On February 24, Tencent Health announced its full integration with the DeepSeek large model through Tencent Cloud, teaming up with Tencent's self-developed Hongyuan large model to comprehensively upgrade medical services. This upgrade covers various scenarios including intelligent triage, pre-consultation, health Q&A, smart medication, symptom self-check, imaging report interpretation, and quality control. The aim is to help over 1,000 hospitals across the country rapidly enhance their level of intelligent applications, providing multi-model deployment and application solutions for medical institutions, health examination centers, medical device companies, and gene sequencing companies.
This upgrade focuses on the entire medical process, from pre-visit health Q&A, symptom self-check, and intelligent triage, to in-visit smart pre-consultation, and post-visit report interpretation and smart medication. Tencent Health is gradually transitioning its services to a "multi-model" mode. Through AI assistants, hospitals can offer personalized medical plans and health management to patients while providing decision support to doctors, including medical history collection, disease prediction, treatment plan recommendations, and medication safety monitoring. This not only enhances the patient experience but also improves doctors' diagnostic efficiency and the quality of medical services.
In the field of medical imaging, the "Xiao Mi AI Assistant" can quickly analyze imaging reports, automatically retrieve historical reports, and compare changes in conditions. It can also recommend relevant examinations based on diagnostic conclusions, interpret the meaning of reported diseases, and support typo detection and one-click correction, significantly enhancing doctors' report writing efficiency.
Moreover, Tencent Health is advancing the "dual model" combination of DeepSeek and Hongyuan in the fields of genomics and medical device digitization. The omics platform AI assistant deeply integrates the bioinformatics analysis workflow development framework and platform knowledge base to provide intelligent assistance throughout the bioinformatics analysis process. Meanwhile, NGES creates an AI solution for medical device companies that combines intelligent Q&A, knowledge management, and real-time information acquisition to support their digital transformation.
To meet the diverse needs of different institutions, Tencent Health provides various methods including Tencent Cloud API, large model knowledge engine calls, and TI-ONE platform deployment, offering public SaaS, API PaaS, and multiple sizes of private deployment models, flexibly adapting to various servers and operating environments. At the same time, Tencent Health conducts deep customization based on specific institutional needs, building a medical professional knowledge base using RAG technology and integrating the DeepSeek and Hongyuan dual foundation large models. Tencent Health itself trains a dedicated medical large model using over 100 billion medical tokens, enhancing retrieval with a knowledge graph covering 98% of diseases, reducing the risk of generating false information, and introducing a feedback-based reward mechanism to make model responses more professional and closer to patient needs.
