ECG Detection Using Spectrogram Conversion and Deep Neural Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing cardiac disease detection systems are limited in their ability to identify a wide range of cardiac irregularities and conditions, particularly failing to detect complex conditions such as Hyperkalemia, Hypercalcemia, Hypokalemia, and Hypocalcemia, and lack a cost-effective, accessible solution for comprehensive monitoring.
Innovation Solution
A deep neural network system that filters one-dimensional ECG data, extracts features from the PQRST complex, classifies cardiac diseases using interval and peak detection techniques, and converts classified data into two-dimensional spectrogram images for enhanced detection and classification of cardiac diseases, including complex conditions, through a filtration, feature extraction, classification, and conversion module.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If existing state of the art methods are used for cardiac disease detection, then the system is simple and accessible, but the detection capability is limited to very few irregularities
Solution Approach 1:
The patent converts one-dimensional ECG data into two-dimensional spectrogram images, adding a temporal-frequency dimension to the data representation. This dimensionality change enables the use of advanced deep learning models (CNNs) that can detect complex cardiac conditions while maintaining system accessibility through remote server processing.
2Measurement precision
If advanced deep neural network techniques are implemented, then detection accuracy for complex conditions improves, but computational requirements and processing time increase
Solution Approach 1:
The system performs preliminary processing of ECG data by converting it to spectrogram images and extracting relevant features before classification. This preliminary action prepares the data in an optimized format that reduces computational burden during the actual detection phase, enabling faster processing of complex cardiac conditions.
Solution Approach 2:
The detection process is divided into distinct segments: ECG data acquisition, spectrogram conversion, feature extraction, and classification. This segmentation allows each module to be optimized independently, with the spectrogram conversion and feature extraction steps preparing data that accelerates the final classification process.
3Reliability
If comprehensive detection of multiple cardiac diseases is implemented, then the system becomes more robust, but the cost and complexity of deployment increase
Solution Approach 1:
The patent introduces a remote server as an intermediary that hosts the computationally intensive deep learning models and spectrogram processing infrastructure. This allows local devices to maintain simplicity and low cost while still accessing advanced detection capabilities through the remote server, enabling comprehensive cardiac disease detection without high deployment costs at each location.
Data Source
AI summary
The invention discloses a system for detection and classification of cardiac diseases using deep neural network techniques, wherein the system (100) comprises filtration module (101) for filtering one-dimensional Electrocardiogram (ECG) data. The filtered ECG data is provided to feature extraction module (102) for extracting a set of pre-defined features from the PQRST complex, wherein the extracted features are classified by the classification module (103) for the purpose of cardiac disease detection using interval and peak detection techniques. Further, the classified one-dimensional ECG data is converted into a two-dimensional spectrogram image by the conversion module (104). The two-dimensional spectrogram image is passed through custom deep neural networks and the diagnostic results may be accessed through a remote server module (105) and viewed by the individuals on a user interface device.