Expert Systems and Double Auction: Enhancing Worm Disease Detection in Children

Muhamad Hariz Muhamad Adnan, Satria Abadi, Hafizul Fahri Hanafi, Che Zalina Zulkifli, Citrawati Jatiningrum, Deshinta Arrova DewiJournal of Advanced Research in Applied Sciences and Engineering Technology 2025 · DOI

Abstract

Helminthiasis is a significant public health problem in Indonesia, with an estimated prevalence ranging from 45 to 65% in 2013, with 19.7% being attributed to A. lumbricoides and 1.5% to T. trichiura. This makes it especially concerning for children of primary school age, as they are particularly at risk. To improve the detection of helminthiasis in children, a research study was conducted to construct an expert system based on four primary characteristics: body weight, eye color, stomach shape, and facial skin color. A questionnaire was created for specialists to fill out and the criteria were then weighed and ranked to identify inconsistencies before being used to detect helminthiasis. To further optimize resource allocation and health outcomes, the concept of double auctions was implemented. This approach suggests that innovative healthcare solutions can be used to increase the efficiency of disease identification and treatment, ultimately leading to improved health outcomes.

Introduction

Childhood helminthiasis, or worm disease, remains a critical health challenge in low-resource settings. In Indonesia, prevalence rates have historically reached up to 65%, leading to severe long-term consequences such as malnutrition, anemia, and impaired cognitive development. The primary barrier to effective intervention is often not the lack of treatment, but rather the difficulty of early detection and the inefficient allocation of medical resources in rural areas.

This paper presents a technological framework designed to bridge this gap through two primary innovations: an AI-driven expert system for non-invasive screening and a "double auction" market model to optimize the distribution of healthcare services.

Methodology

The development of the system followed a structured research planning lifecycle, ensuring a transition from theoretical medical knowledge to a functional digital tool.

The five-phase research planning framework used to develop the system.

The methodology involved five distinct phases:

  1. Literature Review: Analyzing clinical data from 2016–2022 to identify key physical symptoms.
  2. Knowledge Base Construction: Utilizing the Analytical Hierarchy Process (AHP) to weight symptoms based on specialist input.
  3. System Design: Creating the logic for the diagnostic engine and the resource matching framework.
  4. Testing: Validating the internal logic and ensuring the system meets requirements.
  5. Documentation and Maintenance: Preparing the system for deployment and long-term use.

System Architecture and Data Flow

The proposed solution integrates an expert system with a double auction mechanism. The expert system utilizes "IF-THEN" logic to simulate clinical decision-making based on four criteria: body weight, eye color, stomach shape, and facial skin color.

The data flow diagram illustrating how users, admins, and specialists interact with the detection system.

As shown in the data flow diagram, the system facilitates interaction between three primary stakeholders:

  • Users (Parents/Teachers): Input patient data and symptoms to receive detection results.
  • Admins: Manage the knowledge base, including symptom data and rules.
  • Medical Specialists: Provide the underlying expertise and receive patient data for further clinical action.

The Double Auction Framework acts as a market coordinator. Healthcare providers (sellers) bid their services and costs, while patients (bidders) are matched based on the urgency determined by the expert system. This ensures that limited medical resources are directed toward the most critical cases.

Key Results and Discussion

The application of AHP revealed a specific hierarchy of importance for physical indicators:

  • Body Weight: 0.300 (Highest priority)
  • Eye Color: 0.250
  • Stomach Shape: 0.235
  • Facial Skin Color: 0.165

By quantifying these criteria on a scale of 1-4, the system categorizes children into four levels of infection risk. For example, a 10-year-old child weighing less than 24kg is assigned a maximum risk score for the weight category, significantly increasing their priority for medical intervention.

Conclusion and Future Work

The integration of expert systems with market-based resource allocation offers a promising path for managing endemic diseases in resource-constrained environments. While the current design is robust, future work should focus on clinical validation against laboratory results and addressing the digital infrastructure requirements needed for real-world deployment in rural areas.

Summary updated

Abstract

Helminthiasis is a significant public health problem in Indonesia, with an estimated prevalence ranging from 45 to 65% in 2013, with 19.7% being attributed to A. lumbricoides and 1.5% to T. trichiura. This makes it especially concerning for children of primary school age, as they are particularly at risk. To improve the detection of helminthiasis in children, a research study was conducted to construct an expert system based on four primary characteristics: body weight, eye color, stomach shape, and facial skin color. A questionnaire was created for specialists to fill out and the criteria were then weighed and ranked to identify inconsistencies before being used to detect helminthiasis. To further optimize resource allocation and health outcomes, the concept of double auctions was implemented. This approach suggests that innovative healthcare solutions can be used to increase the efficiency of disease identification and treatment, ultimately leading to improved health outcomes.
BibTeX
@article{adnan2025expert,
  author = {Muhamad Hariz Muhamad Adnan and Satria Abadi and Hafizul Fahri Hanafi and Che Zalina Zulkifli and Citrawati Jatiningrum and Deshinta Arrova Dewi},
  title = {{Expert Systems and Double Auction: Enhancing Worm Disease Detection in Children}},
  journal = {Journal of Advanced Research in Applied Sciences and Engineering Technology},
  year = {2025},
  doi = {10.37934/araset.63.4.114},
}
Figure