A Deep Learning-Based Method for Detecting Giant Cells in Ovarian Cancer Polyploid Tumors from H&E Images
By constructing the OCDet model, the subjectivity and complexity of detecting polyploid tumor giant cells in H&E staining images were resolved, achieving efficient and accurate automated detection. This model is applicable to the detection of polyploid tumor giant cells in ovarian cancer, improving the efficiency and accuracy of pathological diagnosis.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- TSINGHUA SHENZHEN INTERNATIONAL GRADUATE SCHOOL
- Filing Date
- 2024-08-21
- Publication Date
- 2026-07-24
AI Technical Summary
Existing technologies for detecting ovarian cancer polyploid tumor giant cells in H&E-stained images suffer from high subjectivity, complex operation, and long processing time. Furthermore, deep learning models lack sufficient accuracy and stability in heterogeneous and small sample sizes.
An OCDet model was constructed, using CSPDarkNet as the feature extraction backbone and combining it with the attention mechanism ECA. The ECA-RER module enhances salient features and removes redundant features, while the ECA-MREF module performs multi-scale feature fusion. The loss function is optimized using the detection head, achieving efficient and accurate detection of polyploid tumor giant cells.
It enables rapid, accurate, and automated detection of polyploid tumor giant cells in H&E stained images, improving the efficiency of pathological diagnosis, reducing human error, and has good scalability and customizability, making it suitable for cell detection under different disease types and staining conditions.
Smart Images

Figure CN119151873B_ABST