EFFECTIVENESS OF ARTIFICIAL INTELLIGENCE MODELS IN ARTERIAL BLOOD GASES (ABG, S) INTERPRETATION FOR CLINICAL DECISIONS MAKING IN NURSING PRACTICE

Abstract

Background: In Critically ill patients in intensive care units and Emergency departments the evaluation of Arterial blood gas is very crucial for nurses, as it provides essential information’s about acid–base metabolism of a body and respiratory balance, but its evaluation can be complex and require advance clinical knowledge. ABG,s interpretation provides vital insights into the respiratory and metabolic status of a patient by measuring key parameters such as pH, partial pressure of carbon dioxide (pCO₂), partial pressure of oxygen (pO₂), and bicarbonate (HCO₃⁻) levels. Aim: This study aims the comparing of Artificial Intelligent based machine learning models for the interpretation of acid–base metabolism from ABG,s data more Effectively. Method: Applied research with a cross-sectional observational design was used using Artificial intelligent based supervised machine learning approach for driving and analysis of data. Results: The results showed that using artificial intelligent different machine learning models were compared for accuracy, precision and Data Recall. Random forest model accuracy 92% and precision 92% and data Recall 87%, Decision Tree Model with 83%,78% and 79% accuracy, precision and Data recall respectively were the models with high accuracy, precision and Data Recall. Conclusion: Artificial intelligent and machine learning models are the effective way to interpret the atrial blood gases for the Quick management of patient respiratory or metabolic acidosis and alkalosis during in an emergency and improve nursing practices.