Edit this page


Back to Projects List

Integration of experimental parameters from the Free WaveForm sequence into DICOM

Key Investigators

Project Description

Tensor-valued diffusion encoding is an emerging technique within diffusion MRI. The novelty lies in the multidimensional encoding of the diffussion process. To support such encoding, we must replace the conventional diffusion encoding sequence with arbitrary gradient waveforms that facilitate a large range of encoding strategies. In doing so, we also need to store new kinds of experimental information to support traceability. In this project, we will extend the MRI pulse sequence to calculate and embed several necessary parameters in the DICOM header.

Constraints and Challenges


  1. Incorporate array-to-character conversion in pulse sequence (Prepared by IN)
  2. Subsample/rotate waveform and store in the WIP memory block
  3. Complete waveform parameterization (moments, concomitant fields, etc)
  4. Validate sequence functionallity in simulator
  5. Validate sequence functionallity on hardware (Boston)
  6. Write Matlab script to extract and validate waveform from DICOM header
  7. Validate waveform integrity (Boston)

Approach and Plan

  1. Objectives 1-4 and 6 will be approached during the project week, in that order.

Progress and Next Steps

Objective 1 was prepared during Hackathon in Sweden by Isaiah Norton, and is almost complete.

Next step: Introduce C-code into pulse sequence, and test compilation conflicts. Translate “Load” type waveform, without rotations, into the WIP parameter space Complement waveform parameters (motion, acceleration, k0 nulling with maxwell terms, etc) and store in header


Example waveform that compares a conventional Stejskal-Tanner experiment (gray) to freely modulated gradient waveforms that produce spherical tensor encoding.

Parameters supported by tensor-valued diffusion encoding at multiple MRI systems

Background and References