Protocol for a randomized controlled trial evaluating the artificial intelligence health education accurately linking system in patients with mild-to-moderate stroke

Abstract

Background: Stroke is a leading cause of death and disability worldwide. Although survival rates from mild-to-moderate stroke are high, long-term functional impairment remains common, requiring sustained self-management beyond traditional rehabilitation. Conventional models depend on institutional medical care, which not only drives up costs but also disrupts continuity of care. Meanwhile, psychological, risk-related, and behavioral factors are often overlooked. Advances in artificial intelligence (AI) and mobile health provide opportunities for individualized, long-term support. Based on this, we developed the AI Health Education Accurately Linking System (AI-HEALS) to evaluate its potential to improve physiological parameters, risk perception, and self-management in patients with mild-to-moderate stroke. Methods: This single-blind randomized controlled trial evaluates AI-HEALS, delivered via WeChat (China's most widely used social media app), to improve the monitoring of key physiological indicators in patients with mild-to-moderate stroke. Eligible participants are randomly allocated either standard care as a control or standard care plus a three-month regimen of AI-HEALS. It features an AI-powered interactive Q&A system that supports both voice and text communication, real-time monitoring of physiological and behavioral indicators, personalized health reminders, and specially designed educational content. These are all offered through the official WeChat account "Stroke Health Management Expert." The primary outcomes are changes in blood pressure, glucose, and blood lipids. Secondary outcomes include risk perception of recurrence of stroke, self-management behaviors, and psychological state of mind. Follow-up assessments are conducted at 3, 6, and 9 months after completion of the intervention to evaluate both short-term and sustained effects. Discussion: This protocol presents a new AI-mHealth approach to delivering stroke care. If proven feasible and effective, AI-HEALS could offer a scalable and sustainable model for improving long-term health outcomes, reducing the risk of recurrence, and optimizing the use of healthcare resources for stroke and other chronic conditions. Clinical trial registration: https://www.chictr.org.cn/showproj.html?proj=251515, Identifier, ChiCTR2500096422.

Ähnliche Arbeiten