Fetal Oximetry Model Training Using Simulated Light Transmission
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for monitoring fetal health, such as fetal heart rate monitoring, are inefficient in determining fetal distress and often provide false positive results, leading to unnecessary interventions like Cesarean deliveries.
Innovation Solution
A two-step process using simulated light transmission data and machine learning to train a fetal oximetry model, first through a simulated model and then an in vivo model, to predict fetal oximetry values without the need for invasive fetal data collection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If current fetal heart rate monitoring methods are used, then fetal health monitoring is provided, but the monitoring is inefficient and provides false positive results indicating fetal distress
Solution Approach 1:
The patent uses simulated light transmission data that copies the characteristics of actual in vivo fetal light transmission data. This simulated data is generated through computer modeling of light interaction with fetal tissue, allowing the machine learning model to be trained without requiring actual invasive fetal measurements. The simulated data preserves the essential optical properties and variability of real fetal tissue while avoiding the risks and limitations of in vivo data collection.
Solution Approach 2:
The patent performs preliminary training of the machine learning model using simulated light transmission data before deploying it for actual fetal oximetry measurements. This preliminary action allows the model to learn the complex relationships between light transmission patterns and fetal oxygen saturation levels in a controlled environment, improving its accuracy and reliability before real-world application.
2Measurement precision
If invasive methods are used to collect fetal data, then accurate fetal oximetry values can be obtained, but the process is costly, difficult, and time-consuming
Solution Approach 1:
The patent creates copies of actual fetal light transmission scenarios through simulated data that replicates the optical properties, tissue layers, and physiological variations of real fetal measurements. This allows comprehensive training data to be generated instantly without the time-consuming process of collecting actual in vivo data from multiple pregnancies.
Solution Approach 2:
The patent performs the data collection and model training process in advance using simulated data, creating a pre-trained model that can be immediately deployed for clinical use. This eliminates the time delay between starting a new fetal monitoring project and having a functional model, as the training is completed beforehand using computationally generated data.
3Adaptability or versatility
If invasive fetal data collection is performed, then training data for oximetry models can be obtained, but the process is costly and difficult
Solution Approach 1:
The patent generates diverse training scenarios by copying and varying key parameters in the simulated fetal tissue model, including different tissue compositions, optical properties, fetal positions, and physiological states. This approach provides comprehensive training data coverage without the logistical challenges of recruiting and monitoring diverse actual fetal subjects.
Solution Approach 2:
The patent performs comprehensive model training in advance using simulated data that covers a wide range of fetal conditions and scenarios. This preliminary training creates a versatile model that can adapt to various fetal characteristics without requiring additional invasive data collection for each new application scenario.
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
This approach significantly reduces the time and cost required to develop an accurate fetal oximetry model, enabling precise fetal distress detection during gestation and labor, while minimizing the need for costly and difficult-to-obtain in vivo data.
Implementation Method 1
simulated light transmission data may correspond to a simulated electronic signal that is similar to an electronic signal that may be provided by a photodetector upon detection of an optical signal that has traveled through tissue
Data Source
AI summary
A plurality of sets of simulated optical inputs that is simulated to travel through an animal model of tissue, thereby generating simulated light transmission data, and corresponding oximetry vales may be used to train a simulated fetal oximetry model to predict, or calculate, oximetry values for subsequently received sets of simulated light transmission data. The simulated fetal oximetry model may be adapted to train an in vivo fetal oximetry model that may be configured to predict, or calculate, fetal oximetry values for subsequently received sets of light transmission data received from an in vivo study of a pregnant mammal and her fetus.