MUMBAI, India, Jan. 23 -- Intellectual Property India has published a patent application (202641001533 A) filed by Koneru Lakshmaiah Education Foundation, Guntur, Andhra Pradesh, on Jan. 7, for 'a real-time clinical prediction system for hemodialysis using ensemble probabilistic model.'
Inventor(s) include T. Hemalatha; and K. V. D Kiran.
The application for the patent was published on Jan. 23, under issue no. 04/2026.
According to the abstract released by the Intellectual Property India: "A Real-Time Clinical Prediction System for Hemodialysis Using Ensemble Probabilistic Model A computer-implemented real-time clinical prediction system and method for risk assessment and outcome prediction in hemodialysis treatment is disclosed. The system integrates heterogeneous clinical data from multiple sources including dialysis machines, patient monitoring devices, laboratory information systems, and electronic health records to generate a unified and temporally synchronized patient state representation. The data undergoes stochastic preprocessing, followed by feature engineering and selection based on temporal curvature and importance scoring. An ensemble probabilistic modeling framework is employed, wherein a plurality of base models is trained independently, their uncertainties quantified through entropy, and outputs fused multiplicatively using entropy-based weights to produce robust probabilistic risk densities for intradialytic adverse events such as blood pressure instability and physiological abnormalities. The system further performs risk inference by computing expected risk and uncertainty-adjusted energy, maps these to discrete risk categories, and generates real-time alerts based on temporal acceleration of risk trajectories. Visualizations and clinician decision support are provided through intuitive dashboards and notification mechanisms. Additionally, a feedback integration mechanism adaptively refines the ensemble models based on clinical interventions, ensuring continuous improvement and stability. The invention enables proactive detection of abnormal clinical patterns, risk stratification, and timely intervention, thereby enhancing patient safety and treatment efficacy in chronic hemodialysis care. The system is deployable on clinical servers or cloud platforms for seamless integration with live dialysis data."
Disclaimer: Curated by HT Syndication.
