肿瘤位置追踪方法、放射治疗设备和存储介质
By constructing a decoupling model, deep neural networks and traditional registration algorithms are used to decouple the patient's breathing and movement. Based on body surface and medical image information, real-time and accurate tracking of tumor location is achieved, solving the problem of inaccurate tumor location estimation in existing technologies and improving the precision of radiotherapy.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANGHAI UNITED IMAGING HEALTHCARE
- Filing Date
- 2024-07-17
- Publication Date
- 2026-07-17
AI Technical Summary
Current technology cannot accurately estimate the location of tumors during radiotherapy, especially when patients are breathing or moving, which increases the risk of radiation off-target, fails to effectively destroy malignant tumors, and may damage normal tissues.
By constructing a decoupling model, deep neural networks are used to decouple the patient's breathing and movement. Based on body surface image information and medical images, the tumor location is tracked in real time. A dataset containing respiratory phases and movement states is constructed using deep neural networks and traditional registration algorithms. The decoupling model is trained to output the current respiratory phase and movement state, eliminating the interference of breathing on movement calculation.
It enables accurate tracking of tumor location during radiotherapy, improving the accuracy of radiotherapy and reducing the risk of damage to normal tissues.
Smart Images
Figure CN121360343B_ABST