Vertebra Labeling in Partial Spine Medical Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing medical image processing methods fail to accurately label vertebrae in images that only include a portion of the spine, such as the thoracic spine, leading to incomplete recognition and labeling.
Innovation Solution
A medical image processing apparatus and method that determines the presence of at least a portion of the upper or lower end vertebra, obtains a secondary image with recognizable vertebra labels, and aligns the images to label the vertebrae in the primary image using the secondary image, even if the primary image only includes a portion of a vertebra.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If existing medical image processing methods are used, then vertebrae can be labeled in images with complete spinal anatomy, but vertebrae cannot be labeled in images that only include a portion of the spine
Solution Approach 1:
The system performs preliminary actions by obtaining a reference medical image that contains complete or more complete spinal anatomy before attempting to label vertebrae in the target partial spine image. This reference image is processed in advance to establish vertebrae labeling that can be subsequently applied to the partial image through alignment, enabling the system to handle partial spine images that would otherwise be unlabeled.
Solution Approach 2:
The invention introduces a reference medical image as an intermediary element between the partial spine target image and the vertebrae labeling process. This reference image serves as a mediator that contains the complete anatomical context needed for accurate labeling, which is then transferred to the partial image through image alignment techniques, bridging the gap between incomplete target data and complete labeling requirements.
2Area of stationary object
If the image only includes a portion of the spine, then the imaging scope is reduced, but the vertebrae cannot be recognized and labeled
Solution Approach 1:
The system achieves universality by making the labeling approach applicable to both complete spine images and partial spine images through a unified process. The reference image methodology allows the same labeling pipeline to handle varying imaging scopes, whether the target image contains the entire spine or only a portion, by adapting to the available anatomical information through alignment with the reference.
Solution Approach 2:
The invention transitions from relying solely on the two-dimensional information within the partial target image to incorporating information from another dimension by introducing the reference medical image. This additional dimensional source provides the missing anatomical context, allowing the system to reconstruct complete vertebrae labeling information even when the target image only covers a portion of the spine.
3Measurement precision
If reference vertebrae are used for labeling, then labeling accuracy improves, but the process complexity increases due to image alignment requirements
Solution Approach 1:
The system employs copying by creating an aligned version of the reference medical image that matches the geometry and orientation of the target partial spine image. This copied and transformed reference data is then overlaid or integrated with the target image, allowing vertebrae labeling to be transferred accurately without requiring complex manual intervention, as the alignment process automatically handles the geometric transformation.
Data Source
AI summary
A determination unit makes a determination as to whether or not at least either one of at least a portion of an upper end vertebra and at least a portion of a lower end vertebra is included in a first medical image of a subject. If the determination is negative, an image obtaining unit obtains a second medical image that allows recognition of a label of the vertebra of the subject. A labeling unit aligns the first medical image with the second medical image and labels the vertebra included in the first medical image.