Cuffless Blood Pressure Estimation via PTT and Feature Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing non-invasive blood pressure estimation methods, particularly those using pulse transit time (PTT), face challenges in accurately distinguishing between extreme and normal blood pressure values due to imprecision in computations and vagueness in class definition, leading to uncertainty in measurements.
Innovation Solution
A system comprising an ECG sensor, a PPG sensor, and a processor with modules for preprocessing, pulse transit time estimation, feature extraction, feature selection, classification, and regression analysis to accurately classify and estimate blood pressure by synchronizing ECG and PPG signals, selecting relevant features, and performing regression analysis to maximize accuracy in differentiating between high and low/normal blood pressure classes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If pulse transit time (PTT) is used for blood pressure estimation, then non-invasive measurement is achieved, but accuracy in distinguishing extreme and normal BP values deteriorates due to imprecision in computations and vagueness in class definition
Solution Approach 1:
The patent transforms the continuous PTT parameter into discrete BP class categories (extreme, high, normal, low) through classification algorithms. This parameter transformation resolves the contradiction by converting an imprecise continuous measurement into actionable discrete categories, maintaining non-invasive operation while improving practical measurement utility for clinical decision-making
Solution Approach 2:
The patent introduces classification algorithms and feature selection mechanisms as intermediaries between the raw PTT measurement and the final BP assessment. These intermediaries process the imprecise PTT data through multiple computational stages (feature extraction, selection, classification) to produce more reliable BP category distinctions, thereby improving measurement precision without sacrificing non-invasive operation
2Measurement precision
If multiple features are extracted from PPG signal for classification, then accuracy in classifying BP classes is improved, but computational complexity and processing time increase
Solution Approach 1:
The patent extracts and selects only the most relevant features from the PPG signal that have proven significance for BP classification, discarding redundant or less informative features. This selective extraction approach maintains high classification accuracy by focusing on critical features (such as those related to PTT and pulse wave characteristics) while reducing the overall computational burden compared to using all possible features
Solution Approach 2:
The patent segments the feature extraction and classification process into distinct modular stages: signal preprocessing, feature extraction, feature selection, and classification. This segmentation allows for optimized processing at each stage, enabling the system to handle multiple features efficiently by processing them in organized batches rather than as a monolithic computational task, thereby managing complexity while maintaining accuracy
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
The system provides a non-invasive, cuff-less technique for estimating blood pressure with improved accuracy, effectively classifying and estimating blood pressure by leveraging the importance of pulse transit time as a physiological parameter, enhancing the precision in distinguishing between high and low/normal blood pressure classes.
Implementation Method 1
The ECG sensor captures an electrocardiogram (ECG) signal of the person
Implementation Method 2
The PPG sensor captures a photoplethysmogram (PPG) signal of the person
Data Source
AI summary
A method and system for blood pressure (BP) estimation of a person is provided. The system is estimating pulse transit time (PTT) using the ECG signal and PPG signal of the person. A plurality of features are extracted from the PPG. The plurality of PPG features and the PTT are provided as inputs to an automated feature selection algorithm. This algorithm selects a set of features suitable for BP estimation. The selected features are fed to a classifier to classify the database into low/normal BP range and a high BP range. The correctly classified normal BP data are then used to create a regression model to predict BP from the selected features. The current methodology uses automated feature selection mechanism and also employs a block to reject extreme BP data. Thus the available accuracy in predicting BP is expected to be more than the existing BP estimation methods.