HDLalert - a healthcare DL alerting system: from user needs to implementation

Area of Science:

• Health Informatics
• Medical Knowledge Management
• User-Centered Design

Background:

• Clinicians and patients need to stay updated with evolving medical knowledge.
• Current medical alert systems often provide irrelevant or excessive information.
• Limited user control over information selection leads to overload and inefficiency.

Purpose of the Study:

• To address the limitations of existing medical alert systems.
• To investigate user needs for tracking medical knowledge changes.
• To develop a novel alerting architecture for personalized information delivery.

Main Methods:

• Conducted international user studies with healthcare professionals and patients in the UK, Germany, and New Zealand.
• Gathered perceptions on the relevance and selection of medical information.
• Developed and evaluated a new context-aware alerting architecture.

Main Results:

• User studies highlighted the critical importance of user context (task and environment) for information relevance.
• The novel architecture demonstrated the ability to deliver finely tailored alert streams.
• The system provides enhanced support for interpreting received medical information.

Conclusions:

• Context-aware design is crucial for effective medical alerting systems.
• A personalized approach to information delivery can significantly improve user experience and knowledge management.
• The proposed architecture offers a promising solution for managing medical knowledge updates.