论文部分内容阅读
目的:实现计算机自动检测出CT图像中的早期肿瘤。方法:提出一个以纹理能量分析为基础的可以检测出CT图像中早期肿瘤的算法——ITEAR(image registration and texture energy analysis)。该算法首先用互信息量配准方法将PET图像和同一层的CT图像配准,并且利用自适应阈值法将PET图像中的感兴趣区(region of interest,ROI)检测出来,然后应用纹理能量分析方法检测出早期肿瘤的准确位置,接着运用边缘检测方法将其边界提取出来。结果:ITEAR算法能准确定位早期肿瘤,并且能检测出早期肿瘤的边界。结论:ITEAR算法可以令计算机自动制定三维适形放疗计划。
Objective: To achieve the computer automatically detect CT images of early tumors. METHODS: An algorithm based on texture energy analysis that detects early tumors in CT images was proposed (image registration and texture energy analysis). The algorithm first registers the PET image and the CT image of the same layer with the mutual information registration method, and detects the region of interest (ROI) in the PET image by using an adaptive threshold method, and then applies the texture energy Analytical methods detect the exact location of early tumors and then use edge detection methods to extract the boundaries. Results: The ITEAR algorithm accurately locates early tumors and detects early tumor boundaries. Conclusion: The ITEAR algorithm allows computers to make three-dimensional conformal radiotherapy plans automatically.