2013 Summer Project Week:DWIConvert

• 

  Projects List

Back to Summer project week Agenda

Key Investigators

• UIowa: Kent Williams, esqr.

Objective

Improve the DICOM - DWI conversion mechanism in Slicer to support more scanners and file output formats.

Specific goals

1. Working with the PREDICT and HDNI longitudinal data sets as sample multi-site, multi-scanner test data corpus, add support for more scanning protocols and software versions.
2. General code clean-up -- DicomToNrrd had been repeatedly modified over the past few years by multiple authors, to the point where there is many instances of 'dead' code (never executed) and confusing paths of execution.

Approach, Plan

1. Approach
   1. Begin with original DicomToNrrd code.
   2. Analyze source code to capture the implicit knowledge about DWI DICOM data sets.
   3. Refactor code to straighten out program logic and remove obsolete and incorrect code.
2. Plan
   1. Incrementally refactor and improve C++ code
   2. Use regression tests, and tests against large collections of DICOM data sets to validate code
   3. Incorporate program into BRAINSTools suite (hosted at github) to facilitate inclusion in Slicer.
   4. Work with Slicer community to come up with new 'problem' datasets for testing.

Progress

• Refactored original monolithic code into a collection of classes, one per scanner manufacturer, with common code isolated from scanner-specific code.
• More modular structure makes it easier to add new scanner models.

Scanner Manufacturer	Software version
Seimans	Avanto Espree Sonata Symphony SymphonyTim TrioTim Verio
GE	GENESIS_SIGNA SIGNA SIGNA EXCITE SIGNA HDx Signa HDxt
Philips	Achieva Intera
Hitachi	In Progress

Delivery Mechanism

This work will be delivered to the NA-MIC Kit as a (please select the appropriate options by noting YES against them below)

1. Part of BRAINSTools suite https://github.com/BRAINSia/BRAINSTools
2. Slicer Module
   1. Built-in YES
   2. Extension -- commandline
   3. Extension -- loadable