Machine learning model based on RCA-PDCA nursing methods and differentiating factors to predict hypotension during...

Area of Science:

• Anesthesiology
• Nursing Science
• Medical Informatics

Background:

• Intraoperative hypotension is a serious complication during cesarean sections, often managed with drugs that have side effects.
• The Root Cause Analysis-Plan, Do, Check, Act (RCA-PDCA) model offers a systematic approach to identify and address problem causes.
• This study explores the application of RCA-PDCA in managing hypotension during cesarean delivery.

Purpose of the Study:

• To evaluate the effectiveness of the RCA-PDCA nursing model in preventing intraoperative hypotension during cesarean sections.
• To develop and assess a machine learning model for predicting intraoperative hypotension in cesarean delivery patients.

Main Methods:

• Retrospective analysis of cesarean section patients' data, including vital signs, demographics, and outcomes.
• Screening of statistically significant features influencing hypotension.
• Development and evaluation of five machine learning models for predictive analysis.

Main Results:

• The RCA-PDCA nursing model significantly reduced the incidence of intraoperative hypotension and postoperative complications compared to general nursing.
• The Random Forest (RF) machine learning model demonstrated the highest predictive performance with 90% accuracy in preventing intraoperative hypotension.
• Patient experience, including comfort and satisfaction, improved with the RCA-PDCA model.

Conclusions:

• The RCA-PDCA nursing method is crucial for preventing intraoperative hypotension in cesarean sections.
• The Random Forest model shows significant promise for predicting hypotension, advancing AI in medical analysis.
• This integrated approach enhances maternal and fetal safety during cesarean delivery.