Ophthalmic Segmentation Using Multi-Modal Tomographic Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current OCT image segmentation techniques struggle to accurately identify layer regions in eyes, especially in diseased eyes, due to low image quality, requiring manual modification by doctors which is time-consuming and labor-intensive.
Innovation Solution
An ophthalmic information processing apparatus that acquires OCT images and additional tomographic information images using different methods, performs segmentation processing to identify layer region boundaries, and displays these boundaries distinctly on a display means, allowing for accurate identification and modification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If segmentation processing is performed on OCT images to identify layer region boundaries, then measurement precision of layer regions is improved, but the accuracy of segmentation deteriorates when image quality is low
Solution Approach 1:
The patent introduces tomographic information images generated by a method different from OCT as an intermediary to facilitate accurate segmentation. These additional images provide complementary information that helps the segmentation processor accurately identify layer region boundaries even when the original OCT image quality is low, thus resolving the contradiction between measurement precision and reliability
Solution Approach 2:
The patent combines multiple types of tomographic information images (OCT images and images generated by other methods) into a unified segmentation processing system. By merging these different image sources, the system achieves more reliable and accurate segmentation results than would be possible with OCT images alone, particularly in low-quality cases
2Measurement precision
If manual modification of segmentation results is performed to improve accuracy, then measurement precision is improved, but productivity deteriorates due to time consumption
Solution Approach 1:
The system performs automated segmentation processing using the segmentation processor that integrates multiple tomographic information images, eliminating the need for manual modification by doctors. The self-service capability maintains high measurement precision while dramatically improving productivity by removing the time-consuming manual intervention step
3Measurement precision
If multiple tomographic information images are acquired and processed, then measurement precision of layer regions is improved, but device complexity increases
Solution Approach 1:
The patent implements a universal processing framework that can handle multiple types of tomographic information images (OCT and other methods) through a single segmentation processor. This multi-functional approach improves measurement precision by utilizing diverse image data while managing device complexity through a unified processing architecture rather than separate specialized systems
Data Source
AI summary
An ophthalmic information processing apparatus includes an acquisition unit, a segmentation processor, and a display controller. The acquisition unit is configured to acquire an OCT image and one or more tomographic information images. The OCT image is a tomographic image of an eye to be examined. The one or more tomographic information images are generated using a method different from a method of generating the OCT image and represent tomographic information of the eye to be examined. The segmentation processor is configured to perform segmentation processing on the OCT image to identify a boundary of a layer region in a depth direction. The display controller is configured to display the OCT image, in which the boundary identified by the segmentation processor is distinguishably depicted, and the one or more tomographic information images on a display means.