成果報告書詳細
管理番号20130000000344
タイトル*平成24年度中間年報 「IT融合による新社会システムの開発・実証プロジェクト (ヘルスケア分野)次世代医用クラウドシステムによる脳卒中に関する統合診断支援プラットフォームの構築」
公開日2014/8/20
報告書年度2012 - 2012
委託先名株式会社NTTデータ 株式会社NTTデータ経営研究所 岩手医科大学
プロジェクト番号P12009
部署名バイオテクノロジー・医療技術部
和文要約
英文要約The objective of this study is to develop a novel cloud service system in order to facilitate early diagnosis and prevention of various cerebrovascular disorders by establishing a medical imaging cloud communication and knowledge system (MICCS) that enables us to store and analyze brain images of patients and healthy elderly subjects. In the fiscal year 2012, several projects were undertaken by the business, application, and system working groups (WGs).
The business WG analyzed the actual operation of medical checkup systems for brain disorders in detail and summarized their business workflows. In addition, this WG identified and disentangled ethical and legal issues related to the MICCS service.
The application WG developed a program that can automatically extract cerebral white matter objects from MRI data as a component of the quantitative analysis application for asymptomatic white matter lesions in elderly subjects. By examining the accuracy of the program multilaterally, we found several issues that can affect the precision of the white matter extraction. Further, to develop an application for predicting stroke events in asymptomatic cerebrovascular lesions, we examined analysis workflows of the computational fluid dynamics (CFD) for unruptured aneurysms and investigated concrete configurations of the high performance computing (HPC) cluster system that will be installed in the next fiscal year.
The system WG built a virtualization system for multiple PC server functions including DICOM viewer software and deployed the Extensible Neuroimaging Archive Toolkit (XNAT) software platform on this system. We confirmed the feasibility of the XNAT pipeline engine by conducting a pilot study to test a workflow for automated signal intensity correction of the MRI data. We also customized the XNAT program to enable it to accept Japanese two-byte character codes. Further, to develop a scalable and secure platform, we investigated the basic functions and specifications of several distributed file systems of mass storage for numerous brain image data, such as OpenStack Swift, GlusterFS, and Hadoop. We also investigated the security requirements of MICCS as a business service according to the guidelines for secure management of healthcare information technology.
ダウンロード成果報告書データベース(ユーザ登録必須)から、ダウンロードしてください。

▲トップに戻る