Skip to content

Commit

Permalink
FIX: Remove hardcoded suffix table. (#833)
Browse files Browse the repository at this point in the history
  • Loading branch information
tsalo authored Jul 21, 2021
1 parent 96131df commit f386214
Showing 1 changed file with 0 additions and 12 deletions.
Original file line number Diff line number Diff line change
Expand Up @@ -210,18 +210,6 @@ non-parametric structural MR images include:
)
}}

| **Name** | `suffix` | **Description** |
| -------------------------------------------- | --------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| T1 weighted images | T1w | In arbitrary units (arbitrary). The contrast of these images is mainly determined by spatial variations in the longitudinal relaxation time of the imaged specimen. In spin-echo sequences this contrast is achieved at relatively short repetition and echo times. To achieve this weighting in gradient-echo images, again, short repetition and echo times are selected; however, at relatively large flip angles. Another common approach to increase T1 weighting in gradient-echo images is to add an inversion preparation block to the beginning of the imaging sequence (for example, `TurboFLASH` or `MP-RAGE`). |
| T2 weighted images | T2w | In arbitrary units (arbitrary). The contrast of these images is mainly determined by spatial variations in the (true) transverse relaxation time of the imaged specimen. In spin-echo sequences this contrast is achieved at relatively long repetition and echo times. Generally, gradient echo sequences are not the most suitable option for achieving T2 weighting, as their contrast natively depends on T2-star rather than on T2. |
| Proton density (PD) weighted images | PDw | In arbitrary units (arbitrary). The contrast of these images is mainly determined by spatial variations in the spin density (1H) of the imaged specimen. In spin-echo sequences this contrast is achieved at short repetition and long echo times. In a gradient-echo acquisition, PD weighting dominates the contrast at long repetition and short echo times, and at small flip angles. |
| T2star weighted images | T2starw | In arbitrary units (arbitrary). The contrast of these images is mainly determined by spatial variations in the (observed) transverse relaxation time of the imaged specimen. In spin-echo sequences, this effect is negated as the excitation is followed by an inversion pulse. The contrast of gradient-echo images natively depends on T2-star effects. However, for T2-star variation to dominate the image contrast, gradient-echo acquisitions are carried out at long repetition and echo times, and at small flip angles. |
| Fluid attenuated inversion recovery images | FLAIR | In arbitrary units (arbitrary). Structural images with predominant T2 contribution (also known as T2-FLAIR), in which signal from fluids (for example, CSF) is nulled out by adjusting inversion time, coupled with notably long repetition and echo times. |
| Inplane T1 | inplaneT1 | In arbitrary units (arbitrary). T1 weighted structural image matched to a functional (task) image. |
| Inplane T2 | inplaneT2 | In arbitrary units (arbitrary). T2 weighted structural image matched to a functional (task) image. |
| PD and T2 weighted images | PDT2 | In arbitrary units (arbitrary). PDw and T2w images acquired using a dual echo FSE sequence through view sharing process ([Johnson et al. 1994](https://pubmed.ncbi.nlm.nih.gov/8010268/)). |
| Homogeneous (flat) T1-weighted MP2RAGE image | UNIT1 | In arbitrary units (arbitrary). UNIT1 images are REQUIRED to use this suffix regardless of the method used to generate them. Note that although this image is T1-weighted, regions without MR signal will contain white salt-and-pepper noise that most segmentation algorithms will fail on. Therefore, it is important to dissociate it from from `T1w`. Please see [`MP2RAGE` specific notes](../99-appendices/11-qmri.md#unit1-images) in the qMRI appendix for further information. |

If the structural images included in the dataset were defaced (to protect
identity of participants) one MAY provide the binary mask that was used to
remove facial features in the form of `_defacemask` files.
Expand Down

0 comments on commit f386214

Please sign in to comment.