Table of contents

• Rationale
• General guidelines
• Guidelines for specific scan types
  • Implemented or partially implemented
    • Multi-echo gradient echo
    • Multi-echo spin echo
    • T1w / T1w-FS / T2w / T2w-FS
    • Quantitative T1 / T2 / wT2
  • Not implemented / Proposed
    • Phase contrast
    • Diffusion
    • Segmentation labels

Rationale

The goal of this standard is to have a uniform interface for software tools for muscle imaging.

Ideally, each tool will have a command-line interface with the following syntax:

muscle_tool <patient_folder>

Where the patient_folder is the location of the root folder containing all the acquired data for a patient, divided in subfolders according to the BIDS standard and as described below. The muscle tools will also output data in the correct format in the same folder structure.

General guidelines

• Each Muscle-BIDS (following: mBIDS) dataset is composed of:
  1. One NIfTI file containing the image data. This file is named <patient>_<acquisition_type>.nii.gz
  2. One JSON header named <patient>_<acquisition_type>.json (subsequently simply termed header) containing the imaging parameters relevant to the specific acquisition type.
  3. One JSON header named <patient>_<acquisition_type>_patient.json containing the patient information. This file can be omitted for anonymization purposes.
  4. One JSON header named <patient>_<acquisition_type>_extra.json containing a dictionary of DICOM fields used to convert the data back to DICOM.
• When it makes sense, 4D NIfTI datasets are used instead of multiple 3D files (for example, multi-echo data are stored as a fourth NIfTI dimension). The quantity that changes in the fourth dimension is identified by the parameter FourthDimension in the header. E.g. "FourthDimension": "EchoTime". The value of the FourthDimension field must have a corresponding entry in the header, which is a list of values corresponding to each index along the fourth dimension. E.g. "EchoTime": [3.9, 6.3].
• Multislice 2D acquisitions are saved as a 3D NIfTI volume. To store the actual slice thickness, the AcquisitionVoxelSize parameter can be set in the header.
• Signal component suffixes:
  • Magnitude: No suffix
  • Phase (scaled -π/π): _ph
  • Real: _real
  • Imaginary: _imag
• TODO: Definition of multi-station data (e.g. whole-body acquisitions divided into multiple stacks). Proposal: add a _stack suffix to the filename.

Guidelines for specific scan types

Implemented or partially implemented

Multi-echo gradient echo

• NIfTI structure: 4D (x,y,z,echo)
• Filename suffix: megre
• Folder: anat
• JSON required fields:
  • EchoTime (array) in ms
  • WaterFatShift in pixels
  • MagneticFieldStrength
  • [proposed] ReadoutMode: Monopolar/Bipolar
  • [proposed] PrecessionDirection: Counter-/Clockwise

Multi-echo spin echo

• NIfTI structure: 4D (x,y,z,echo)
• Filename suffix: mese
• Folder: anat
• JSON required fields:
  • EchoTime (array) in ms
  • RefocusingFlipAngle in degrees
  • [optional] ExcitationProfile and RefocusingProfile (arrays) providing the slice profiles in degrees.

T1w / T1w-FS / T2w / T2w-FS

• NIfTI structure: 3D (x,y,z)
• Filename suffix: t1w / t1w-fs / t2w / t2w-fs
• Folder: anat

Quantitative T1 / T2 / wT2

• NIfTI structure: 3D (x,y,z)
• Filename suffix: t1 / t2 / wt2
• Folder: quant

Not implemented / Proposed

Phase contrast

• NIfTI structure: 4D (x,y,z,t) + multiple suffixes for venc direction
• Filename suffix: pc_mag / pc_ph_1 / pc_ph_2 / pc_ph_3
• Folder: flow ??
• The velocity information is stored as phase data scaled between -π and +π.
• JSON required fields:
  • Venc in cm/s
  • EncodingDirection (3D vector) for each phase volume, indicating the direction of the positive velocity encoding for that volume. TBD: patient coordinate system or image coordinate system.

Diffusion

• NIfTI structure: 4D (x,y,z,direction)
• Filename suffix: diff
• Folder: diff ??
• JSON required fields:
  • MixingTime in ms
  • EncodingDirection (array of 3D vectors). The norm of the vector is the b-value. The normalized vector indicates the direction of the diffusion gradient in patient coordinates.

Segmentation labels

• NIfTI structure:
  • 3D (x,y,z) with integer values, from 0 to N, each value corresponding to a different label
  • 4D (x,y,z,label) stack of binary (0/1) values, each dimension corresponding to a label
• Filename suffix: seg ??
• Folder: seg ??
• JSON required fields:
  • Labels (array of strings). List of the labels represented in the masks. The first value in the list corresponds to either a gray level of 0 or to the 1st volume in the fourth dimension. E.g. ["Background", "SOL", "VM", "VL"]
  • Note: the string representation of the labels must follow a standardized format. While it is possible that the same anatomical structure is represented by different labels (e.g. SOL or Soleus), the labels must be known. This allows flexibility in the implementation of segmentation tools, while keeping easy interoperability because all values are easily convertible. A list of standardized labels is visible here. Please contact Francesco Santini if you would like to add your own definitions.