MUMBAI, India, Sept. 14 -- Intellectual Property India has published a patent application (202641106620 A) filed by Dr. Kavitha S; Kalavathi. P; Ms. S Sandhya; Boddu Balaji; Dr. V. Subramanyam; Dr. Rupasree Peruru; Ms. R Ramya; Dr. C. Priyanka; Mr. Heamchandar B; and Preethi. P on September 04, 2026, for Clinagent: Multi-Agent System For Autonomous Hospital Discharge Management.

Inventors include Dr. Kavitha S; Boddu Balaji; Dr. V. Subramanyam; Dr. Rupasree Peruru; Ms. R Ramya; Dr. C. Priyanka; Mr. Heamchandar B; Preethi. P; Kalavathi. P; and Ms. S Sandhya.

The application for the patent was published on September 11, 2026, under issue no. 37/2026.

Abstract: Hospital discharge processes have an inherent cross-departmental nature. Traditional Hospital Information System (HIS) and Electronic Health Record (EHR) systems generally suffer from two key flaws: they lack autonomous reasoning capabilities and have insufficient cross-departmental coordination. To address these limitations, this paper proposes a self-developed multi-agent architecture, ClinAgent: it includes six types of business agents for nursing, pharmacy, billing, laboratory, medical insurance, and patient services, paired with one central coordination agent that uses fuzzy logic, neural network prediction, and reinforcement learning to implement process reasoning. Communication between all agents follows the standard FIPA-ACL protocol. This study built a simulated hospital environment using Python, integrating custom discrete event simulation, scikit-learn-style fuzzy inference, PyTorch, and NumPy stochastic modeling tools. This study conducted a controlled experiment to verify the performance of the ClinAgent intelligent medical scheduling system. The study adopted a hybrid evaluation framework. One hundred anonymized patient records from the MIMIC-IV database were used for validation, while 5,000 synthetic discharge cases were generated using a custom workflow simulation framework informed by healthcare workflow statistics. Four quantitative performance improvements were ultimately measure,. Results showed 57.1% reduction in discharge duration, 59.3% reduction in patient waiting time, 98.6% task completion rate within six hours, and 20% improvement in workload balance. These findings confirm that multi-agent systems can effectively restructure traditional medical workflows.

Disclaimer: Curated by HT Syndication.