pwighton/miba

Steps to create a MIBA

Using the following datasets:

ds004230
cox1blocked

1) Run `petprep_hmc`

Run petprep_hmc on the datasets to generate motion corrected pet timeseries.

See the repo for details.
Also see this issue documenting container running notes.
The container ID used was 384176a6cc6b (martinnoergaard/petprep_hmc:latest as of 20231005)

On ds004230:

docker run -it --rm \
    -v /home/paul/lcn/20230918-bloodstream-r/ds004230-plus-cox1blocked/ds004230:/data/input \
    -v /home/paul/lcn/20230918-bloodstream-r/ds004230-plus-cox1blocked/petprep-hmc:/data/output \
    -v /home/paul/lcn/license.txt:/opt/freesurfer/license.txt \
    -e SUBJECTS_DIR=/data/output \
    -e PATH=/opt/freesurfer/bin:/opt/freesurfer/fsfast/bin:/opt/freesurfer/tktools:/opt/fsl/bin:/opt/freesurfer/mni/bin:/opt/freesurfer/bin:/opt/fsl/bin:/usr/local/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/local/fsl/bin \
    martinnoergaard/petprep_hmc:latest \
        --bids_dir /data/input \
        --output_dir /data/output \
        --n_procs 12

On cox1blocked:

docker run -it --rm \
    -v /home/paul/lcn/20230918-bloodstream-r/ds004230-plus-cox1blocked/cox1blocked:/data/input \
    -v /home/paul/lcn/20230918-bloodstream-r/ds004230-plus-cox1blocked/petprep-hmc2:/data/output \
    -v /home/paul/lcn/license.txt:/opt/freesurfer/license.txt \
    -e SUBJECTS_DIR=/data/output \
    -e PATH=/opt/freesurfer/bin:/opt/freesurfer/fsfast/bin:/opt/freesurfer/tktools:/opt/fsl/bin:/opt/freesurfer/mni/bin:/opt/freesurfer/bin:/opt/fsl/bin:/usr/local/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/local/fsl/bin \
    martinnoergaard/petprep_hmc:latest \
        --bids_dir /data/input \
        --output_dir /data/output \
        --n_procs 12 \
        --skip_bids_validator

2) Merge datasets

Merge ds004230 and cox1blocked into a single dataset.

We do this after petprep_hmc since the inconsistent ses naming (rescan vs blocked) interferes with petprep_hmc
We do this before bloodstream since we're using hierarchical modeling in #3 which implies the processsing is not independent (pw: ? verify)

The following subjects were used going forward:

sub-PS19
sub-PS20
sub-PS21
sub-PS24
sub-PS26
sub-PS27
sub-PS28
sub-PS38
sub-PS39
sub-PS42
sub-PS50
sub-PS51
sub-PS52
sub-PS53
sub-PS54
sub-PS55
sub-PS56

3) Run `bloodstream`

Run bloodstream to model and upsample artierial input functions (AIF)

See the repo for details and this PR for functionality to run non-interactivley in a container
The config file was config_2023-08-28_id-tfK8--martin.json.
The container ID used was 796cd379c2c0 (pwighton/bloodstream:20230929 as of 20230929)

docker run -it --rm \
  -v ${HOME}:/home/jovyan/work \
  pwighton/bloodstream:20230929 \
    /home/jovyan/run-bloodstream.R \
      -s /home/jovyan/work/lcn/20230918-bloodstream-r/ds004230-plus-cox1blocked \
      -c /home/jovyan/work/lcn/20230918-bloodstream-r/config_2023-08-28_id-tfK8--martin.json

4) Run `FreeSurfer 7.4.1`

For each subject, run:

recon-all -all $SUB
gtmseg -s $SUB
mni152reg --s $SUB

Place the global-expert-options.txt file in $SUBJECTS_DIR before running
The script mlsc-recon-all.bash can be used to submit these processing jobs to a SLURM compute cluster, which reads the file ds004230-plus-cox1blocked-subject-list.txt to determine what to process.

