一种基于肝分段和肝癌轮廓融合的放疗靶区数据库建立方法

提出一种基于肝分段和肝癌轮廓融合的放疗靶区数据库建立和全过程数据质量管理方法，为后续人工智能靶区勾画或评估当前手工勾画提供数据支持。从肝癌数据库中取出原始图像，分别对其做带有肝脏放疗靶区勾画和分区分段轮廓标注工作，并通过图像融合技术使肝癌的放射治疗精确到肝段，最后使用深度学习的方法训练Unet网络以得到精准肝分割的神经网络模型，以实现针对肝癌的精准放疗。

A method for establishing radiotherapy target area database based on the fusion of liver segment and liver cancer contour

Abstract: A method for radiotherapy target database establishment and the whole process data quality management based on the fusion of liver segment and liver cancer contour is proposed to provide data support for subsequent artificial intelligence-based target segmentation or evaluation of current manual delineation. The original images are selected from the liver cancer database for target segmentation in liver radiotherapy and labeling the contour in sections and segments. Through image fusion technology, the radiotherapy of liver cancer can be accurate to the liver segment. Unet is trained with deep learning method to obtain the neural network model for accurate liver segmentation, so as to achieve precise radiotherapy for liver cancer.