Biosignal Analysis Device for ECG Data Segmentation and Noise Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Long-term continuous recording of ECG signals using patch-type ECG measurement devices results in large datasets, with significant portions being unnecessary due to noise, requiring efficient methods to identify and exclude analysis of non-relevant sections to reduce analysis time and resource consumption.
Innovation Solution
A medical data providing device and method that processes biosignals by dividing them into segments, converting them into the frequency domain, and determining whether sections are necessary for analysis based on frequency components, using techniques like Fourier transforms and machine learning models to identify noise and harmonic components, thereby setting sections as either necessary or unnecessary for analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If long-term continuous recording of ECG signals is performed, then comprehensive medical data is obtained, but data size becomes very large and analysis time increases significantly
Solution Approach 1:
The patent divides the continuous ECG signal into multiple signal segments based on detected peak points. Each segment represents a complete cardiac cycle or a portion thereof, allowing selective analysis of only relevant segments rather than processing the entire continuous signal, thus reducing analysis time while maintaining diagnostic value.
Solution Approach 2:
The patent extracts and removes sections identified as analysis-unnecessary based on frequency domain characteristics. By converting signals to frequency domain and detecting harmonic components, the system identifies and excludes noisy or artifact-containing segments, keeping only the essential data for analysis.
2Reliability
If all recorded ECG signal sections are analyzed, then no important information is missed, but processing resources and time are wasted on noisy sections
Solution Approach 1:
The patent performs preliminary frequency domain conversion and harmonic component detection on each signal segment before full analysis. This preliminary screening identifies segments containing noise or artifacts, allowing the system to skip detailed analysis of these segments while maintaining reliability by thoroughly analyzing only the clean segments.
Solution Approach 2:
The patent changes the representation parameter of the ECG signal from time domain to frequency domain using Fourier transform. This parameter transformation enables efficient detection of harmonic components and noise characteristics, allowing rapid identification of analysis-unnecessary sections without compromising the ability to detect genuine cardiac abnormalities.
3Measurement precision
If frequency domain conversion and harmonic detection are performed on all signal segments, then noise sections are accurately identified, but processing time increases
Solution Approach 1:
The patent applies frequency domain conversion and harmonic detection only to segmented portions of the ECG signal rather than the entire continuous signal at once. By processing smaller segments individually and selectively, the system achieves accurate noise detection where needed while minimizing overall processing time through parallel processing and selective application.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach reduces the time, power, and resources required for analysis by identifying and excluding unnecessary sections, thereby streamlining the analysis process and conserving storage space by only processing relevant data.
Implementation Method 1
the processor converts a divided first signal segment into the frequency domain... The processor may perform a Fourier transform or a fast Fourier transform with a certain sample frequency of the biosignal.
Implementation Method 2
The processor may perform a Fourier transform or a fast Fourier transform with a certain sample frequency of the biosignal.
Data Source
AI summary
A medical data providing device includes a memory, a communication unit, and a processor. The processor is configured to read a biosignal stored in the memory, divide the biosignal at certain time intervals, convert a signal segment into the frequency, and determines whether the signal segment is an analysis necessary section or an analysis unnecessary section based on a frequency component of the signal segment. The processor is further configured to analyze the signal segment of the analysis necessary section to generate data including an analysis result, and transmits data including the analysis result to a certain device.