5) Run `cox1-preproc.bash`

Edit cox1-preproc.bash and set:

MAP_FILE
IMG_REF_DIR
BLOOD_REF_DIR
CALC_FRAMEWISE_AIF_PY

Run cox1-preproc.bash which:

Takes as input a mapping file (cox1-preproc-mapping.txt) containing
- fs subjects
- pet study dirs
- pet .json sidecar file
- motion corrected pet imaging file
- bloodstream file
And:
- Creates the file $SUBJECTS_DIR/tsec.txt
  - Currently assumes frame timing constant across study
- For each line in provenance file:
  - Creates the folder ${SUBJECTS_DIR}/${FS+SUBJECT}/${PET_STUDY}
  - Copies ${PET_IMAGE} to ${SUBJECTS_DIR}/${FS+SUBJECT}/${PET_STUDY}/pet.nii.gz
  - Creates a mean image across time from ${SUBJECTS_DIR}/${FS+SUBJECT}/${PET_STUDY}/pet.nii.gz
  - Creates mean AIF per PET frame (${SUBJECTS_DIR}/${FS_SUBJECT}/${PET_STUDY}/aif.bloodstream.dat) by running calc_framewise_aif.py on the bloodstream file

5) Run `cox1-proc`

Run cox1-proc on each pet session, which:

Registers pet.mn.nii.gz to the anatomical using mri_coreg
Maps the subject's aparc+aseg to PET space
Creates a brainmask in PET space
Runs mri_gtmpvc
Runs mri_glmfit to peform an Logan analysis
Maps the results ("Distribution Volume Ratio"; DVR) to fsaverage space

6) Compute Vnd via Lassen plot

Vnd is the x-intercept of the lassen plot
regions to use/merge is specified in occupancy-merge.json

export SUBJECTS_DIR=/home/paul/lcn/20230918-bloodstream-r/fs-subs
export MERGE_DATA=/home/paul/lcn/git/miba/occupancy-merge.json
export CALC_VND=/home/paul/lcn/git/miba/calc_vnd.py

for SUBJECT in sub-PS50 sub-PS51 sub-PS52 sub-PS53 sub-PS54 sub-PS55 sub-PS56
do
    $CALC_VND \
      -m $MERGE_DATA \
      --seg-base $SUBJECTS_DIR/$SUBJECT/pet1/apas.nii.gz \
      --dvr-base $SUBJECTS_DIR/$SUBJECT/pet1/glmfit.logan/dvr/gamma.nii.gz \
      --seg-block $SUBJECTS_DIR/$SUBJECT/pet2/apas.nii.gz \
      --dvr-block $SUBJECTS_DIR/$SUBJECT/pet2/glmfit.logan/dvr/gamma.nii.gz \
      -o $SUBJECTS_DIR/$SUBJECT/pet-lassen.json \
      -ofig $SUBJECTS_DIR/$SUBJECT/pet-lassen.png
done

This generates the following values:

Subject	Vnd	Occupancy
sub-PS50	1.1319486599814665	0.7869641734671312
sub-PS51	1.1192285589034248	0.7959955315691787
sub-PS52	1.0496236020838752	0.7542152189502245
sub-PS53	1.153529523200513	0.8902525060612867
sub-PS54	1.5011441337708775	0.9089416391564102
sub-PS55	0.9736405141537436	0.6999053740530418
sub-PS56	1.2181914901342907	0.7805440613678035

Values from Nafiseh:

Subject	Vnd	Occupancy
sub-PS50	1.446	0.8343
sub-PS51	1.349	0.8216
sub-PS52	1.197	0.8483
sub-PS53	1.112	0.9189
sub-PS54	2.326	0.988
sub-PS55	1.139	0.7026
sub-PS56	2.379	0.8267
sub-PS57	2.429	0.7932

7) Run `miba_gen.py`

To generate the average across the first session of all subjects

In volumetric space:

