一种融合深度学习的染色体核型智能分割方法及系统
By integrating deep learning methods with image enhancement and region segmentation techniques, chromosome adhesion regions are screened and segmented, solving the problems of time-consuming, labor-intensive, and subjective manual interpretation in existing technologies, and achieving high accuracy and high stability in chromosome karyotype segmentation.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- HUNAN INST OF INFORMATION TECH
- Filing Date
- 2026-04-21
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies rely on manual interpretation in chromosome karyotype analysis, which is time-consuming, labor-intensive, and subject to subjective differences. This makes it difficult to meet the clinical needs for high accuracy and stability, especially when faced with complex situations such as chromosome morphological diversity, blurred boundaries, and chromosome adhesion and overlap, where the segmentation accuracy and robustness are insufficient.
By employing a deep learning-integrated approach, the pixel area and band count of chromosomes are extracted through image enhancement and region segmentation techniques. The structural coverage index is used to screen candidate adhesion regions, and a deep learning segmentation model is combined for refined processing to identify and segment chromosome adhesion regions.
It improves the accuracy of chromosome image segmentation, reduces error propagation, enhances the segmentation ability of adhered chromosomes, and achieves higher segmentation precision and stability.
Smart Images
Figure CN122090448B_ABST