CTSC IGT, BWH

Back to CTSC Imaging Informatics Initiative

Mission

Mark Anderson at Surgical Planning and Channing labs currently manages data for many investigators, pulling data from PACS into the research environment. There is interest in setting up a parallel channel by which the data are also enrolled into an XNAT database and accessed from client, and comparing its ease of use with the existing infrastucture. To explore XNAT as a possible long-term informatics solution for the NCIGT project, Mark will be uploading retrospective data for a number of NCIGT efforts (and PIs):

• NCIGT_Brain_Function (SS/AG)
  • Key Investigators:
  • Brief Description:
  • Use: What kinds of queries will be important?
• NCIGT_Tumor_Resection (HK/AG)
  • Key Investigators:
  • Brief Description:
  • Use: What kinds of queries will be important?
• NCIGT_Prostate (HE/CT)
  • Key Investigators:
  • Brief Description:
  • Use: What kinds of queries will be important?
• NCIGT_Prostate_Fully_Segmented (HE/CT)
  • Key Investigators:
  • Brief Description:
  • Use: What kinds of queries will be important?
• NCIGT_Brain_Biopsy (FT)
  • Key Investigators:
  • Brief Description:
  • Use: What kinds of queries will be important?

Use-Case Goals

Step 1. Data Management

• Anonymize, apply DICOM metadata and upload retrospective datasets; confirm appropriate organization and naming scheme via web GUI.

Step 2. Query & Retrieval

• Make specific queries using XNAT web services,
• Download data conforming to specific naming convention and directory structure, using XNAT web services

Each effort listed above will have different requirements for being able to query, retrieve and use data collections. Brief description of how retrospective data will be used within the NCIGT is described below:

• NCIGT_Brain_Function:
• NCIGT_Tumor_Resection:
• NCIGT_Prostate:
• NCIGT_Prostate_Fully_Segmented:
• NCIGT_Brain_Biopsy:

Step 3. Disseminating & Sharing

• In addition to NCIGT mandate to share data, each effort listed above will have different requirements for being able to make data available to collaborating and other interested groups.

Step 4. Moving data from central.xnat.org to BWH instance of XNAT

Outcome Metrics

Step 1. Data Management

Step 2. Query & Retrieval

Step 3. Dissemination & Sharing

Participants

• Mark Anderson
• Tina Kapur

Workflows

Current Data Management Process

Data on local disk.

Target Data Management Process (Step 1.) Option A. interactive upload using various tools and web gui

See here

Target Data Management Process (Step 1.) Option B. batch scripted upload via DICOM Server

See here

Target Data Management Process (Step 1.) Option C. batch scripted upload via web services

See here