基于深度学习的胰腺图像自动三维重建

打开文本图片集
摘 要: 胰腺图像的三维重建对于辅助疾病诊断具有重要意义。提出一种全自动的胰腺图像三维重建方法,利用改进的U-Net深度学习网络对图像进行分割,并结合面绘制算法生成三维可视化模型。实验结果表明,该方法重建准确度较高,执行效率快,对辅助诊疗具有积极的作用。
关键词: 胰腺; 深度学习; 面绘制; 三维重建
中图分类号:TP391.9 文献标识码:A 文章编号:1006-8228(2022)01-68-04
Automatic 3D reconstruction of pancreas image based on deep learning
Ji Jianbing1, Zhang Jinglin1, Yang Yuanyuan2
(1. College of Information Engineering, Fujian Business University, Fuzhou, Fujian 350012, China;
2. Department of General Surgery, Fujian Medical University Union Hospital)
Abstract: The 3D reconstruction of the pancreas images is of great significance for assisting disease diagnosis. A fully automatic 3D reconstruction method of the pancreas is proposed, which uses an improved U-Net deep learning network to segment the images, and creates a 3D visual model by surface rendering algorithm. The experiment results show that the proposed method has high reconstruction accuracy and fast execution efficiency. Therefore, it has a positive effect on assisting diagnosis and treatment.
Key words: pancreas; deep learning; surface rendering; 3D reconstruction
0 引言
胰腺癌的治疗主要在于对局部肿瘤做有效的切除手术,但手术存在很大的风险。(剩余4382字)