export SUBJECTS_DIR=/home/paul/lcn/20230918-bloodstream-r/fs-subs

/home/paul/lcn/git/miba/miba_gen.py \
  --mean \
  -i ${SUBJECTS_DIR}/sub-PS19/pet1/glmfit.logan/dvr/dvr.fsaverage.nii.gz \
     ${SUBJECTS_DIR}/sub-PS20/pet1/glmfit.logan/dvr/dvr.fsaverage.nii.gz \
     ${SUBJECTS_DIR}/sub-PS21/pet1/glmfit.logan/dvr/dvr.fsaverage.nii.gz \
     ${SUBJECTS_DIR}/sub-PS24/pet1/glmfit.logan/dvr/dvr.fsaverage.nii.gz \
     ${SUBJECTS_DIR}/sub-PS26/pet1/glmfit.logan/dvr/dvr.fsaverage.nii.gz \
     ${SUBJECTS_DIR}/sub-PS27/pet1/glmfit.logan/dvr/dvr.fsaverage.nii.gz \
     ${SUBJECTS_DIR}/sub-PS28/pet1/glmfit.logan/dvr/dvr.fsaverage.nii.gz \
     ${SUBJECTS_DIR}/sub-PS38/pet1/glmfit.logan/dvr/dvr.fsaverage.nii.gz \
     ${SUBJECTS_DIR}/sub-PS39/pet1/glmfit.logan/dvr/dvr.fsaverage.nii.gz \
     ${SUBJECTS_DIR}/sub-PS42/pet1/glmfit.logan/dvr/dvr.fsaverage.nii.gz \
     ${SUBJECTS_DIR}/sub-PS50/pet1/glmfit.logan/dvr/dvr.fsaverage.nii.gz \
     ${SUBJECTS_DIR}/sub-PS51/pet1/glmfit.logan/dvr/dvr.fsaverage.nii.gz \
     ${SUBJECTS_DIR}/sub-PS52/pet1/glmfit.logan/dvr/dvr.fsaverage.nii.gz \
     ${SUBJECTS_DIR}/sub-PS53/pet1/glmfit.logan/dvr/dvr.fsaverage.nii.gz \
     ${SUBJECTS_DIR}/sub-PS54/pet1/glmfit.logan/dvr/dvr.fsaverage.nii.gz \
     ${SUBJECTS_DIR}/sub-PS55/pet1/glmfit.logan/dvr/dvr.fsaverage.nii.gz \
     ${SUBJECTS_DIR}/sub-PS56/pet1/glmfit.logan/dvr/dvr.fsaverage.nii.gz \
   -o cox1.miba.vt.fsaverage.nii.gz

On the surface of the left hemisphere

export SUBJECTS_DIR=/home/paul/lcn/20230918-bloodstream-r/fs-subs

/home/paul/lcn/git/miba/miba_gen.py \
  --mean \
  -i ${SUBJECTS_DIR}/sub-PS19/pet1/glmfit.logan/dvr/dvr.fsaverage.lh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS20/pet1/glmfit.logan/dvr/dvr.fsaverage.lh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS21/pet1/glmfit.logan/dvr/dvr.fsaverage.lh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS24/pet1/glmfit.logan/dvr/dvr.fsaverage.lh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS26/pet1/glmfit.logan/dvr/dvr.fsaverage.lh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS27/pet1/glmfit.logan/dvr/dvr.fsaverage.lh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS28/pet1/glmfit.logan/dvr/dvr.fsaverage.lh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS38/pet1/glmfit.logan/dvr/dvr.fsaverage.lh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS39/pet1/glmfit.logan/dvr/dvr.fsaverage.lh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS42/pet1/glmfit.logan/dvr/dvr.fsaverage.lh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS50/pet1/glmfit.logan/dvr/dvr.fsaverage.lh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS51/pet1/glmfit.logan/dvr/dvr.fsaverage.lh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS52/pet1/glmfit.logan/dvr/dvr.fsaverage.lh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS53/pet1/glmfit.logan/dvr/dvr.fsaverage.lh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS54/pet1/glmfit.logan/dvr/dvr.fsaverage.lh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS55/pet1/glmfit.logan/dvr/dvr.fsaverage.lh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS56/pet1/glmfit.logan/dvr/dvr.fsaverage.lh.nii.gz \
   -o cox1.miba.vt.fsaverage.lh.nii.gz

