Target-Controlled Infusion Occlusion Correction With Model Rollback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing target-controlled infusion systems fail to accurately model drug distribution in a patient's body when an occlusion occurs in the infusion line, leading to incorrect drug concentration modeling due to delayed detection of the occlusion.
Innovation Solution
A system and method that store parameters derived from a mathematical model at multiple points in time, allowing the control device to revert to a previous set of parameters upon occlusion detection, accurately modeling drug distribution by assuming no drug reaches the patient during the occlusion period.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If occlusion detection is performed by monitoring pressure in the infusion line, then occlusion can be detected, but detection is delayed until pressure exceeds a threshold
Solution Approach 1:
The system performs preliminary actions by storing mathematical model parameters at multiple time points before occlusion occurs. When occlusion is detected, the system can immediately revert to the most recent stored parameters, eliminating the need to wait for pressure threshold exceedance and reducing detection delay.
Solution Approach 2:
The system implements feedback by continuously monitoring pressure and using this information to determine when to revert to previously stored model parameters. This closed-loop approach allows the system to respond dynamically to occlusion conditions while maintaining accurate drug concentration modeling.
2Device complexity
If the mathematical model continues to assume drug delivery after occlusion occurs, then modeling simplicity is maintained, but drug concentration accuracy deteriorates
Solution Approach 1:
The system extracts the inaccurate assumptions from the mathematical model by detecting occlusion conditions and separating the pre-occlusion model parameters from post-occlusion operation. This allows the system to maintain a simple continuous modeling approach while removing the harmful assumption that drug delivery continues during occlusion.
Solution Approach 2:
The system discards the inaccurate model parameters that assume continuous drug delivery and recovers the correct parameters by reverting to previously stored values. This enables the system to maintain modeling simplicity while ensuring accuracy by periodically discarding and recovering appropriate parameter sets based on occlusion detection.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A system for controlling a target-controlled infusion for administering a drug to a patient (P), comprises at least one infusion device (31-33) for administering a drug to the patient (P), the at least one infusion device (31, 33) comprising a sensor device (311, 321, 331) for measuring a value indicative of a pressure in an infusion line (310, 320, 330) connected to the at least one infusion device (31-33) for delivering said drug to the patient (P). A control device (2) is configured to control operation of the at least one infusion device (31-33) in order to establish a drug concentration within the patient (P) based on a target concentration, wherein the control device (2) is configured to execute a target-controlled infusion protocol using a mathematical model modelling a drug distribution in the patient's body for controlling operation of the at least one infusion device (31-33). The control device (2) is further configured to detect an occlusion in the infusion line (310, 320, 330) connected to the at least one infusion device (31-33) based on said value measured by the sensor device (311, 321, 331). Herein, the control device (2) is configured to store, at each of a multiplicity of points in time (Ti...Ti+3) during execution of said target-controlled infusion protocol, a set of parameters derived from said mathematical modelin a memory (21), wherein the control device (2) is configured, in case an occlusion is detected, to use the set of parameters of a point in time (Ti...Ti+3) prior to the detection of the occlusion in the mathematical model to model the drug distribution in the patient's body based on the set of parameters of the point in time (Ti...Ti+3) prior to the detection of the occlusion.Fig. 9D