Infusion System Occlusion Detection Using AI
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional infusion systems face challenges in timely detection of occlusions at low infusion rates, often resulting in delayed identification and high false alarm rates due to interference from disturbance variables like static friction and plastic disposable variations.
Innovation Solution
An infusion system equipped with multiple sensors and an AI unit using statistical models to calculate target variables such as absolute pressure, occlusion probability, and occlusion location, which reduces false alarms by distinguishing true occlusions from disturbances through pattern recognition and trend analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If fixed threshold values are used for occlusion detection, then the system is simple to operate, but occlusions at low infusion rates are detected too late
Solution Approach 1:
The system transitions from static fixed threshold values to dynamic adaptive thresholds that automatically adjust based on infusion rate. The control unit modifies detection thresholds in real-time according to the current infusion rate, enabling timely occlusion detection at low rates while maintaining system simplicity.
Solution Approach 2:
The detection threshold parameter is changed dynamically based on infusion rate. Instead of using a single fixed threshold, the system adjusts the threshold parameter adaptively - using lower thresholds at low infusion rates and higher thresholds at high infusion rates, thereby optimizing detection timing across different operating conditions.
2Measurement precision
If very low threshold values are used for occlusion detection, then occlusion detection sensitivity is improved, but false alarm rate increases
Solution Approach 1:
The detection threshold parameter is adaptively changed based on infusion rate. At low infusion rates, lower thresholds are applied to maintain sensitivity, while at high infusion rates, higher thresholds are used to reduce false alarms. This dynamic parameter adjustment resolves the contradiction between sensitivity and reliability.
Solution Approach 2:
The system employs dynamic threshold adjustment rather than static thresholds. The control unit continuously adapts the detection threshold to match current operating conditions, particularly infusion rate, thereby maintaining optimal detection sensitivity while minimizing false alarms across varying operational states.
3Reliability
If multiple sensors and AI processing are added to improve occlusion detection accuracy, then detection reliability is improved, but device complexity increases
Solution Approach 1:
The control unit is designed to perform multiple functions: it manages infusion pump operation, processes sensor data, adjusts detection thresholds dynamically, and generates alarms. By making the control unit multi-functional, the system achieves high detection reliability without adding separate dedicated components for each function.
Solution Approach 2:
The system combines multiple sensors (pressure sensor, force sensor, motor current sensor, position sensor) and integrates their data processing within a single control unit. The control unit merges sensor data processing, threshold adaptation, and alarm generation functions, achieving reliable occlusion detection while minimizing structural complexity through functional integration.
Data Source
AI summary
An infusion system includes an infusion pump configured to convey a fluid to be administered through a fluid conduit, a number of sensors for producing associated sensor data, and an AI unit configured to calculate a number of target variables depending on the number of sensor data using a statistical model as a basis, the statistical model having been trained using training data.