Fangzhou Launches XingShi Medical LLM to Scale Precision Chronic Disease Management

Fangzhou launches "XingShi" medical LLM to streamline chronic disease management

Fangzhou Inc. (06086.HK) introduced its "XingShi" Large Language Model (XS LLM) at the 10th H2H Healthcare Ecosystem Conference. The system focuses on chronic disease management, where operational inefficiency, fragmented workflows, and limited physician time block scale. XS LLM is built to increase service efficiency, personalize patient interactions, and help clinicians focus on higher-value care.

The model combines multimodal capabilities-image and speech recognition, natural language processing, large-scale medical knowledge storage, and reasoning. It is positioned as the core digital brain of Fangzhou's platform, supporting a closed-loop service model from patient intake through follow-up.

What XS LLM does

• AI Knowledge Agent: Curates and reasons over medical guidelines, literature, and structured knowledge to support consistent, evidence-based decisions.
• AI Guidance Agent: Directs patients to the right care path, provides education, and drives adherence across the care plan.
• AI Pre-Consult Agent: Gathers histories, symptoms, and structured data before visits to shorten cycle time and reduce administrative friction.
• AI Doctor Assistant: Summarizes notes, suggests differentials, drafts orders, and supports documentation so clinicians can spend more time with patients.
• AI-EMR Agent: Translates unstructured inputs into structured records, updates problem lists, and supports coding and quality checks.

Products now live on the platform

• AI Medication Finder: Real-time medication guidance, alternatives, and interaction checks for safer prescribing.
• AI Health Manager: Ongoing monitoring and education for chronic conditions, with timely nudges to improve adherence.
• AI Doctor Assistant: Provider-facing support for documentation, retrieval of prior data, and clinical reasoning aids.
• AI Academic Assistant: Faster literature review and guideline lookup to keep teams current.
• AI-Powered Search: Enterprise-grade retrieval across knowledge bases, EMR excerpts, and care protocols.

Why this matters for healthcare management

Chronic diseases consume resources across long timelines and multiple touchpoints. AI that captures structured data up front, supports guideline adherence, and automates documentation can remove bottlenecks where costs accumulate. For context on the global burden of noncommunicable diseases, see the World Health Organization overview here.

• Capacity and access: Free physician time by shifting repetitive tasks to AI agents and standardizing pre-visit intake.
• Consistency and safety: Use knowledge-grounded reasoning to reduce variation and surface guideline-concordant options.
• Patient stickiness: Personalized education and reminders improve adherence and reduce avoidable visits.
• Cost discipline: Fewer reworks and faster documentation lower administrative spend per encounter.

Implementation playbook

• Start with two pilots: One patient-facing (e.g., hypertension coaching), one provider-facing (e.g., pre-consult intake for diabetes). Define entry and exit criteria.
• Integrate with your EMR: Use APIs and FHIR/HL7 where available. Focus on problem lists, meds, vitals, labs, and care plans for fast impact.
• Set safety rails: Human-in-the-loop review for clinical outputs, source citation visibility, and hard stops for high-risk decisions.
• Measure what moves cost and quality: Average time to complete notes, physician after-hours charting time, no-show rate, refill issues, readmissions within 30 days, and guideline-concordant care rates.
• Train your teams: Short sessions on prompt patterns, exception handling, and escalation paths. Consider targeted upskilling resources for managers and clinicians here.

Data governance and risk

• Privacy and security: Define permitted data flows, apply de-identification where possible, and restrict model outputs containing PHI.
• Bias and auditability: Monitor outputs by demographic segments; log prompts, sources, and decisions for audits.
• Regulatory posture: Align usage with local clinical AI guidance, document intended use, and keep a post-deployment monitoring plan.
• Change management: Communicate clearly: what the AI does, what it does not, and how responsibility is assigned across the care team.

Strategic context

Fangzhou positions XS LLM as the technical base layer for its H2H (Hospital-to-Home) ecosystem. With China's "AI+" policy encouraging industrial adoption, the company indicates future R&D will focus on chronic disease scenarios where scale and personalization matter most.

About Fangzhou Inc.

Fangzhou Inc. (06086.HK) is an online chronic disease management platform in China, serving 52.8 million registered users and 229,000 physicians as of June 30, 2025. The company provides AI-enabled precision medicine solutions across patient education, medication guidance, and clinical support. Learn more at investors.jianke.com.