Modular algorithm for automatic positioning of features in images
By using personalized artificial intelligence models and fully convolutional neural networks (CNNs) to detect and locate vertebrae, sacrum, and femoral heads, the accuracy and efficiency issues of anatomical feature localization in two-dimensional medical images in existing technologies have been solved, achieving faster execution speed and flexible software integration.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- MAZOR ROBOTICS
- Filing Date
- 2024-12-19
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies for automatically locating anatomical features in two-dimensional medical images suffer from inaccuracies, lack of data variability, complex development requirements, slow runtime, incompatible software integration, and inability to extend to other application functions.
A personalized artificial intelligence model is used to implement a fully convolutional neural network (CNN) using processing circuits and storage media to detect and locate vertebrae, sacrum and femoral head. The multi-level algorithm program improves accuracy and flexibility, simplifies the development process and software integration.
It improves the accuracy of anatomical feature localization in spinal images, simplifies the development process of AI systems, increases execution speed, and simplifies software integration and function expansion.
Smart Images
Figure CN122422902A_ABST