JMIR Publications

ABSTRACT

Background:

Effective patient education is crucial in preventing venous thromboembolism (VTE), improving patient outcomes, and reducing healthcare costs. Traditional educational methods, however, often lack engagement and fail to address individual patient needs comprehensively.

Objective:

This study aimed to develop and preliminarily validate an immersive, large language model (LLM)-based patient education system for VTE.

Methods:

An interactive intelligent patient education platform, ChatVTE, was engineered by integrating a retrieval-augmented large language model (LLM; Qwen1.5-7B) with Text-to-Speech (TTS) and Lip-Synchronization (Lip-Sync) technologies. The system’s performance was initially assessed through a comparative evaluation against Chat Generative Pre-trained Transformer (ChatGPT). This involved utilizing a standardized set of VTE-related questions, administered from December 10 to 31, 2024, with responses rigorously evaluated by four VTE domain experts using a 5-point Likert scale for accuracy, completeness, consistency, and safety. In a subsequent phase, a prospective cohort of 25 adult inpatients diagnosed with VTE was consecutively enrolled from the Departments of Pulmonary Vascular and Thrombotic Diseases and General Surgery at the Sixth Medical Center of the Chinese People’s Liberation Army General Hospital between March 1 and May 31, 2025. These participants engaged with the ChatVTE system throughout their inpatient stay and completed post-intervention assessments upon discharge.

Results:

Expert evaluation demonstrated ChatVTE’s superior performance over ChatGPT in terms of accuracy, completeness, consistency (all P<.001), and safety (P=.009). Among the 25 enrolled patients (mean age: 55.4±13.2 years), ChatVTE garnered high average scores (mean score >4.0/5.0) across 8 of 9 evaluated experience dimensions; however, the emotional support domain received a comparatively lower score (1.92/5.0).

Conclusions:

This study validates the feasibility of ChatVTE in the management of VTE patients, demonstrating its potential to enhance the quality of patient-provider interaction and the efficacy of knowledge dissemination. The system exhibits promising application prospects in both patient education and the facilitation of shared clinical decision-making.