Graph Neural Network for Tissue Slide Feature Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional machine learning techniques face difficulties in extracting high-level pathological information from whole slide images (WSI) for diagnosing anticancer drug responses, requiring medical expertise and being resource-intensive, while deep learning methods are inefficient due to high capacity and resource requirements.
Innovation Solution
A method and system using a graph neural network (GNN) to extract and analyze graphic data from tissue slide images, including training and inference nodes and edges, to derive efficient and high-level information without requiring extensive medical expertise or resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If traditional machine learning techniques are used to extract features from WSI information, then simple mathematical models can be used for training, but difficulty arises in extracting features needed for high-level pathological information diagnosis
Solution Approach 1:
The patent introduces an attention mechanism as an intermediary component between the input WSI data and the deep neural network. This attention mechanism selectively weights different regions and features of the tissue slide image, enabling the model to focus on diagnostically relevant areas while maintaining computational efficiency. The attention mechanism acts as a mediator that bridges the gap between simple feature extraction and complex high-level pathological analysis.
2Measurement precision
If deep learning techniques input all pixels of WSI into a neural network, then comprehensive feature learning is achieved, but speed and efficiency are greatly degraded and much more resources are needed
Solution Approach 1:
The patent extracts and processes only the most relevant features from the WSI data using the attention mechanism, rather than processing all pixels. By taking out and focusing on salient regions and features, the model achieves high-level pathological information extraction with significantly reduced computational burden, improving processing speed while maintaining diagnostic accuracy.
Solution Approach 2:
The patent segments the WSI data into multiple patches or regions, and the attention mechanism selectively processes only those segments that contain diagnostically relevant information. This segmentation approach divides the large-scale WSI into manageable units, allowing efficient processing while capturing essential pathological features.
3Measurement precision
If deep learning techniques input all pixels of WSI into a neural network, then comprehensive feature learning is achieved, but much more computational resources are needed
Solution Approach 1:
The patent extracts and processes only the most relevant features from the WSI data using the attention mechanism, rather than processing all pixels. By taking out and focusing on salient regions and features, the model achieves high-level pathological information extraction with significantly reduced computational burden, improving processing speed while maintaining diagnostic accuracy.
Data Source
AI summary
An image analysis method and an image analysis system are disclosed. The method may include extracting training graphic data including at least one first node corresponding to a plurality of histological features of a training tissue slide image, and at least one first edge defined by a relationship between the histological features The method may also include determining a parameter of a readout function by training a graph neural network (GNN) using the training graphic data and training output data corresponding to the training graphic data. The method may also include extracting inference graphic data including at least one second node corresponding to a plurality of histological features of an inference tissue slide image, and at least one second edge decided by a relationship between the histological features, The method may further include deriving inference output data by the readout function after inputting the inference graphic data to the GNN.