Machine Learning Mapping Function for Patient Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for comparing patient data at different physiological states are costly, risky, and prone to inaccuracies due to the need for repeated measurements or time-consuming simulations, especially when planning therapy before its administration.
Innovation Solution
A system and method that maps patient data from one physiological state to another using a trained mapping function, which extracts features from patient data at the first state to predict the quantity of interest at the second state without requiring direct measurement or simulation at the second state, employing machine-learning based approaches with training data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If repeated measurements are performed at different physiological states, then data accuracy is improved, but patient risk and cost increase
Solution Approach 1:
The patent creates a virtual copy of the patient's physiological state through computational modeling. Instead of repeatedly measuring the actual patient at different physiological states (which increases risk), the system generates simulated measurements from a digital twin that replicates patient-specific anatomy and physiology. This allows accurate data acquisition without exposing the patient to additional clinical procedures or risks.
Solution Approach 2:
The patent performs preliminary actions by creating and validating the computational model during periods when the patient is not undergoing stressful procedures. The model is trained using data from less invasive measurements, and then used to predict outcomes for more stressful physiological states. This preliminary modeling approach allows accurate prediction of high-risk state measurements without actually subjecting the patient to those risky conditions during data collection.
2Measurement precision
If computational fluid dynamics simulation is performed for each physiological state, then measurement accuracy is improved, but time consumption and resource intensity increase
Solution Approach 1:
The patent performs preliminary actions by pre-computing and storing patient-specific anatomical models and boundary conditions during less resource-intensive periods. The computational model is prepared in advance with all necessary geometric and physiological parameters, allowing rapid simulation when needed. This preliminary setup eliminates the need to perform complete, resource-intensive CFD simulations each time a physiological state assessment is required.
Solution Approach 2:
The patent implements a dynamic computational model that can efficiently adapt to different physiological states. Rather than performing static, full-scale CFD simulations for each state, the system uses a dynamic framework that leverages pre-computed anatomical data and adjusts only the necessary physiological parameters for each simulation scenario. This dynamic approach dramatically reduces computation time while maintaining accuracy across multiple physiological states.
3Reliability
If boundary conditions are adjusted to represent different physiological states, then simulation relevance is improved, but difficulty in obtaining correct boundary conditions increases
Solution Approach 1:
The patent introduces an intermediary computational layer that translates complex boundary condition requirements into manageable parameters. Instead of directly dealing with the complexity of obtaining accurate boundary conditions for each physiological state, the system uses an intermediate model that incorporates patient-specific anatomical data and applies simplified physiological relationships. This intermediary approach maintains simulation relevance while reducing the complexity of boundary condition specification.
Solution Approach 2:
The patent systematically changes key physiological parameters (such as flow rates, pressure gradients, and tissue properties) to represent different physiological states. Rather than attempting to capture all aspects of each physiological state through complex boundary conditions, the method identifies and adjusts the most influential parameters that drive the simulation results. This parameter-based approach simplifies the complexity of boundary condition management while maintaining physiological accuracy.
Data Source
AI summary
Systems and methods for determining a quantity of interest of a patient comprise receiving patient data of the patient at a first physiological state. A value of a quantity of interest of the patient at the first physiological state is determined based on the patient data. The quantity of interest represents a medical characteristic of the patient. Features are extracted from the patient data, wherein the features which are extracted are based on the quantity of interest to be determined for the patient at a second physiological state. The value of the quantity of interest of the patient at the first physiological state is mapped to a value of the quantity of interest of the patient at the second physiological state based on the extracted features.