Blood Pump Suction Detection via Flow Waveform Interpolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing suction detection methods for blood pumps in mechanical circulatory support systems are limited by inaccurate false positives, high energy consumption, and the need for extensive signal processing, which is not feasible in embedded systems used for ventricular assist devices.
Innovation Solution
A method that estimates the flow rate of a blood pump by interpolating data sets defining pump power to flow for various pump speed values, using a quadratic equation to identify suction events by locating a suction marker reference point in the flow waveform and calculating a probability of occurrence based on suction marker location.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If waveform correlation methods are used to detect suction events, then detection accuracy is improved, but device complexity and energy consumption increase due to extensive signal processing requirements
Solution Approach 1:
The patent extracts only the essential feature needed for suction detection - the flow waveform shape characteristics - rather than performing comprehensive waveform correlation analysis. By focusing on extracting and analyzing specific waveform features (flow rate, flow shape) during the diastolic phase, the system achieves adequate detection accuracy without requiring extensive signal processing capabilities that would increase device complexity and energy consumption.
2Measurement precision
If waveform correlation methods are used to detect suction events, then detection accuracy is improved, but energy consumption increases due to extensive signal processing
Solution Approach 1:
The patent extracts only the essential feature needed for suction detection - the flow waveform shape characteristics - rather than performing comprehensive waveform correlation analysis. By focusing on extracting and analyzing specific waveform features (flow rate, flow shape) during the diastolic phase, the system achieves adequate detection accuracy without requiring extensive signal processing capabilities that would increase device complexity and energy consumption.
3Device complexity
If other detection methods (non-waveform correlation) are used, then device complexity is reduced, but detection accuracy deteriorates due to inability to discern suction events from other physiological conditions
Solution Approach 1:
The patent applies local quality by focusing analysis on a specific portion of the flow waveform - the diastolic phase - rather than attempting to analyze the entire waveform or use complex correlation methods. By concentrating computational resources on analyzing flow characteristics during the diastolic phase specifically, the system achieves better discrimination of suction events from other physiological conditions while maintaining simpler processing requirements.
4Measurement precision
If complex signal processing is performed, then detection accuracy is improved, but response time increases due to processing delays
Solution Approach 1:
The patent extracts only the essential feature needed for suction detection - the flow waveform shape characteristics - rather than performing comprehensive waveform correlation analysis. By focusing on extracting and analyzing specific waveform features (flow rate, flow shape) during the diastolic phase, the system achieves adequate detection accuracy without requiring extensive signal processing capabilities that would increase device complexity and energy consumption.
Data Source
Figure 1
Figure 2A
Figure 2B
AI summary
A system and method for detecting and mitigating a suction condition are disclosed. The method may include estimating a flow waveform of the pump, identifying pulses in the flow waveform, determining a negative flow based on a valid identification of a pulse, and evaluating a characteristic of the pulse for an existence of a suction condition. In various embodiments, a suction marker is located based on a minimum in a diastolic phase, and the suction marker location is used to identify a probability of a suction condition. A speed of the pump may be adjusted to mitigate the suction condition. A system and method for estimating flow is further disclosed. The method may include interpolating data sets defining pump power to flow for various pump speed values.