Harnessing an Artificial Intelligence-Based Large Language Model With Personal Health Record Capability for...

Area of Science:

• Cardiology
• Digital Health
• Artificial Intelligence in Healthcare

Background:

• Myocardial infarction (MI) is a major global health concern, with recovery management posing significant challenges.
• Existing support systems offer generic advice, failing to address individual patient needs.
• AI-powered large language models (LLMs) offer potential for personalized post-MI health information delivery.

Purpose of the Study:

• To evaluate user experiences and perceptions of the iflyhealth AI tool for post-MI care.
• To assess patient and family engagement with the tool during recovery.
• To identify facilitators and barriers to the adoption and continued use of the iflyhealth app.

Main Methods:

• A qualitative study involving 20 participants (12 users, 8 nonusers) post-MI surgery.
• Semistructured interviews were conducted to explore first impressions, usage, and adoption/nonadoption reasons.
• Data were analyzed using Colaizzi's method to identify key themes.

Main Results:

• Participant experiences varied based on digital literacy, prior tech exposure, and recovery needs.
• Users valued enhanced access to health information, personalized advice, and responsiveness to health changes.
• Barriers included digital literacy issues, usability concerns, data privacy, interface complexity, and perceived relevance of advice.

Conclusions:

• The iflyhealth AI app shows promise for post-MI patient support, offering personalized information and improved responsiveness.
• Key benefits include enhanced access to tailored health advice and better adaptation to changing conditions.
• Addressing challenges in digital literacy, usability, and privacy is crucial for maximizing the app's potential in patient-centered MI management.