| 潘清,王硕,吕雪倩.基于多尺度分形维数加权损失和注意力机制的鸡胚胎血管分割方法[J].高技术通讯(中文),2026,36(7):701~712 |
| 基于多尺度分形维数加权损失和注意力机制的鸡胚胎血管分割方法 |
| Segmentation of chicken embryo vascular network based on multiscale fractal dimension weighted loss and attention mechanism |
| |
| DOI：10. 3772 / j. issn. 1002 - 0470. 2026. 07. 004 |
| 中文关键词: 血管分割； 注意力机制； 分型维数； 多尺度损失函数； 活体显微镜成像 |
| 英文关键词: vessel segmentation, attention mechanism, fractal dimension, multiscale loss function, in vivo microscopic imaging |
| 基金项目: |
| 作者 | 单位 | | 潘清 | （浙江工业大学信息工程学院杭州 310023） | | 王硕 | | | 吕雪倩 | |
|
| 摘要点击次数: 69 |
| 全文下载次数: 52 |
| 中文摘要: |
| 活体显微镜图像具有噪声高、动态性强的特点，而传统血管分割方法在应对这些挑战时经常会出现血管断连、预测错误的问题。为了解决这一问题，本研究设计了一种基于多尺度分形维数加权损失和注意力机制的深度学习模型，该模型通过分形维数加权损失和浅层注意力机制将血管网络的空间复杂度融入训练过程，同时通过多尺度结构和深层注意力机制保留了模型多层特征提取的优势和全局信息。在提出的私有数据集鸡胚血管网络图像数据上进行训练和测试，并通过血管断连数量和传统指标测试了其分割性能。结果显示，本研究提出的模型在整体上将Dice系数(Dice coefficient)从0.896提升至0.906，平均交并比(mean intersection of union，mIoU)从0.813提升至0.830，准确率(accuracy)从0.885提升至0.938，ROC曲线下面积(area under curve，AUC)从0.914提升至0.930，血管断连点的平均数量从15.7减少到11.3。这表明模型在维持分割性能的同时，能够有效地减少血管断连的数量，提高了血管网络结构的连接性。 |
| 英文摘要: |
| In vivo microscopy images are characterized by high noise and strong dynamics, and traditional vascular segmentation methods often encounter problems such as vascular discontinuity and prediction errors in addressing these challenges. To address this issue, this study designs a deep learning model based on multi-scale fractal dimension-weighted loss and attention mechanism. This model integrates the spatial complexity of the vascular network into the training process through fractal dimension-weighted loss and shallow attention mechanisms, while preserving the advantages of multi-layer feature extraction and global information through multi-scale structures and deep attention mechanisms. The model is trained and tested on a proprietary dataset of chicken embryo vascular network images, and its segmentation performance is evaluated through the number of vascular discontinuities and traditional metrics. The results show that the proposed model increases the Dice (Dice coefficient) index from 0.896 to 0.906, the mean intersection over union (mIoU) from 0.813 to 0.830, the accuracy from 0.885 to 0.938, the area under curve (AUC) from 0.914 to 0.930, and reduces the average number of vascular discontinuity points from 15.7 to 11.3. This indicates that the model effectively reduces the number of vascular discontinuities while maintaining segmentation performance, thereby improving the connectivity of the vascular network structure. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
