Physiological Modeling for Noninvasive Cardiac Output Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional physiological modeling systems for clinical use are limited in accurately describing interactions between multiple physiological systems and are often invasive, inaccurate, and prone to clinical complications, as they rely on statistical or probabilistic methods and invasive measurement techniques like Swan-Ganz catheters.
Innovation Solution
A computer-readable medium program that emulates cardiopulmonary function by providing a generic model of the cardiopulmonary system, including pulmonary circulation, systemic circulation, heart chambers, autonomic nervous system, metabolism, gas exchange, and reflex, allowing for the measurement and simulation of cardiopulmonary variables, and a clinical patient modeling system that uses pre-generated physiological models with differential equations to generate decision support data for diagnosing and treating patients, enabling noninvasive estimation of cardiac output.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If invasive measurement techniques (Swan-Ganz catheters, direct Fick method) are used to measure cardiac output, then measurement precision is improved, but device complexity and clinical complications increase
Solution Approach 1:
The patent replaces invasive mechanical measurement systems (Swan-Ganz catheters, direct Fick method requiring fluid inhalation or injection) with a noninvasive computational model that uses mathematical algorithms to estimate cardiac output from readily available clinical parameters such as oxygen consumption, hemoglobin concentration, and cardiac index, thereby eliminating the need for invasive probes while maintaining measurement accuracy
Solution Approach 2:
The patent creates a virtual copy of the physiological system through computational modeling, where the actual physiological parameters are replicated in a mathematical framework that can be simulated and analyzed without physically invasive procedures, allowing repeated measurements and scenario testing without affecting the patient
2Ease of operation
If statistical or probabilistic models based on available patient data are used, then ease of operation is improved, but measurement precision deteriorates
Solution Approach 1:
The patent transforms the modeling approach by changing from statistical/probabilistic parameters to deterministic physiological parameters based on fundamental biological principles and conservation laws, allowing the model to accurately predict physiological responses by manipulating input parameters such as oxygen consumption rates, hemoglobin levels, and cardiac indices through controlled mathematical relationships
3Manufacturing precision
If mathematical models describing single physiological systems are developed, then manufacturing precision is improved, but adaptability deteriorates
Solution Approach 1:
The patent merges previously separate single-system physiological models into an integrated multi-system framework that simultaneously models cardiovascular, respiratory, and metabolic systems, allowing the system to capture interactions between different physiological domains while maintaining the validation accuracy of individual subsystem models through modular architecture
4Measurement precision
If invasive probes are placed at different locations in and on the body, then measurement precision is improved, but object-affected harmful factors increase
Solution Approach 1:
The patent substitutes physical invasive probes with a computational measurement system that derives physiological parameters through mathematical calculations from noninvasive clinical data, completely eliminating probe-body interactions and associated complications such as infections, discomfort, and positioning errors while maintaining measurement precision
Data Source
AI summary
When generating a model of physiological systems in a patient, differential equations representing parameters and variables in the systems are linked together to form one or more sub-models (e.g., one for each physiological system), which in turn are linked together to form the patient model. Simulations of hypothetical clinical situations are then run on the model to solve for the variables, and the solutions are output as decision support data for review by a clinician to facilitate a determination of a treatment of diagnosis for the patient. Additionally, model predictions can be compared to actual measurements, when available, and the model can be refined or optimized as a function of the comparison.