OCT图像的多病灶并行检测方法、系统、设备和介质
The OCT image detection method, which combines feature extraction algorithms and multi-task learning algorithms, solves the problem of parallel detection of multiple lesions, achieves efficient and accurate detection of multiple lesions, and reduces the reliance on doctors' professional skills.
CN117649382BActive Publication Date: 2026-07-17CHINA UNITED NETWORK COMM GRP CO LTD +1
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHINA UNITED NETWORK COMM GRP CO LTD
- Filing Date
- 2023-11-29
- Publication Date
- 2026-07-17
AI Technical Summary
Technical Problem
Current technology cannot detect multiple lesions that may exist in OCT images in parallel, resulting in low detection efficiency and an increased possibility of misdiagnosis.
Method used
A combination of feature extraction and multi-task learning algorithms is used to extract and classify features from OCT images. An information fusion attention mechanism is introduced to fuse patient information with the classification results of lesion types, enabling parallel detection of multiple lesions.
Benefits of technology
It improves the efficiency and accuracy of multi-lesion detection in OCT images, reduces the possibility of misdiagnosis, and reduces reliance on doctors' professional skills.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN117649382B_ABST
Abstract
本发明公开一种OCT图像的多病灶并行检测方法、系统、设备和介质，涉及医学图像处理技术领域。OCT图像的多病灶并行检测方法包括：获取待检测的OCT图像。对待检测的OCT图像进行预处理。采用特征提取算法对预处理后的待检测的OCT图像数据进行第一次特征提取。采用多任务学习算法对第一特征集进行第二次特征提取。获取与待检测的OCT图像对应的患者信息，分别将至少一个病灶类型的分类结果中的每一个分类结果与患者信息进行融合。以及，获取待检测的OCT图像包括的至少一个病灶类型的检测结果。将特征提取算法与多任务学习算法进行结合，将患者信息引入到多病灶并行检测方法中，实现对多个病灶的检测，提升检测的效率与准确率。
Need to check novelty before this filing date? Find Prior Art