Neural Network ECG Classifier for Arrhythmia Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current cardiac monitoring systems face challenges in accurately detecting arrhythmias due to the need for large bandwidth data transmission and human review of biometric data, which can lead to incorrect identification of cardiac events, potentially resulting in inappropriate treatment or missed diagnoses.
Innovation Solution
An arrhythmia monitoring system that includes an external heart monitoring device with ECG electrodes, ECG processing circuitry, and a neural network-based rhythm change classifier to detect predetermined rhythm changes in ECG signals, transmitting relevant data to a remote computer system for further analysis and classification, utilizing additional sensors and baseline data for enhanced accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If large bandwidth data transmission is used to transmit all biometric data for cardiac monitoring, then the completeness of cardiac event detection is improved, but the system complexity and data processing burden increase significantly
Solution Approach 1:
The patent extracts and transmits only the most relevant cardiac event data (arrhythmias, Bradycardia, tachycardia) to the remote server, rather than transmitting all biometric data. This selective extraction reduces bandwidth requirements and processing complexity while maintaining detection reliability for critical cardiac events.
Solution Approach 2:
The local device acts as an intermediary that performs preliminary analysis and filtering of biometric data before transmission. It identifies and extracts only the most relevant cardiac events, serving as a mediator between continuous monitoring and remote analysis, thereby reducing data transmission burden.
2Adaptability or versatility
If human reviewers manually analyze all biometric data to identify cardiac events, then the flexibility in detecting various cardiac conditions is improved, but the time consumption and potential for human error increase
Solution Approach 1:
The system performs preliminary automated analysis and identification of cardiac events at the local device before remote review. This preliminary action pre-processes the data, identifies potential cardiac events, and prepares them for efficient remote analysis, reducing the time required for comprehensive cardiac event detection.
Solution Approach 2:
The patent replaces manual human review with automated machine learning algorithms and neural networks for initial cardiac event detection and classification. This substitution eliminates human review time and reduces human error while maintaining adaptability in detecting various cardiac conditions through programmable analysis.
3Measurement precision
If all biometric data is transmitted to remote servers for analysis, then the accuracy of cardiac event classification is improved, but the bandwidth requirements and data processing load increase
Solution Approach 1:
The system extracts and transmits only the most relevant cardiac event features and parameters needed for accurate classification, rather than transmitting complete raw biometric datasets. This selective extraction maintains classification accuracy by preserving critical diagnostic information while significantly reducing data transmission volume.
Solution Approach 2:
The patent segments the biometric data into essential cardiac event features and non-essential data. Only the segmented essential features containing diagnostic information are transmitted to remote servers, reducing overall data volume while preserving the information necessary for accurate arrhythmia classification.
Data Source
AI summary
Some embodiments of the current disclosure are directed toward cardiac diagnosis and/or arrhythmia monitoring, and more particularly, systems, devices and methods for arrhythmia monitoring with a trained classifier including at least one neural network. In some embodiments, an external heart monitoring device may include a plurality ECG electrodes to sense surface ECG activity, ECG processing circuitry to process the surface ECG activity to provide at least one ECG signal, a non-transitory computer-readable medium comprising a rhythm change classifier comprising at least one neural network, and at least one processor to receive the ECG signal(s), detect with the rhythm change classifier time data corresponding to a predetermined rhythm change in the ECG signal(s), determine based on the detected time data at least one ECG signal portion corresponding to the predetermined rhythm change, and transmit the at least one determined ECG signal portion to a remote computer system.