NPWT Canister Full Detection Using SVM Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current negative pressure wound therapy systems face challenges in accurately detecting a full canister condition without generating false alarms, particularly distinguishing it from other zero-flow conditions such as a blocked suction conduit or a tightly sealed wound dressing, which can lead to unnecessary interruptions in therapy.
Innovation Solution
A method utilizing a support vector machine-based classification algorithm that calculates a negative pressure variation score and correlates it with pump turns to differentiate between a canister full condition and other zero-flow scenarios, eliminating the need for additional sensors like pressure sensors or flow meters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional canister full detection methods are used, then the system can detect when the canister is full, but it generates false alarms by confusing canister full condition with other zero-flow conditions such as blocked suction conduit or tightly sealed wound dressing
Solution Approach 1:
The patent segments the zero-flow condition detection into multiple distinct detection pathways by introducing a canister port pressure sensor that specifically monitors pressure at the canister port. This separate detection mechanism distinguishes canister full conditions from other zero-flow scenarios such as blocked suction conduits or tightly sealed wound dressings, thereby reducing false alarms while maintaining detection accuracy.
2Measurement precision
If additional sensors like pressure sensors or flow meters are added to improve detection accuracy, then the system can reliably distinguish canister full condition from other zero-flow conditions, but the device complexity and cost increase
Solution Approach 1:
The canister port pressure sensor serves multiple functions: it detects canister full conditions, monitors system pressure, and distinguishes between different zero-flow conditions. This multi-functional approach achieves reliable canister full detection without requiring additional specialized sensors, thereby maintaining device simplicity while improving measurement precision.
3Reliability
If the system interrupts therapy upon detecting zero-flow condition, then it prevents operation with a full canister, but it causes unnecessary interruptions when the zero-flow condition is due to other causes such as blocked suction conduit or tightly sealed wound dressing
Solution Approach 1:
The system dynamically adjusts its response to zero-flow conditions based on real-time pressure data from the canister port pressure sensor. When the sensor indicates a canister full condition, the system interrupts therapy; when it indicates other causes such as blocked suction conduit or tightly sealed wound dressing, the system maintains therapy continuity. This dynamic, condition-based response eliminates unnecessary interruptions while ensuring therapy safety.
Data Source
Figure 1
Figure 2a~2b
Figure 2c~2e
AI summary
The present invention relates to a method of determining a canister full condition in a negative pressure wound therapy system during a negative pressure wound therapy. Said method includes determining and recording a number of pump turns associated with an electrical pump used for generating a negative pressure in the negative pressure wound therapy system, determining and recording a plurality of negative pressure values by means of a pressure sensor, calculating and recording a negative pressure variation score by means of the recorded negative pressure values, and generating a first or a second canister full detection data set. Moreover, the method includes executing a classification algorithm which allows to discriminate a canister not full condition from a canister full condition of the negative pressure wound therapy system using the generated data sets. The invention further relates to another method of determining a canister full condition in a negative pressure wound therapy system during a negative pressure wound therapy. Said further method includes generating a negative pressure at a wound site by means of an electrical pump, generating a first or a second canister full detection data set, and executing a classification algorithm, which allows to discriminate a canister not full condition from a canister full condition of the negative pressure wound therapy system using the generated data sets, whereby the classification algorithm includes a support vector machine. The invention further relates to a negative pressure wound therapy system adapted to execute said methods of determining a canister full condition a negative pressure wound therapy system.