Cardiovascular State Estimation via Probabilistic Signal Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Mechanical and biomedical sensors face challenges in filtering out contaminating signals, particularly those caused by sensor movement and nonlinear, non-stationary noise, which complicates the extraction of reliable physiological information, and they typically measure a limited range of medical parameters.
Innovation Solution
The use of a probabilistic model that combines data from multiple sensors through a dynamic state-space model and a probabilistic digital signal processor to filter and estimate additional biomedical parameters, such as stroke volume and cardiac output, by iteratively updating probability distribution functions and integrating sensor data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If simple filters are used to remove artifacts, then device complexity is reduced, but measurement precision deteriorates because artifacts resembling real processes cannot be removed reliably
Solution Approach 1:
The patent transforms the filtering approach from simple frequency-based filtering to a complex parameter-based filtering system that analyzes multiple signal characteristics simultaneously. The system evaluates amplitude, frequency, duration, and morphological parameters of artifacts to distinguish them from genuine physiological signals, thereby achieving high-precision artifact removal without oversimplification
Solution Approach 2:
The patent introduces an intermediary artifact database that serves as a reference library containing characteristics of known artifacts. This database acts as a mediator between the raw signal and the filtering decision, enabling the system to compare incoming signals against stored artifact patterns and remove only those matching known artifact characteristics while preserving genuine physiological signals
2Measurement precision
If multiple sensors are used to measure additional parameters, then measurement precision improves, but device complexity increases due to data fusion requirements
Solution Approach 1:
The patent merges data from multiple sensors (ECG, PPG, accelerometer, gyroscope) into a unified physiological state determination system. By combining these diverse data sources and analyzing them together through integrated algorithms, the system achieves more accurate physiological parameter measurement than any single sensor could provide alone
Solution Approach 2:
The patent creates a universal data processing framework that handles multiple sensor types and multiple physiological parameters through a single integrated system. This multi-functional approach allows the same processing architecture to extract information from various sensor modalities and compute multiple physiological parameters (heart rate, stroke volume, blood pressure, etc.) simultaneously
3Device complexity
If conventional filtering is used on sensor data, then device complexity is minimized, but loss of information increases because valuable physiological information may be filtered out along with noise
Solution Approach 1:
The patent applies local quality by treating different portions of the signal with different processing approaches. Genuine physiological signals are preserved with minimal filtering, while artifact-contaminated portions are selectively removed. The system adapts its filtering intensity and type based on the local characteristics of each signal segment, ensuring that valuable physiological information is not lost
Data Source
AI summary
The invention comprises a method and apparatus for estimating state of a cardiovascular system, comprising a cardiac stroke volume analyzer, comprising: (1) a blood pressure sensor generating a time-varying pressure state waveform output from a limb of the person; (2) a system processor connected to the blood pressure sensor; and (3) a dynamic state-space model; the system processor receiving cardiovascular input data, from the blood pressure sensor, related to a transient pressure state of the cardiovascular system; at least one probabilistic model, of the dynamic state-space model, operating on the time-varying pressure state waveform output to generate a probability distribution function to a non-pressure state of the cardiovascular system; iteratively updating the probability distribution function using output from the blood pressure sensor; and processing the probability distribution function to generate a non-pressure state output related to stroke volume of a heart of the person and arterial compliance of the person.