Probabilistic Digital Signal Processor for Biomedical Parameter Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Biomedical sensors face challenges in extracting reliable physiological information due to overlapping noise frequencies and severe artifacts, such as signal dropouts from sensor movement, which conventional filtering techniques struggle to address, especially in mobile or ambulatory patients.
Innovation Solution
A probabilistic model-based system using a dynamic state-space model and probabilistic digital signal processor to estimate physiological parameters by integrating sensor data with physical models, effectively filtering noise and extracting additional information from biomedical signals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional filtering techniques are used to process biomedical signals, then device complexity is reduced, but measurement precision and reliability deteriorate due to overlapping noise frequencies and severe artifacts
Solution Approach 1:
The patent transforms the signal processing approach by changing from deterministic filtering parameters to probabilistic model parameters. The system uses a probabilistic digital signal processor that maintains multiple hypotheses about signal characteristics simultaneously, allowing it to adapt to non-stationary noise and artifacts while preserving measurement precision without excessive complexity increase
Solution Approach 2:
The patent replaces conventional mechanical/filter-based noise reduction with a probabilistic computational model. Instead of using fixed filter mechanisms that struggle with overlapping frequencies, the system uses probabilistic reasoning to distinguish signal from noise, achieving better precision while managing complexity through software-based probabilistic processing
2Reliability
If simple filters are used to remove artifacts, then device complexity is minimized, but reliability worsens because artifacts resembling real processes cannot be removed reliably
Solution Approach 1:
The patent segments the signal processing into multiple probabilistic hypotheses, each representing a different interpretation of the signal components. The system evaluates multiple segments of evidence simultaneously and combines them to make reliable artifact removal decisions, allowing distinction between real physiological processes and artifacts even when they overlap in frequency
Solution Approach 2:
The probabilistic digital signal processor implements feedback by continuously updating probability distributions based on new signal data. The system uses feedback loops where posterior probabilities from one time step become prior probabilities for the next, enabling adaptive and reliable artifact removal that responds to changing signal conditions
3Measurement precision
If conventional filtering is applied to mobile patient data, then processing speed is maintained, but measurement precision deteriorates due to non-stationary noise characteristics
Solution Approach 1:
The patent implements dynamic signal processing where the probabilistic model parameters are continuously updated as new data arrives. The system adapts to non-stationary noise characteristics in real-time by adjusting its probability distributions dynamically, maintaining both measurement precision and processing speed for mobile patient monitoring
Solution Approach 2:
The system performs preliminary probabilistic modeling and hypothesis generation before final parameter estimation. By pre-computing probability distributions and preparing multiple hypotheses in advance, the system can quickly resolve ambiguities in real-time without sacrificing measurement precision during actual signal processing
Data Source
AI summary
A probabilistic digital signal processor for medical function is described. Initial probability distribution functions are input to a dynamic state-space model, which operates on state and/or model probability distribution functions to generate a prior probability distribution function, which is input to a probabilistic updater. The probabilistic updater integrates sensor data with the prior to generate a posterior probability distribution function passed to a probabilistic sampler, which estimates one or more parameters using the posterior, which is output or re-sampled in an iterative algorithm. For example, the probabilistic processor operates using a physical model on data from a medical meter, where the medical meter uses a first physical parameter, such as blood oxygen saturation levels from a pulse oximeter, to generate a second physical parameter not output by the medical meter, such as a heart stroke volume, a cardiac output flow rate, and/or a blood pressure.