Embryo Development Timing Assessment via Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for assessing embryo development in IVF processes are time-consuming due to manual annotation of time-lapse embryo imaging, which requires skilled embryologists to identify and measure morphokinetic parameters from a series of images, hindering efficient selection of high-quality embryos for transfer.
Innovation Solution
A computer-implemented method using machine learned classifiers to determine estimated timings of developmental events in embryos from time series images, employing feature information abstraction through approaches like scale invariant feature transform, machine learned feature clustering, and neural networks, to automate the annotation process and provide confidence estimates for the timings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual annotation by skilled embryologists is used to assess embryo quality, then measurement precision and reliability are improved, but productivity and time consumption worsen
Solution Approach 1:
The patent replaces the manual mechanical process of embryologists visually inspecting and annotating time-lapse images with an automated computer vision system. The system uses deep learning neural networks to automatically detect, track, and measure morphokinetic parameters from embryo images, eliminating the need for human intervention in the measurement process while maintaining assessment accuracy.
Solution Approach 2:
The patent creates a digital copy of the manual annotation process through trained neural network models. These models learn from annotated training data and reproduce the assessment methodology, allowing automated reproduction of embryologist evaluations without requiring the actual human experts to perform each measurement manually.
2Measurement precision
If manual annotation of morphokinetic parameters is performed, then measurement precision is improved, but loss of time increases
Solution Approach 1:
The patent performs preliminary training of the neural network system using annotated training data before actual embryo assessment. During this preliminary phase, the system learns the correct identification and measurement of morphokinetic parameters. Once trained, the system can rapidly assess new embryos without requiring time-consuming manual annotation for each case.
Solution Approach 2:
The patent substitutes the time-consuming manual process of visually scanning and annotating each image frame with an automated neural network system that processes images instantly, dramatically reducing the time required for embryo quality assessment while maintaining measurement precision.
3Productivity
If automated annotation techniques are implemented, then productivity is improved, but device complexity increases
Solution Approach 1:
The patent segments the complex embryo assessment task into distinct functional modules: image acquisition, feature extraction, neural network classification, parameter measurement, and quality assessment. This modular approach allows each component to be developed and optimized independently, making the overall system more manageable despite its complexity.
Solution Approach 2:
The patent introduces an intermediary layer of feature extraction and preprocessing that bridges the raw images and the final assessment decisions. This intermediary processing layer simplifies the data representation and prepares it for the neural network, reducing the effective complexity of the decision-making process while maintaining high productivity.
Data Source
AI summary
A method of processing a series of images of an embryo to determine estimated timings of developmental events for the embryo, wherein the method comprises: determining feature information for each image, the feature information for each image representing the content of the image; establishing machine learned classifiers for associating each image with a respective likelihood of the image being associated with one or more developmental events based on the feature information for the image; applying the machine learned classifiers to the feature information for each image to determine a respective likelihood of the image being associated with one or more developmental events, and determining estimated timings for the plurality of developmental events for the embryo from the respective likelihoods of the respective images being associated with respective ones of the plurality of developmental events. The method may further comprise determining an indication of a confidence estimate for the timings.