On the surface of the right hemisphere

cd $SUBJECTS_DIR
/home/paul/lcn/git/miba/miba_gen.py \
  --mean \
  -i ${SUBJECTS_DIR}/sub-PS19/pet1/glmfit.logan/dvr/dvr.fsaverage.rh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS20/pet1/glmfit.logan/dvr/dvr.fsaverage.rh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS21/pet1/glmfit.logan/dvr/dvr.fsaverage.rh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS24/pet1/glmfit.logan/dvr/dvr.fsaverage.rh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS26/pet1/glmfit.logan/dvr/dvr.fsaverage.rh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS27/pet1/glmfit.logan/dvr/dvr.fsaverage.rh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS28/pet1/glmfit.logan/dvr/dvr.fsaverage.rh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS38/pet1/glmfit.logan/dvr/dvr.fsaverage.rh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS39/pet1/glmfit.logan/dvr/dvr.fsaverage.rh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS42/pet1/glmfit.logan/dvr/dvr.fsaverage.rh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS50/pet1/glmfit.logan/dvr/dvr.fsaverage.rh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS51/pet1/glmfit.logan/dvr/dvr.fsaverage.rh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS52/pet1/glmfit.logan/dvr/dvr.fsaverage.rh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS53/pet1/glmfit.logan/dvr/dvr.fsaverage.rh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS54/pet1/glmfit.logan/dvr/dvr.fsaverage.rh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS55/pet1/glmfit.logan/dvr/dvr.fsaverage.rh.nii.gz \
     ${SUBJECTS_DIR}/sub-PS56/pet1/glmfit.logan/dvr/dvr.fsaverage.rh.nii.gz \
   -o cox1.miba .vt.fsaverage.rh.nii.gz

To generate Vs (Vt - Vnd) for each baseline/blocked subject

First generate Vs (Vt - Vnd) for each subject:

export SUBJECTS_DIR=/home/paul/lcn/20230918-bloodstream-r/fs-subs
SUB_LIST=(sub-PS50 sub-PS51 sub-PS52 sub-PS53 sub-PS54 sub-PS55 sub-PS56)
VND_LIST=(1.446 1.349 1.197 1.112 2.326 1.139 2.379)

for i in "${!SUB_LIST[@]}"
do
  SUB=${SUB_LIST[i]}
  VND=${VND_LIST[i]}
  echo $SUB $VND
  /home/paul/lcn/git/miba/miba_gen.py \
    --diff-const \
    -i ${SUBJECTS_DIR}/${SUB}/pet1/glmfit.logan/dvr/dvr.fsaverage.nii.gz \
    -cv ${VND} \
    -o cox1.${SUB}.vnd.fsaverage.nii.gz
  /home/paul/lcn/git/miba/miba_gen.py \
    --diff-const \
    -i ${SUBJECTS_DIR}/${SUB}/pet1/glmfit.logan/dvr/dvr.fsaverage.lh.nii.gz \
    -cv ${VND} \
    -o cox1.${SUB}.vnd.fsaverage.lh.nii.gz
  /home/paul/lcn/git/miba/miba_gen.py \
    --diff-const \
    -i ${SUBJECTS_DIR}/${SUB}/pet1/glmfit.logan/dvr/dvr.fsaverage.rh.nii.gz \
    -cv ${VND} \
    -o cox1.${SUB}.vnd.fsaverage.rh.nii.gz
done

Then average:

/home/paul/lcn/git/miba/miba_gen.py \
  --mean \
  -i cox1.sub-PS50.vnd.fsaverage.nii.gz \
     cox1.sub-PS51.vnd.fsaverage.nii.gz \
     cox1.sub-PS52.vnd.fsaverage.nii.gz \
     cox1.sub-PS53.vnd.fsaverage.nii.gz \
     cox1.sub-PS54.vnd.fsaverage.nii.gz \
     cox1.sub-PS55.vnd.fsaverage.nii.gz \
     cox1.sub-PS56.vnd.fsaverage.nii.gz \
  -o cox1.miba.vnd.fsaverage.nii.gz

/home/paul/lcn/git/miba/miba_gen.py \
  --mean \
  -i cox1.sub-PS50.vnd.fsaverage.rh.nii.gz \
     cox1.sub-PS51.vnd.fsaverage.rh.nii.gz \
     cox1.sub-PS52.vnd.fsaverage.rh.nii.gz \
     cox1.sub-PS53.vnd.fsaverage.rh.nii.gz \
     cox1.sub-PS54.vnd.fsaverage.rh.nii.gz \
     cox1.sub-PS55.vnd.fsaverage.rh.nii.gz \
     cox1.sub-PS56.vnd.fsaverage.rh.nii.gz \
  -o cox1.miba.vnd.fsaverage.rh.nii.gz

