• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>Proceedings of the world medical conference >Novel mining algorithm for multiple level classification of brain tumors
【24h】

Novel mining algorithm for multiple level classification of brain tumors

机译：脑肿瘤多层次分类的新型挖掘算法

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

This paper proposes the various levels of rumors in CT scan brain images, which can assist the medical image diagnosis system. A tumor is a pattern of abnormal cells which are resulting from unwanted changes in the genetic material (i.e. chromosomes) that makes the body cells lose the ability to control their growth. If the tumor does not invade the nearby tissues and body parts, it is called benign tumor, or non-cancerous growth. In contrast, if the tumor invades and destroys the nearby cells, it is called malignant tumor, which is usually life threatening. Brain tumors can be classified into two main categories, primary and secondary. Primary brain tumors (gliomas) are the tumors that start in the brain, whereas the secondary brain tumors result from cancer that starts elsewhere in the body and spreads to the brain (i.e. metastasized). Such a kind of tumors is usually malignant and more common among brain tumor incidences. In Canada, brain cancer that result from malignant tumors causes a large amount of deaths every year; 1650 deaths (out of 2,500 diagnosed cases) were expected in 2005. Regardless of their growth rates, both the malignant and benign tumors have similar effects on the brain. Image segmentation is, arguably, the most important component in the medical image mining process; it is certainly the start point. Segmentation is concerned with the automated division of images into non-overlapping regions. A high speed parallel fuzzy C-means (FCM) algorithm proved to be superior over the other clustering approaches in terms of segmentation efficiency. This proposed algorithm has the advantages of both the sequential FCM and parallel FCM for the clustering process in the segmentation techniques. This algorithm is very fast when the image size is large and it requires less execution time. We have also achieved less processing speed and minimizing the need for accessing secondary storage compared to the previous results. The reduction in the computation time is primarily due to the selection of actual cluster centre and the accessing minimum secondary storage.. Abnormal brain images from four tumor classes namely metastase, meningioma, glioma and astrocytoma are used in this work. In this work, we take advantage of association rule mining. The method proposed here makes use of association rule mining technique to classify the CT scan brain images. It combines the low-level features extracted from images and high level knowledge from specialists. The Experimental results on pre-diagnosed database of brain images shows high accuracy (up to 98%), allowing us to claim that the use of associative classifier is an efficient technique to assist in the diagnosing task. This paper presents a fast association rule mining algorithm which is suitable for medical image data sets. In particular, it assesses the feasibility of using association rule algorithms to extract hidden information fiom medical image data setsThe main objective of this work is to devise a computational technique that analyzes CT data by means of association rule mining to come up with a set of information that help in assessing brain tumors.
机译：本文提出了CT扫描脑部图像中各种级别的谣言，它们可以帮助医学图像诊断系统。肿瘤是异常细胞的一种模式，这种异常细胞是由于遗传物质（即染色体）发生意外变化而导致机体失去控制其生长的能力。如果肿瘤没有侵入附近的组织和身体部位，则称为良性肿瘤或非癌性生长。相反，如果肿瘤侵入并破坏附近的细胞，则称为恶性肿瘤，通常威胁生命。脑肿瘤可分为两个主要类别，即原发性和继发性。原发性脑部肿瘤（神经胶质瘤）是始于脑部的肿瘤，而继发性脑部肿瘤则是由始于身体其他部位并扩散至脑部（即已转移）的癌症引起的。这类肿瘤通常是恶性的，在脑部肿瘤的发病率中更为常见。在加拿大，由恶性肿瘤引起的脑癌每年导致大量死亡。预计2005年将有1650例死亡（在2500例确诊病例中）。无论其增长率如何，恶性和良性肿瘤对大脑的影响都相似。可以说，图像分割是医学图像挖掘过程中最重要的组成部分。这当然是起点。分割与将图像自动划分为非重叠区域有关。事实证明，在分割效率方面，高速并行模糊C均值（FCM）算法优于其他聚类方法。该算法在分割技术的聚类过程中具有顺序FCM和并行FCM的优点。当图像较大时，此算法非常快，并且需要较少的执行时间。与之前的结果相比，我们还实现了更低的处理速度，并最大程度地减少了访问二级存储的需求。计算时间的减少主要是由于选择了实际的簇中心和访问了最少的次要存储空间。这项工作使用了来自四种肿瘤类别的异常脑图像，即转移，脑膜瘤，神经胶质瘤和星形细胞瘤。在这项工作中，我们利用关联规则挖掘。本文提出的方法利用关联规则挖掘技术对CT扫描脑图像进行分类。它结合了从图像中提取的低级功能和专家的高级知识。在预先诊断的大脑图像数据库上的实验结果显示出很高的准确性（高达98％），这使我们可以断言关联分类器的使用是一种有助于完成诊断任务的有效技术。本文提出了一种适用于医学图像数据集的快速关联规则挖掘算法。特别是，它评估了使用关联规则算法从医学图像数据集中提取隐藏信息的可行性。这项工作的主要目的是设计一种计算技术，该技术通过关联规则挖掘来分析CT数据以得出一组信息。有助于评估脑瘤。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号