Ultrasound Tomographic Image Correction via Volume Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current medical imaging techniques, such as ultrasonic image diagnosis, struggle to accurately grasp unclear image sensing regions in tomographic images, leading to distorted three-dimensional shape reconstruction due to probe pressure and unclear image discrimination.
Innovation Solution
An image processing apparatus that acquires tomographic images, position and orientation information, and three-dimensional volume data to specify unclear image sensing regions, replacing them with corresponding data from three-dimensional volume images, thereby correcting deformation and enhancing image clarity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Shape
If three-dimensional shape data is reconstructed from tomographic images obtained by ultrasonic probe, then three-dimensional visualization is achieved, but deformation occurs due to probe pressure causing distorted reconstruction
Solution Approach 1:
The probe contact period is segmented into three phases: before contact, during contact, and after contact. Tomographic images obtained during the contact period are excluded from reconstruction, while only images from before and after contact periods are used. This temporal segmentation eliminates deformation artifacts caused by probe pressure while maintaining three-dimensional shape reconstruction capability.
2Ease of operation
If tomographic images are obtained by moving ultrasonic probe freely, then image sensing flexibility is improved, but position and orientation information becomes unknown leading to reconstruction difficulties
Solution Approach 1:
The system uses the ultrasonic probe itself to detect contact state with the object surface by analyzing reflected ultrasonic waves. When the probe is in contact, the reflection characteristics change, providing self-generated position information without requiring external sensors. This maintains operational flexibility while enabling accurate positional awareness for reconstruction.
3Loss of information
If unclear image sensing regions are present in tomographic images, then diagnostic information is lost, but replacing them with three-dimensional volume data may impede clear image observation
Solution Approach 1:
The composite image applies different data sources to different regions: clear image sensing regions retain original high-resolution ultrasonic tomographic image data, while unclear regions (shadow regions and posterior echo regions) are replaced with corresponding data from three-dimensional volume images. This local differentiation maintains overall image reliability while recovering lost diagnostic information in problematic areas.
Data Source
AI summary
Based on the position and orientation information of an ultrasonic probe, an image generation unit acquires, from three-dimensional volume data, an image of a slice corresponding to a tomographic image of an object obtained by the ultrasonic probe. An image composition unit composes the image of the slice with the tomographic image to generate and output a composite image. In this composition, a region of the tomographic image is specified as an unclear image sensing region, and the image in the unclear image sensing region is replaced with the image in a region of the image of the slice corresponding to the unclear image sensing region, thereby composing the tomographic image with the image of the slice.