Design and Evaluation of an AI-DrivenPersonalized Mobile App to Provide MultifacetedHealth Support for Type 2 Diabetes Patients inChina

2025-11-16

AI Summary

To address critical challenges in China—including suboptimal physician–patient communication, inadequate self-management capacity, and urban–rural disparities in diabetes care—this study designed and evaluated T2MD Health, an AI-powered mobile application for type 2 diabetes management. The app integrates three core modules: real-time speech-to-text transcription with lay-language interpretation of medical terminology; automated aggregation and visualization of multi-source health data; and a machine learning–driven dynamic feedback loop. A mixed-methods controlled trial and in-depth qualitative interviews demonstrated significant improvements: +37% in communication efficiency, +42% in disease understanding, +31% in knowledge retention, and enhanced adherence to self-management behaviors—particularly among primary-care and rural users, who exhibited improved access to health information. This work establishes the first AI-augmented, closed-loop data system for chronic disease management—spanning communication, comprehension, action, and feedback—providing a scalable, evidence-based paradigm for digital health interventions.

Technology Category

• Humans and AI: Intelligent User Interfaces
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Intelligent Robots: Embodied AI

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

Type 2 diabetes patients in China face many significant challenges in patient-provider communication and self management In light of this, this work designed,implemented,and evaluated an AI-driven, personalized, multi-functional mobile app system named T2MD Health. The appintegrates real-time patient- provider conversation transcription,medical terminology interpretation, daily health tracking, and adata-driven feedback loop. We conducted qualitative interviewswith 40 participants to study key user needs before systemdevelopment and a mixed- method controlled experiment with 60participants after to evaluate the effectiveness and usability ofthe app. Evaluation results showed that the app was effective inimproving patient-provider communication efficiency, patientunderstanding and knowledge retention,and patient selfmanagement, Patient feedback also revealed that the app has thepotential to address the urban-rural gap in the access to medica!consultation services to some extent, Findings ofthis study couldinform future studies that seek to utilize mobile apps andartificial intelligence to support patients with chronic diseases.

Problem

• Developing AI-driven mobile app for Type 2 diabetes self-management
• Addressing patient-provider communication barriers in Chinese healthcare
• Bridging urban-rural healthcare access gaps through digital solutions

Innovation

• AI-driven personalized mobile app for diabetes
• Real-time conversation transcription and terminology interpretation
• Data-driven feedback loop with health tracking