Dual-Path Neural Network for Automated Ventricular Ectopic Beat Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Interpreting and summarizing mobile cardiac telemetry (MCT) study results is a time-consuming process due to the large amounts of electrocardiogram (ECG) data generated, which requires high-performance algorithms for efficient analysis and classification of heartbeats, such as Ventricular Ectopic Beats (VEBs).
Innovation Solution
A computer-implemented method using a machine learning architecture that combines a convolutional neural network for analyzing waveform data and a fully-connected neural network for analyzing time interval data to classify heartbeats, improving accuracy and efficiency in classifying cardiac events without requiring time-consuming patient-specific labeling of training data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional methods are used to interpret and summarize MCT study results, then comprehensive analysis can be performed, but the process becomes time-consuming due to large amounts of ECG data
Solution Approach 1:
The patent divides the ECG data analysis into multiple independent processing paths: a first path processes waveform data using a convolutional neural network, while a second path processes time interval data using a fully-connected neural network. This segmentation allows parallel processing of different data types, reducing overall analysis time while maintaining comprehensive evaluation for accurate heartbeat classification.
Solution Approach 2:
The patent replaces traditional manual or rule-based ECG analysis methods with machine learning-based automated classification systems. The convolutional neural network and fully-connected neural network automatically learn patterns from ECG data, substituting time-consuming manual interpretation with efficient computational algorithms that maintain high diagnostic accuracy.
2Productivity
If high-performance algorithms are used for efficient analysis and classification, then processing speed improves, but system complexity increases
Solution Approach 1:
The complex analysis task is divided into two specialized neural networks with distinct architectures: a convolutional neural network for waveform pattern recognition and a fully-connected neural network for time interval analysis. Each network is optimized for its specific data type, making individual components more manageable while achieving high overall processing efficiency through parallel operation.
Solution Approach 2:
The patent employs a universal machine learning framework that can handle multiple types of ECG data (waveform and time interval) through a unified classification system. The dual-path architecture serves multiple functions: it processes different data types simultaneously, provides robust classification across various heartbeat conditions, and can be deployed in diverse clinical settings without requiring separate specialized systems.
3Measurement precision
If patient-specific labeling of training data is required, then classification accuracy may improve, but the time and resources needed for data preparation increase significantly
Solution Approach 1:
The patent performs preliminary data processing and feature extraction automatically during the model training phase, preparing the data in advance for classification. The convolutional neural network automatically extracts relevant features from waveform data, and the fully-connected network processes time interval features, eliminating the need for manual patient-specific labeling while maintaining high classification sensitivity through pre-computed meaningful representations.
Solution Approach 2:
The machine learning models perform self-training by automatically learning from the ECG data structures themselves. The convolutional and fully-connected networks autonomously identify patterns and relationships in the waveform and time interval data without requiring external manual annotation, enabling the system to achieve high accuracy through self-directed learning from the inherent structure of the medical data.
Data Source
AI summary
Techniques for classifying heartbeats using patient electrocardiogram (ECG) data are described. ECG data is received, including waveform data and time interval data relating to a plurality of heartbeats for the patient. A convolutional neural network in a first path of a machine learning architecture generates a first plurality of output values by analyzing the waveform data. A fully-connected neural network in a second path of the machine learning architecture generates a second plurality of output values by analyzing the time interval data. The plurality of heartbeats in the ECG data are classified by concatenating the first plurality of output values and the second plurality of output values using the machine learning architecture.