Bio-signal Feature Extraction via Second-Order Differential Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current bio-signal monitoring technologies face challenges in accurately estimating bio-information, such as blood pressure, vascular age, and arterial stiffness, due to limitations in detecting characteristic points from pulse wave signals, especially when high-frequency components are lost or noise is present, leading to reduced measurement accuracy.
Innovation Solution
An apparatus and method that utilize a sensor to obtain bio-signals and a processor to extract features by detecting inflection points, zero-crossing points, and local minimum points from second-order differential signals, even in cases where high-frequency components are absent, by employing specific criteria for determining these points within predetermined periods.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional pulse wave analysis methods are used to estimate bio-information, then the measurement process is simple, but the measurement precision deteriorates when high-frequency components are lost or noise is present
Solution Approach 1:
The patent segments the pulse wave signal analysis by dividing it into multiple differential processing stages. The processor performs first-order differentiation to obtain a first differential signal, then second-order differentiation to obtain a second differential signal, and continues with higher-order differentiations. Each stage extracts specific characteristic points (inflection points, zero-crossing points, local minimum points) that correspond to different physiological information, thereby improving measurement precision through systematic signal decomposition
Solution Approach 2:
The patent transforms the analysis from the time domain to the differential domain by applying multiple orders of differentiation. This dimensional transformation allows the extraction of characteristic points that are not easily identifiable in the original pulse wave signal, particularly when high-frequency components are attenuated or noise is present. The differential signals reveal hidden features that improve bio-information estimation accuracy
2Reliability
If characteristic points are detected from original pulse wave signals, then the detection process is straightforward, but the reliability deteriorates when noise or signal degradation occurs
Solution Approach 1:
The patent applies preliminary differentiation processing to the pulse wave signal before characteristic point detection. By performing first-order, second-order, and higher-order differentiations in advance, the system prepares transformed signals that enhance the visibility and detectability of characteristic points. This preliminary transformation ensures that even when noise or signal degradation occurs in the original pulse wave, the differentiated signals maintain reliable characteristic point features for accurate detection
Solution Approach 2:
The patent changes the signal parameters by applying multiple orders of differentiation to transform the pulse wave signal into differential signals with enhanced characteristic features. The first differential signal, second differential signal, and higher-order differential signals each have different parameter characteristics that make characteristic points (inflection points, zero-crossing points, local minimum points) more distinct and reliable for detection, especially under noisy conditions
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 enhances the accuracy of bio-information estimation by stabilizing the detection of characteristic points, even in noisy or ideal conditions, thereby improving the reliability of measurements for blood pressure and other health indicators.
Implementation Method 1
a light source configured to emit light onto the object, and a detector configured to detect the light emitted onto the object and reflected or scattered from the object
Data Source
AI summary
An apparatus for estimating bio-information includes: a sensor configured to obtain a bio-signal from an object; and a processor configured to obtain a second-order differential signal of the bio-signal, to detect at least one of an inflection point in a predetermined period of the second-order differential signal, and a zero-crossing point in the predetermined period of the second-order differential signal, to extract a feature based on the detected at least one of the inflection point and the zero-crossing point, and to estimate bio-information based on the extracted feature.