzed by both ChatGPT and a specialist. Five interpretation criteria were used: pH, primary disorder, compensation, likely diagnosis, and clinical recommendation.

Results: Concordance was ≥90% in COPD, asthma, and pulmonary edema; 80-90% in DKA, AKI, and lactic acidosis; <70% in toxicologic and mixed acid-base cases. ChatGPT's recommendations were clinically safe even when diagnostic clarity was limited.

Conclusion: ChatGPT shows high concordance with clinical interpretation in typical ABG cases but has limitations in complex or contextual diagnoses. These findings support its potential as a supportive tool in emergency medicine.

Keywords: Arterial blood gas; Artificial intelligence; ChatGPT; Clinical decision support; Diagnostic concordance; Emergency medicine.