A Visual Decision Support Tool for Appendectomy Care

Area of Science:

• Pediatric Surgery
• Clinical Decision Support Systems
• Predictive Analytics in Healthcare

Background:

• Appendectomy is the most frequent abdominal surgery in U.S. children.
• Effective postoperative management of pediatric appendicitis requires robust decision-making tools.
• Current treatment planning can benefit from data-driven predictive insights.

Purpose of the Study:

• To develop a predictive statistical model for pediatric appendicitis outcomes.
• To create a prototype decision aid application for clinical use.
• To enhance postoperative care planning and patient communication.

Main Methods:

• Developed a predictive model trained on 3724 anonymized pediatric appendicitis care records.
• Evaluated the model on a separate dataset of 2205 cases from a tertiary care center.
• Built an interactive decision support tool with visualization techniques for clinician workflows.

Main Results:

• The predictive model demonstrated high specificity (97.0%).
• Achieved true sensitivity of 25.1% and precision of 58.8%.
• The decision support tool is designed for ease of use and integration into clinical practice.

Conclusions:

• The developed model and decision aid show potential for improving postoperative management of pediatric appendicitis.
• The tool supports clinicians in making care decisions and communicating with patients and stakeholders.
• The system is designed for continuous improvement with evolving data.