欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
乳腺癌是威胁我国妇女身体健康最主要的恶性肿瘤之一,乳腺钼靶图像中肿块的自动检测是乳腺癌的计算机辅助诊断领域的研究热点。提出一种基于海岛冲刷模型的肿块检测新算法,通过模拟一个海水上涨并不断侵蚀大陆与海岛的迭代过程,逐步剥离乳房中的脂肪甚至腺体组织;通过模拟海岛居民防洪筑坝的行为,不断维护疑似肿块形态的完整性,从而最终达到对肿块的分离与检测。在整个检测流程中,使用了模糊神经网络技术对一些应变参数进行自适应调节。实验结果表明,与一些传统的检测方法相比,海岛冲刷模型对肿块尤其是一些隐匿性肿块拥有较高的检测精度,在每幅图像有3.85个假阳性时能获得94.31%的真阳性检出率。
英文摘要：
Breast cancer is one of the most dangerous cancers facing Chinese women today and computer-aided diagnosis（CAD） to detect mammographic masses is being under extensive study.A novel detection algorithm for mammographic masses based on island scouring model（ISM） is proposed in this paper.It gradually peels off the fatty tissue and the glandular tissue from the breast,by simulating an iterative process of sea level rising and scouring lands and protects the integrity of the shape of the suspicious masses by mimicking the behavior that the people build up the dam against the sea;finally the suspicious masses could be segmented and detected.During the detection process,adaptive-network-based fuzzy inference system（ANFIS） is applied to adjust the detection parameters adaptively.The experimental result revealed that compared with the conventional detection method,ISM possessed a totally higher detection precision especially for some subtle masses.It could achieve a true positive rate of 94.31% as well as 3.85 false positives per image.
同期刊论文项目
同项目期刊论文
Improving the performance of computer-aided detection of subtle breast masses using an adaptive cuei
期刊信息
位置：