Medical Image ROI Mapping for Source-Image Correspondence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing medical image analysis systems face challenges in accurately determining the position of a key image within an original medical image, leading to unclear correspondence and potential loss of image information, which hinders effective training of deep learning models.
Innovation Solution
A medical image analysis apparatus and method that extracts association information from key images to specify the region of interest in the original medical image, using techniques like character and image recognition, registration, and deep learning models to accurately determine the position and add annotations, enabling precise specification of the region of interest.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If key images are used for training deep learning models, then training data availability is improved, but the positional relationship with the original image becomes unclear
Solution Approach 1:
The patent introduces an intermediary processing system that includes image recognition units and registration units. These intermediaries analyze the key images and original medical images to establish correspondence relationships, acting as a mediator that bridges the gap between key images and original images while preserving positional information.
Solution Approach 2:
The patent performs preliminary analysis of key images to extract association information before using them for training. The system预先 (in advance) identifies and extracts positional relationship information, annotation information, and other association data, so that when the training is performed, the positional relationships are already established and preserved.
2Measurement precision
If key images are created from medical images, then region of interest identification is improved, but image information may be defective or lost
Solution Approach 1:
The patent implements a feedback mechanism where the system analyzes the key image, extracts association information, and then uses this information to identify and preserve relevant features in the original medical image. The feedback loop ensures that any information loss during key image creation can be compensated by retrieving and restoring it through the association information extraction and feedback processes.
Solution Approach 2:
The patent segments the medical image processing into distinct components: key image creation, association information extraction, and region of interest specification. By segmenting the process, the system can preserve different types of information at different stages - structural information in key images and positional/annotation information extracted separately, preventing overall information loss.
3Measurement precision
If annotations are added to key images, then region of interest marking is improved, but positional relationship with original image may be lost
Solution Approach 1:
The patent merges the annotation information from key images with the positional relationship information extracted through image recognition and registration processes. By combining these elements, the system creates a comprehensive data structure that includes both the annotated region of interest and its precise positional relationship with the original medical image, preventing information loss.
Data Source
AI summary
A medical image analysis apparatus, a medical image analysis method, and a program for specifying a region of interest intended by a doctor in a medical image of a source of creation of a key image are provided. A medical image analysis apparatus includes at least one processor, and at least one memory in which an instruction to be executed by the at least one processor is stored, in which the at least one processor is configured to acquire a key image that is created from a medical image and that includes a region of interest, extract association information between the key image and the medical image of a source of creation of the key image by analyzing the key image, and specify the region of interest in the medical image based on the association information.