Facing the post-COVID-19 challenges of healthcare workforce shortages and rising clinical care burdens, Chi Mei Medical Center has developed the “A+ Clinical Guardian Net System,” Taiwan’s first AI-driven platform integrating generative AI (GPT-4) into inpatient pharmaceutical care and patient safety management.
The system automatically consolidates electronic medical records, laboratory results, and medication data daily, enabling real-time generation of personalized medication risk summaries and care recommendations. This allows pharmacists to quickly identify high-risk patients and prioritize interventions, thereby improving clinical efficiency and care quality.
At the core of the system is the “AI Medication Safety Guard” module, which has been fully embedded into pharmacists’ daily workflows. It features real-time generation, automated scheduling, risk-based prioritization, and multi-criteria filtering. The system also incorporates Global Trigger tools to detect potential adverse events, enhancing patient safety surveillance. Initial outcomes show a reduction in average pharmacist evaluation time from 10 minutes to 6 minutes per patient, saving up to 1.3 hours per day while significantly enhancing high-risk case identification.
In addition to medication management, the Guardian Net System also extends to patient safety alerts and healthcare quality monitoring, supporting multidisciplinary clinical decision-making and forming an integrated intelligent care network.
Built on the Microsoft Azure cloud platform, the system offers high-performance computing and compliance-grade data security. It integrates the Medispan drug database to enhance dosage checking and drug interaction alerts. A flexible prompt design mechanism ensures content accuracy and consistency even as AI models are updated.

1. Web API interaction
2. Modular design
3. In-house development team
4. Model-independent design
5. Compliant with International Clinical Standards