Intelligent Assistance in Diagnosis of Multi-dimensional Medical Images

多维医学影像智能辅助诊断

• 批准号: 15070101
• 负责人: KOBATAKE Hidefumi
• 金额: $ 23.62万
• 依托单位: Tokyo University of Agriculture and Technology
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Scientific Research on Priority Areas
• 财政年份: 2003
• 资助国家: 日本
• 起止时间: 2003 至 2006
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/en/grant/KAKENHI-PROJECT-15070101/
• 关键词: computer aided diagnosis; pattern recognition; multi-organ; multi-disease; visualization; CT image; 画像診断; 画像、文章、音声等認識; 知能機械; 計測工学

The objective of this research project is to develop a multi-organ, multi-disease CAD system that incorporates anatomical knowledge of the human body and diagnostic knowledge of various types of diseases.Digital atlases of human anatomy (DAHA) have been developed. Based on them various methods of segmentation of organs in the chest and/or abdominal 3D CT images have been developed. They include the bronchus, the blood vessels in the lung area, the diaphragm, the heart, the coronary artery, the stomach wall, the liver, the spleen, the left and right kidneys, the gallbladder, the pancreas, the colon, the portal vein (PV) between the spleen and liver, the esophagus, the abdominal aorta, and the inferior vena cava (IVC).The characteristics of abnormal regions depend on the type of disease. In order to identify these abnormalities, diagnostic knowledge of various types of diseases is essential. A database of such diagnostic knowledge on the major organs in chest and/or abdominal region, i. e. a digital representation of diagnostic knowledge (DRDK), has been developed.The DANA and the DRDK are large databases that can be achieved by the collaboration of several research groups. Various types of segmentation methods, organ modeling methods, lesion detection methods, diagnostic methods, etc., must be integrated into one system to develop a multi-organ, multi-disease CAD system. A common platform that can integrate all of those plug-in units has been developed. It is called Pluto. And a prototype of a comprehensive CAD system for multi-organ and multi-disease has been constructed on this platform.These results suggest that the goal of the research project has been achieved.

本研究课题的目的是开发一个多器官、多疾病的CAD系统，该系统融合了人体解剖学知识和各种疾病的诊断知识。基于它们，已经开发了在胸部和/或腹部3D CT图像中分割器官的各种方法。它们包括支气管、肺区的血管、隔膜、心脏、冠状动脉、胃壁、肝脏、脾脏、左右肾脏、胆囊、胰腺、结肠、脾和肝脏之间的门静脉（PV）、食道、腹主动脉和下腔静脉（IVC）。异常区域的特征取决于疾病的类型。为了识别这些异常，各种类型疾病的诊断知识是必不可少的。关于胸部和/或腹部区域中的主要器官的这种诊断知识的数据库，即。e.诊断知识的数字表示（DRDK）已经被开发。DANA和DRDK是可以通过几个研究小组的合作来实现的大型数据库。各种类型的分割方法、器官建模方法、病变检测方法、诊断方法等，必须集成到一个系统中，以开发多器官，多疾病的CAD系统。已经开发了一个可以集成所有这些插件单元的通用平台。它被称为冥王星。并在此平台上构建了一个多器官多疾病的综合CAD系统原型，初步实现了课题的研究目标。

项目成果

Rajalida Lipikorn: "A Modified Exoskeleton and a Hausdorff Distance Matching Algorithm for Shape-Based Object Recognition"Proc.of International Conference on Imaging Science, Systems, and Technology (CISST'03). Vol.2. 507-511 (2003)

Rajalida Lipikorn：“用于基于形状的物体识别的改进外骨骼和豪斯多夫距离匹配算法”Proc.of 国际成像科学、系统和技术会议 (CISST03)。

中川潤哉: "多時相の3次元マルチスライスCT像からの肝腫瘤の児童抽出手法の開発"電子情報通信学会論文誌. J87-D-II・1. 260-270 (2004)

Junya Nakakawa：“从多时相 3D 多层 CT 图像中提取儿童肝脏肿块的方法的开发”电子、信息和通信工程师学会会刊 J87-D-II・1（2004 年）。 ）

豊田修一: "指示データの集約化・視覚化による診療支援システムの開発"電子情報通信学会論文誌D-II. J89-D-II・7. 1111-1122 (2003)

Shuichi Toyoda：“通过汇总和可视化指令数据开发医疗保健支持系统”IEICE Transactions D-II·7（2003）。

H.Takizawa: "A RECOGNITION METHOD OF LUNG NODULE SHADOWS IN X-RAY CT IMAGES USING 3D OBJECT MODELS"International Journal of IMAGE AND GRAPHICS. 3・4. 533-545 (2003)

H.Takizawa：“使用 3D 对象模型的 X 射线 CT 图像中的肺结节阴影识别方法”国际图像与图形杂志 3・4（2003 年）。

X.Zhou: "Automated volume measurements of pulmonary emphysema on 3-D chest CT images"Proc.of CARS 2003,International Congress Series 1256. 1256. 1043-1048 (2003)

X.Zhou：“3-D胸部CT图像上肺气肿体积的自动测量”Proc.of CARS 2003，国际大会系列1256. 1256. 1043-1048 (2003)