UDEC-MO: an uncertainty-guided deep embedded clustering framework for bulk and single-cell multi-omics data.

The prevalence of type 2 diabetes mellitus (T2DM) continues to rise and has emerged as a major global public health challenge []. This chronic condition not only imposes a substantial burden on individual health, marked by increased risk of disability, reduced quality of life, and shortened life expectancy [], but also results in significant medical and socioeconomic costs []. According to the International Diabetes Federation, China has the highest number of people with diabetes and diabetes-related deaths worldwide, and the age of onset is trending younger []. As a lifelong condition, effective management of T2DM largely depends on patients’ self-management capabilities [], which encompass dietary regulation, regular physical activity, medication adherence, blood glucose monitoring, foot care, and routine eye examinations. However, awareness, treatment, and glycemic control rates in China remain suboptimal [], highlighting a substantial gap in diabetes self-management practices. Therefore, the implementation of professionalized, standardized, and evidence-based self-management strategies is critically important for individuals living with T2DM. In recent years, digital health interventions (DHIs) have shown great promise in supporting diabetes self-management [,]. By leveraging smartphones and other mobile devices, DHIs can overcome the time and space limitations of traditional community-based care, promote behavioral changes, and enhance the accessibility and equity of health services. To this end, our research team developed the Artificial Intelligence-based Health Education Accurately Linking System (AI-HEALS), implemented on the WeChat (Tencent) platform under the name Peking Diabetes Butler. This mobile application integrates knowledge graphs and natural language processing technologies and comprises an AI chatbot, tailored health education delivery, and blood glucose monitoring reminders []. It aims to provide patients with personalized and continuous support for diabetes self-management.