Neurosurgery intelligent planning method and system based on AI image recognition
By using AI image recognition technology, combined with the U-Net model and Lucas-Kanade algorithm, and employing dual-color fluorescent probes and dynamic optical flow fields to plan neurosurgical surgical paths, the challenges of vascular boundary identification and path planning in traditional microsurgery have been solved, achieving accurate identification and safe and efficient surgical procedures.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- FOURTH MILITARY MEDICAL UNIVERSITY
- Filing Date
- 2026-04-29
- Publication Date
- 2026-07-17
AI Technical Summary
In neurosurgical and vascular microsurgery, traditional methods struggle to accurately identify vascular boundaries, distinguish between arteries and veins, and lack quantitative basis for marking danger zones. Furthermore, the highly subjective nature of surgical path planning leads to high risks of vascular injury and low efficiency.
AI image recognition technology is used to acquire preoperative images through a high-resolution microscope. Combined with the U-Net model and the improved Lucas-Kanade algorithm, the system identifies vessel wall movement and bifurcation danger zones. Dual-color fluorescent probes are used to distinguish between arteries and veins. The system also plans safety boundaries and paths by combining dynamic optical flow fields and optical flow phase differences.
It enables precise identification of vascular boundaries and accurate differentiation of arteries and veins, objectively marks dangerous areas, plans safe and efficient surgical paths, reduces the risk of vascular injury, and improves surgical efficiency.
Smart Images
Figure CN122415564A_ABST