[Development and practice evaluation of blood acid-base imbalance analysis software]

Area of Science:

• Medical Informatics
• Clinical Chemistry
• Computational Medicine

Background:

• Acid-base imbalance diagnosis relies on complex calculations from blood gas and electrolyte data.
• Manual analysis can be time-consuming and prone to errors, impacting timely patient care.

Purpose of the Study:

• To develop and evaluate a computer-aided diagnostic software for systematic, rapid, and accurate acid-base imbalance analysis.
• To assess the clinical utility and efficiency of the developed software.

Main Methods:

• VBA programming language was used to create the diagnostic software.
• Retrospective analysis of arterial blood gas (pH, HCO(3)(-), PaCO₂) and electrolyte (Na⁺, Cl⁻) data from 220 patients.
• Comparison of software-generated diagnoses with manual calculations using the H-H compensation formula.

Main Results:

• The software achieved a total accuracy of 95.5% in diagnosing acid-base imbalance types.
• High consistency was observed between software and manual judgments (Kappa = 0.935, P=0.000).
• Software analysis time (18.14 ± 3.80 seconds) was significantly shorter than manual analysis (43.79 ± 23.86 seconds).

Conclusions:

• The developed software provides a rapid, accurate, and convenient alternative to manual acid-base imbalance diagnosis.
• Implementation of this software can enhance clinical work efficiency and diagnostic quality.
• The tool holds significant clinical application and promotion value for healthcare professionals.