Neural Network Training with Label Merging for Partial Annotations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manual annotation of medical images for training neural networks is time-consuming and costly, making it prohibitive for large-scale studies, and existing models require fully-annotated images for effective training, which limits their performance with partially-annotated data.
Innovation Solution
A method that uses a label merging function to map labels from fully-annotated images to partially-annotated images, allowing the integration of both types of data for training neural networks, thereby increasing the size of the training set and improving segmentation performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual annotation is used to create fully-annotated training images, then segmentation precision is improved, but loss of time and cost increase significantly
Solution Approach 1:
The patent applies partial annotation by creating two types of training images: fully-annotated images with complete segmentation labels for all organic structures, and partially-annotated images with labels for only a subset of structures. This partial action approach reduces the time-consuming manual annotation workload while maintaining effective model training through the use of label merging functions that integrate both annotation types.
2Measurement precision
If fully-annotated images are required for model training, then segmentation performance is improved, but productivity decreases due to limited available training data
Solution Approach 1:
The patent merges fully-annotated images and partially-annotated images into a unified training dataset using label merging functions. These functions map labels from fully-annotated images to corresponding regions in partially-annotated images, creating combined training sets that expand data availability while maintaining segmentation performance through integrated label information.
3Measurement precision
If manual annotation is performed for large-scale studies, then measurement precision is improved, but loss of time becomes prohibitive
Solution Approach 1:
The patent implements partial annotation strategies where only critical or representative organic structures are annotated in partially-annotated images, rather than requiring complete annotation of all structures. This selective annotation approach maintains sufficient segmentation accuracy for large-scale studies while dramatically reducing the time investment required compared to full manual annotation.
4Measurement precision
If more fully-annotated training data is collected, then segmentation precision is improved, but loss of time and resources increase
Solution Approach 1:
The patent combines fully-annotated and partially-annotated images through label merging functions to create expanded training datasets. This merging approach allows the model to learn from both complete and partial annotation examples, improving segmentation precision without requiring proportional increases in time-consuming manual annotation work for fully-annotated images.
Data Source
AI summary
Methods and systems for training a model labeling two or more organic structures within an image. One method includes receiving a set of training images. The set of training images including a first plurality of images and a second plurality of images. Each of the first plurality of images including a label for each of the two or more organic structures and each of the second plurality of images including a label for only a subset of the two or more organic structures. The method further includes training the model using the first plurality of images, the second plurality of images, and a label merging function mapping a label from the first plurality of images to a label included in the second plurality of images.