System and method of deploying neural network in optimizing vein characteristic during apheresis
A neural network system optimizes fluid flow rates in apheresis by using donor-specific input parameters and supervised classification to predict vein characteristics, addressing the limitations of existing systems and enhancing vein monitoring efficiency.
Patent Information
- Application Number
- PCT/US2026/015304
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2025-02-19
- Filing Date
- 2026-02-13
- Publication Date
- 2026-08-27
AI Technical Summary
Existing apheresis systems lack donor-specific predictive capabilities and adaptive vein monitoring, often resulting in slower than desired flow rates due to reliance on a constant initial flow rate and failing to account for individual donor characteristics.
A neural network system is deployed to optimize fluid flow rates by receiving donor-specific input parameters, applying activation functions, selecting relevant input layers based on data qualification thresholds, and adjusting weights through supervised classification to predict optimal vein characteristics.
The system enables accurate, donor-specific optimization of fluid flow rates during apheresis procedures, reducing computational resources and improving system response times while enhancing vein monitoring.