參考：FSL里的FAST、核磁數據處理: FSL系列之segmentation工具FAST、fslwiki_FAST、FSL-Introduction
相關文檔：FSL Structural Segmentation ☆☆☆☆☆
The various output images are:
- Partial volume maps: A (non-binary) partial volume image for each class, where each voxel contains a value in the range 0-1 that represents the proportion of that class's tissue present in that voxel. This is the default output. 局部體積圖：每個類別的（非二進制）局部體積圖像，其中每個體素包含0-1范圍內的值，該值表示該體素中該類別組織的比例。 這是默認輸出。
# Partial volume maps (Default)
/usr/share/fsl/5.0/bin/fast -t 1 -n 3 -H 0.1 -I 4 -l 20.0 -o /home/captain/Desktop/7_30_fsl/sample-001_brain_fast /home/captain/Desktop/7_30_fsl/sample-001_brain
# output the files as below:
sample-001_brain_fast_mixeltype.nii.gz
sample-001_brain_fast_pve_0.nii.gz
sample-001_brain_fast_pve_1.nii.gz
sample-001_brain_fast_pve_2.nii.gz
sample-001_brain_fast_pveseg.nii.gz
sample-001_brain_fast_seg.nii.gz
- Binary segmentation: single image: This is the "hard" (binary) segmentation, where each voxel is classified into only one class. A single image contains all the necessary information, with the first class taking intensity value 1 in the image, etc. 二進制分割：單個圖像：這是“硬”（二進制）分割，其中每個體素僅分類為一類。 一幅圖像包含所有必要的信息，第一類圖像中的強度值為1，依此類推。
# pure
/usr/share/fsl/5.0/bin/fast -t 1 -n 3 -H 0.1 -I 4 -l 20.0 --nopve
-o /home/captain/Desktop/7_30_fsl/sample-001_brain_fast
/home/captain/Desktop/7_30_fsl/sample-001_brain
# output the files as below:
sample-001_brain_fast_seg.nii.gz
- Binary segmentation: One image per class: This is also a hard segmentation output; the difference is that there is one output image per class, and values are only either 0 or 1. 二進制分割：每類一張圖像：這也是硬分割輸出； 區別在于每個類只有一個輸出圖像，并且值僅為0或1。
# Binary segmentation: Also output one image per class
/usr/share/fsl/5.0/bin/fast -t 1 -n 3 -H 0.1 -I 4 -l 20.0 -g --nopve
-o /home/captain/Desktop/7_30_fsl/sample-001_brain_fast
/home/captain/Desktop/7_30_fsl/sample-001_brain
# output the files as below: (White matter、Gray matter、CSF)
sample-001_brain_fast_seg.nii.gz
sample-001_brain_fast_seg_0.nii.gz
sample-001_brain_fast_seg_1.nii.gz
sample-001_brain_fast_seg_2.nii.gz
- Restored input: This is the estimated restored input image after correction for bias field. 恢復的輸入：這是在對偏差字段進行校正后估計的恢復的輸入圖像。
# Restored input
/usr/share/fsl/5.0/bin/fast -t 1 -n 3 -H 0.1 -I 4 -l 20.0 --nopve -B -o /home/captain/Desktop/7_30_fsl/sample-001_brain_fast /home/captain/Desktop/7_30_fsl/sample-001_brain
# output the files as below:
sample-001_brain_fast_restore.nii.gz
sample-001_brain_fast_seg.nii.gz
- Bias field: This is the estimated bias field. 偏差字段：這是估計的偏差字段。
# Estimated Bias field
/usr/share/fsl/5.0/bin/fast -t 1 -n 3 -H 0.1 -I 4 -l 20.0 --nopve -b -o /home/captain/Desktop/7_30_fsl/sample-001_brain_fast /home/captain/Desktop/7_30_fsl/sample-001_brain
# output the files as below:
sample-001_brain_fast_bias.nii.gz
sample-001_brain_fast_seg.nii.gz
Advanced Options
-
Bias field iterations determines the number of passes made during the initial bias field estimation stage. A greater number of iterations can help esitmate particularly strong bias fields. 偏置場迭代確定了初始偏置場估計階段的通過次數。
大量的迭代可以幫助確定特別強的偏置場。 -
