A plane growth-based joint image segmentation method
By using a planar growth-based method combined with binarization and hole filling, the problems of blurred boundaries and holes in joint image segmentation were solved, achieving accurate segmentation of joint images and improving the accuracy of joint replacement surgery.
CN117095017BActive Publication Date: 2026-02-06NANJING TUODAO MEDICAL TECHNOLOGY CO LTD
Patent Information
- Application Number
- CN202311154600.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-09-08
- Publication Date
- 2026-02-06
- Estimated Expiration
- 2043-09-08
AI Technical Summary
Technical Problem
Existing techniques are difficult to accurately separate the bones on both sides of the joint from the cartilage in the middle, and have defects such as cavities, which affect the accuracy of joint replacement surgery.
Method used
A planar growth-based method is adopted. By acquiring joint region images, performing binarization processing, calculating the area of the foreground region, selecting seed points for planar growth, and combining hole filling and morphological closing operations, accurate segmentation of joint images is achieved.
Benefits of technology
It enables accurate segmentation of joint images, eliminates cavity defects, and improves the accuracy and effectiveness of joint replacement surgery.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN117095017B_ABST
Abstract
The application discloses a joint image segmentation method based on plane growth, comprising the following steps: S1, acquiring a joint region image and performing binary processing to obtain a foreground region and a background region; S2, calculating the area of the foreground region of each axial plane slice image in the joint region image, and obtaining the position corresponding to the slice image corresponding to the minimum value between the two maximum values of the plurality of areas as the segmentation boundary; S3, selecting a plurality of center points in the body head-to-foot direction in the background region, and taking the point closest to the segmentation boundary in the plurality of center points as a seed point; S4, performing plane growth with the seed point to obtain a segmentation plane, and segmenting the joint region image according to the segmentation plane. The application can accurately segment the bone on both sides of the joint and the cartilage part in the middle, and the segmented image is free of defects such as holes.
Need to check novelty before this filing date? Find Prior Art
Citation Information
Patent Citations
Method and system for abdominal bone segmentation in medical image
CN105701795A