Development of an Expert System Calculator for Pediatric Blood Draw: A Conceptual Study

Abstract

Pediatric patients are particularly vulnerable to unnecessary blood loss during diagnostic phlebotomy because of their limited circulating blood volume and frequent laboratory testing. Efficient specimen management is therefore essential to maximize diagnostic information while minimizing blood collection. This study presents a proof-of-concept expert system designed to optimize pediatric blood collection by integrating patient-specific blood-volume constraints with laboratory test and specimen requirements. A rule-based forward-chaining inference engine was developed in Python and implemented it on a Raspberry Pi 3 Model B+ with a touchscreen interface. The knowledge base incorporated test-specific tube requirements, minimum and preferred analytical volumes, tube dead volume, whole-blood requirements, anticoagulant compatibility, and collection time points. Patient-specific total blood volume and maximum allowable blood draw were calculated using weight-based parameters. We evaluated system performance using 20 intentionally constructed simulated pediatric cases involving endocrine test panels. The rule-based system identified lower-volume collection configurations that remained within the calculated maximum allowable blood volume in 15 of 20 cases (75%), with blood-volume reductions ranging from 0.36 to 19.63 mL among feasible cases. Five cases (25%) exceeded the calculated maximum allowable volume, demonstrating conflicts between laboratory minimum-volume requirements and patient-specific blood-volume constraints. Two of these cases included deliberately extreme hematocrit values of approximately 0.90 to stress-test system behavior. These findings demonstrate the computational feasibility of an explainable, rule-based framework for reducing unnecessary blood collection while identifying infeasible testing configurations. Because evaluation was limited to a small, simulated cohort, the findings do not establish clinical effectiveness or generalizability. Future work will focus on formal optimization, dynamic specimen-yield modeling, expanded laboratory testing, and validation using real pediatric clinical data.

Received: 22 November 2025 | Revised: 4 January 2026| Accepted: 21 January 2026

Conflicts of Interest

The authors declare that they have no conflicts of interest to this work.

Data Availability Statement

Data are available from the corresponding author upon reasonable request.

Author Contribution Statement

Sihe Wang: Conceptualization, Methodology, Software, Investigation, Resources, Data curation, Writing – review & editing, Visualization, Supervision, Project administration. Richard Desatnik: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Data curation, Writing – original draft, Writing – review &; editing, Visualization. Motaz Hassan: Validation, Formal analysis, Writing – original draft, Writing – review &; editing, Visualization. Amanpreet Singh Wasir: Validation, Formal analysis, Writing – original draft, Writing – review &; editing, Visualization. Ajay Mahajan: Conceptualization, Methodology, Software, Investigation, Resources, Data curation, Writing – review & editing, Visualization, Supervision, Project administration.