Bias field smoothing controls the amount of smoothness expected in the estimated bias field. The value entered is the Full-Width Half-Maximum (FWHM) in mm. A larger value here will impose more smoothness on the estimated bias field. 偏置場平滑控制估計的偏置場中
期望的平滑度。輸入的值為以毫米為單位的全寬半最大值（FWHM）。此處較大的值將對估計的偏置場施加更大的平滑度。 -
Use a-priori probability maps tells FAST to start by registering the input image to standard space and then use standard tissue-type probability maps (from the MNI152 dataset) instead of the initial K-means segmentation, in order to estimate the initial parameters of the classes. This can help in cases where there is very bad bias field. By default the a-priori probability maps are only used to initialise the segmentation; however, you can also optionally tell FAST to use these priors in the final segmentation - this can help, for example, with the segmentation of deep grey structures. 使用
先驗概率圖告訴FAST首先將輸入圖像注冊到標準空間，然后使用標準組織類型概率圖（來自MNI152數據集）代替初始K均值分割，以便估算類。這在偏置場非常差的情況下可以提供幫助。默認情況下，先驗概率圖僅用于初始化分割。但是，您也可以選擇告訴FAST在最終的細分中使用這些先驗-例如，這可以幫助細分深灰色結構。 -
Use file of initial tissue-type means tells FAST to use a text file with mean intensity values (separated by newlines) for the starting mean values of the different classes to be segmented. This is then used instead of the automated K-means starting parameter estimation. 初始組織類型均值的使用文件告訴FAST使用具有平均強度值（用換行符分隔）的
文本文件作為要分割的不同類別的起始均值。然后使用它代替自動K均值開始參數估計。
# Partial volume maps (Default)
/usr/share/fsl/5.0/bin/fast -t 1 -n 3 -H 0.1 -I 4 -l 20.0 -o
/home/captain/Desktop/7_30_fsl/sample-001_brain_fast
/home/captain/Desktop/7_30_fsl/sample-001_brain
# output the files as below:
sample-001_brain_fast_mixeltype.nii.gz
sample-001_brain_fast_pve_0.nii.gz
sample-001_brain_fast_pve_1.nii.gz
sample-001_brain_fast_pve_2.nii.gz
sample-001_brain_fast_pveseg.nii.gz
sample-001_brain_fast_seg.nii.gz
常用參數:
-S or --channels= 圖像通道數, 通常設為1
-t or --type= 圖像種類, T1的話設置為1, T2設置為2, PD設置為3 Type of image (e.g. T1, T2, PD) - Note space in fast4 version e.g. t 2 not t2 → Fast3
-o or --out= 輸出文件的基本名稱,生成的文件會自動附上后綴, 例如seg, pve_0, pve_1等.
-n or --class= 組織類型數.T1的話通常設置為3, (Grey Matter, White Matter and CSF) T2的話通常設置為4. 如果圖像很不清晰, 可以設置為2.
-b 輸出的估計偏置場(estimated bias field)
-B 輸出偏置圖像并保存(bias-corrected image)
-l or –lowpass= 偏置場平滑, m是FWHM, 單位是mm, m ? 2 √n
-I 或者 –iter 循環次數
- ADNI_023_S_1190_FDG.nii（左）原始數據
- ADNI_023_S_1190_FDG_brain.nii （右）去除顱骨
- ADNI_023_S_1190_FDG_brain_pve_0.nii（左）腦脊液
- ADNI_023_S_1190_FDG_brain_pve_1.nii（中）灰質
- ADNI_023_S_1190_FDG_brain_pve_2.nii（右）白質
注：partial volume (PVE)
- ADNI_023_S_1190_FDG_brain_mixeltype.nii （左）
- ADNI_023_S_1190_FDG_brain_pveseg.nii（中）
- ADNI_023_S_1190_FDG_brain_seg.nii（右）（= *_pveseg.nii）
ADNI_023_S_1190_FDG_brain_pveseg.nii：（Advanced Options）
-
noUse a-priori probability maps for initiallsation;noand for final segmentation (左) -
√Use a-priori probability maps for initiallsation;noand for final segmentation (中) -
√Use a-priori probability maps for initiallsation;√and for final segmentation (右)
ADNI_023_S_1190_FDG_brain_pveseg.nii：（Advanced Options）
- Bias field smoothing (FWHM in mm) ：20.0 （左） 默認值，較優
- Bias field smoothing (FWHM in mm) ：8.0 （右）
