Machine Learning Ejection Fraction Analysis from Implantable Cardiac Current Curves
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for monitoring cardiac pump function in heart failure patients require invasive procedures, are time-consuming, and involve patient risk, making regular monitoring challenging, especially for remote detection of new onset heart failure.
Innovation Solution
A computer-implemented method using a trained machine learning algorithm that analyzes one-channel cardiac current curve data from implantable medical devices to determine ejection fraction and its variation, enabling remote and automated monitoring, with notifications sent to healthcare providers for abnormal readings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If imaging methods (cardiac echo, MRI, CT, catheter examination) are used to measure ejection fraction, then measurement precision is improved, but device complexity and patient risk increase
Solution Approach 1:
The patent replaces complex mechanical imaging systems (echo, MRI, CT, catheter) with a simplified electrical measurement system using implantable medical devices to record cardiac current curves. The machine learning algorithm processes these electrical signals to derive ejection fraction, substituting mechanical/imaging approaches with electrical and computational methods.
Solution Approach 2:
The patent introduces machine learning algorithms as an intermediary between simple cardiac current curve recordings and ejection fraction determination. The algorithm acts as a mediator that translates basic electrical signals into clinically meaningful ejection fraction values without requiring complex imaging hardware.
2Measurement precision
If imaging procedures are performed regularly to monitor cardiac pump function, then measurement precision is improved, but loss of time and patient burden increase
Solution Approach 1:
The implantable medical device automatically records cardiac current curves and transmits them without requiring patient presence or active cooperation. The system serves itself by autonomously performing measurements and data transmission, eliminating the need for patient travel to clinical facilities.
Solution Approach 2:
The patent enables continuous monitoring by having the implantable device continuously record cardiac current curves and automatically transmit data. This provides ongoing ejection fraction assessment without interrupting patient daily life or requiring periodic visits to healthcare facilities.
3Ease of operation
If remote transmission of 12-lead ECG is implemented, then ease of operation is improved, but reliability decreases due to patient compliance requirements
Solution Approach 1:
The implantable medical device automatically performs ECG recording and data transmission without requiring patient action. The device self-initiates the measurement process and autonomously transmits data, eliminating compliance issues associated with patient-dependent remote monitoring.
Solution Approach 2:
The implantable medical device serves as an intermediary between the patient's heart and the external monitoring system. It captures cardiac electrical activity internally and handles transmission automatically, removing the need for patient cooperation in the data collection process.
4Productivity
If machine learning algorithm is trained on cardiac current curve data, then productivity is improved through automated analysis, but device complexity increases
Solution Approach 1:
The patent replaces manual analysis of cardiac data with automated machine learning algorithms. The computational system processes cardiac current curves and derives ejection fraction automatically, substituting human expertise and time-consuming analysis with algorithmic processing.
Data Source
AI summary
A computer implemented method for determining an ejection fraction, comprising the steps of receiving a first data set comprising pre-acquired cardiac current curve data, in particular one-channel cardiac current curve data, captured by an implantable medical device, applying a machine learning algorithm to the pre-acquired cardiac current curve data, and outputting a second data set representing the ejection fraction and/or the variation of the ejection fraction or a classification of the ejection fraction and/or a classification of the variation of the ejection fraction by the machine learning algorithm. Furthermore, a corresponding system and a method for providing a trained machine learning algorithm is provided.