改进YOLOv5检测宫颈癌双染图像，提升准确率与稳定性。
Research on Cervical Cancer p16/Ki-67 Immunohistochemical Dual-Staining Image Recognition Algorithm Based on YOLO
- 融合Swin-Transformer与注意力机制，增强特征提取能力。
- [email protected]达92.6%，[email protected]:0.95达70.5%，显著优于基线模型。
- 适用于病理图像识别，尤其适合医疗影像自动化分析场景。
p16/Ki-67双染法是宫颈癌筛查的新方法，具有高敏感性和特异性。然而，直接使用YOLOv5s算法处理双染细胞图像时存在误检和识别不准的问题。本文提出基于YOLOv5的宫颈癌双染图像识别模型DSIR-YOLO，通过融合Swin-Transformer模块、GAM注意力机制、多尺度特征融合及EIoU损失函数，显著提升检测性能，[email protected]和[email protected]:0.95分别达到92.6%和70.5%。在五折交叉验证中，该模型的准确率、召回率、[email protected]和[email protected]:0.95较YOLOv5s分别提升2.3%、4.1%、4.3%和8.0%，方差更小，稳定性更高。同时研究了数据集质量对检测结果的影响，通过控制像素密封性、尺度差异、未标注细胞及对角标注等，模型各项指标提升分别为13.3%、15.3%、18.3%和30.5%。
原文摘要 · Abstract (English)
The p16/Ki-67 dual staining method is a new approach for cervical cancer screening with high sensitivity and specificity. However, there are issues of mis-detection and inaccurate recognition when the YOLOv5s algorithm is directly applied to dual-stained cell images. This paper Proposes a novel cervical cancer dual-stained image recognition (DSIR-YOLO) model based on an YOLOv5. By fusing the Swin-Transformer module, GAM attention mechanism, multi-scale feature fusion, and EIoU loss function, the detection performance is significantly improved, with [email protected] and [email protected]:0.95 reaching 92.6% and 70.5%, respectively. Compared with YOLOv5s in five-fold cross-validation, the accuracy, recall, [email protected], and [email protected]:0.95 of the improved algorithm are increased by 2.3%, 4.1%, 4.3%, and 8.0%, respectively, with smaller variances and higher stability. Compared with other detection algorithms, DSIR-YOLO in this paper sacrifices some performance requirements to improve the network recognition effect. In addition, the influence of dataset quality on the detection results is studied. By controlling the sealing property of pixels, scale difference, unlabelled cells, and diagonal annotation, the model detection accuracy, recall, [email protected], and [email protected]:0.95 are improved by 13.3%, 15.3%, 18.3%, and 30.5%, respectively.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。