Enabling Health Information Recommendation Using Crowdsourced Refinement in Web-Based Health Information...

Area of Science:

• Digital Health
• Health Informatics
• Software Engineering

Background:

• Search engines and social media often provide biased health information due to commercial algorithms.
• Web-based health applications offer a trusted alternative for health information access.
• A research gap exists in integrating effective content-ranking algorithms into specialized health applications for personalized delivery.

Purpose of the Study:

• To introduce a generic methodology for developing and implementing health information recommendation features in web-based health applications.
• To provide a framework for creating personalized and relevant health content delivery systems.

Main Methods:

• A comprehensive methodology covering design, development, operation, review, and optimization was proposed.
• The methodology was applied to the EndoZone platform, creating the EndoZone Informatics recommendation system for endometriosis information.
• The case study demonstrated the practical implementation of recommendation functionalities.

Main Results:

• The EndoZone Informatics system successfully delivers personalized health content, incorporating user feedback effectively.
• The methodology proved useful for enabling advanced recommendation features, with flexibility in adjusting recommendation logic.
• Initial design flaws were identified and rectified during the review and optimization stages.

Conclusions:

• A generic methodology for designing and implementing health information recommendation functionality in web applications was developed.
• This approach leverages user data and feedback for personalized content ranking within trusted health platforms.
• The successful implementation in EndoZone Informatics advances personalized health information delivery at scale.