针对肝脏肿瘤CT图像灰度差别小,边缘不明显的特点,为了提高分割精度提出一种基于极值自适应中值滤波与现有区域生长算法相结合的一种新的肝癌分割算法.采用一组临床肝癌CT图像样本进行图像预处理,利用此算法进行第一次滤波,再采用傅里叶变换方法进行第二次滤波,最后与现有算法结合完成分割.采用重叠错误率、相对误差和Dice相似性系数作为图像分割结果评价指标.通过定性分析和定量评价显示,基于本研究算法能精确分割出肝癌区域,获得的图像分割评价指标均优于其他常用算法.

Liver cancer is a common malignant tumor and one of the main causes of cancer deaths in the world. Accurately segmenting liver tumor regions from the liver has a very important guiding role for doctors in disease diagnosis and surgical planning. Thus, the method of automated segmentation of liver tumors has important value for clinical diagnosis and treatment. This paper proposes a method that combines fully convolutional neural networks based on two-dimensional convolution and those based on three-dimensional convolution. The results show that the liver tumor segmentation effect of this method is much better than that of only using single two-dimensional convolutional neural networks or three-dimensional convolutional neural networks.

Hepatocellular carcinoma (HCC) is the third most common cancer in the world, which seriously threatens people's life and health. With the rapid development of medical imaging technology, it is of great significance to classify the benign and malignant tumors of HCC through computer-aided techniques. In this paper, a classification algorithm using the multiple deep features fusion is designed. By designing the feature fusion layer network based on the deep learning framework, the deep features and textural features are integrated; the end-to-end classification algorithm using deep learning feature fusion is realized, which has higher classification accuracy and algorithm stability, and can get better classification results than the traditional textural features.

教育现代化的进行在网络信息技术飞速发展的态势下呈现高速发展,资源建设作为教育信息化的核心则至关重要.Moodle平台是国内外应用最为广泛的一款免费、开源的课程管理系统,应用广泛,其中医学类微视频资源的建设可以让学生利用碎片化时间进行学习,减少授课教师的授课压力,应进行深入研究与建设.