摘要
宫颈癌是严重威胁妇女健康的常见恶性肿瘤,早期筛查至关重要。针对我国优质医疗资源分布不均问题,通过人工智能辅助分析,研究细胞病理图像的染色归一化和分割技术。改进的Macenko染色归一化方法提高了图像的一致性,改进的Mask R-CNN模型增强了细胞重叠和遮掩情况下的分割性能。实验结果表明,这些改进显著提升了图像分割的准确性和稳定性。
Cervical cancer is a common malignant tumor that seriously threatens women’s health.So early screening is very important.In order to solve the problem of uneven distribution of high-quality medical resources in China,the study analyzes the staining normalization and segmentation techniques of cytopathological images with AI-assisted analysis.The improved Macenko staining normalization method improves the image consistency,and the improved Mask R-CNN model enhances the segmentation performance in the case of cell overlap and masking.The results of the experiment show that these improvement significantly improve the accuracy and stability of image segmentation.
作者
汤傲伟
Tang Aowei(Wuhan University,Wuhan 430000,China)
出处
《黑龙江科学》
2024年第18期128-130,133,共4页
Heilongjiang Science