/home/paul/lcn/git/miba/miba_gen.py \
  --mean \
  -i cox1.sub-PS50.vnd.fsaverage.lh.nii.gz \
     cox1.sub-PS51.vnd.fsaverage.lh.nii.gz \
     cox1.sub-PS52.vnd.fsaverage.lh.nii.gz \
     cox1.sub-PS53.vnd.fsaverage.lh.nii.gz \
     cox1.sub-PS54.vnd.fsaverage.lh.nii.gz \
     cox1.sub-PS55.vnd.fsaverage.lh.nii.gz \
     cox1.sub-PS56.vnd.fsaverage.lh.nii.gz \
  -o cox1.miba.vnd.fsaverage.lh.nii.gz

To generate BPND (Vs / Vnd) for each baseline/blocked subject

SUB_LIST=(sub-PS50 sub-PS51 sub-PS52 sub-PS53 sub-PS54 sub-PS55 sub-PS56)
VND_LIST=(1.446 1.349 1.197 1.112 2.326 1.139 2.379)

for i in "${!SUB_LIST[@]}"
do
  SUB=${SUB_LIST[i]}
  VND=${VND_LIST[i]}
  echo $SUB $VND
  /home/paul/lcn/git/miba/miba_gen.py \
    --div-const \
    -i cox1.${SUB}.vnd.fsaverage.nii.gz \
    -cv ${VND} \
    -o cox1.${SUB}.bpnd.fsaverage.nii.gz
  /home/paul/lcn/git/miba/miba_gen.py \
    --div-const \
    -i cox1.${SUB}.vnd.fsaverage.lh.nii.gz \
    -cv ${VND} \
    -o cox1.${SUB}.bpnd.fsaverage.lh.nii.gz
  /home/paul/lcn/git/miba/miba_gen.py \
    --div-const \
    -i cox1.${SUB}.vnd.fsaverage.rh.nii.gz \
    -cv ${VND} \
    -o cox1.${SUB}.bpnd.fsaverage.rh.nii.gz
done

Then average:

/home/paul/lcn/git/miba/miba_gen.py \
  --mean \
  -i cox1.sub-PS50.bpnd.fsaverage.nii.gz \
     cox1.sub-PS51.bpnd.fsaverage.nii.gz \
     cox1.sub-PS52.bpnd.fsaverage.nii.gz \
     cox1.sub-PS53.bpnd.fsaverage.nii.gz \
     cox1.sub-PS54.bpnd.fsaverage.nii.gz \
     cox1.sub-PS55.bpnd.fsaverage.nii.gz \
     cox1.sub-PS56.bpnd.fsaverage.nii.gz \
  -o cox1.miba.bpnd.fsaverage.nii.gz

/home/paul/lcn/git/miba/miba_gen.py \
  --mean \
  -i cox1.sub-PS50.bpnd.fsaverage.rh.nii.gz \
     cox1.sub-PS51.bpnd.fsaverage.rh.nii.gz \
     cox1.sub-PS52.bpnd.fsaverage.rh.nii.gz \
     cox1.sub-PS53.bpnd.fsaverage.rh.nii.gz \
     cox1.sub-PS54.bpnd.fsaverage.rh.nii.gz \
     cox1.sub-PS55.bpnd.fsaverage.rh.nii.gz \
     cox1.sub-PS56.bpnd.fsaverage.rh.nii.gz \
  -o cox1.miba.bpnd.fsaverage.rh.nii.gz

/home/paul/lcn/git/miba/miba_gen.py \
  --mean \
  -i cox1.sub-PS50.bpnd.fsaverage.lh.nii.gz \
     cox1.sub-PS51.bpnd.fsaverage.lh.nii.gz \
     cox1.sub-PS52.bpnd.fsaverage.lh.nii.gz \
     cox1.sub-PS53.bpnd.fsaverage.lh.nii.gz \
     cox1.sub-PS54.bpnd.fsaverage.lh.nii.gz \
     cox1.sub-PS55.bpnd.fsaverage.lh.nii.gz \
     cox1.sub-PS56.bpnd.fsaverage.lh.nii.gz \
  -o cox1.miba.bpnd.fsaverage.lh.nii.gz

8) Generate figures

Figures of the average across the first session of all subjects

freeview \
  --volume \
    ${FREESURFER_HOME}/subjects/fsaverage/mri/orig.mgz \
    cox1.vt.fsaverage.nii.gz:colormap=heat:opacity=0.5:heatscale=1.8,2.3,2.8:heatscale_options=truncate \
  --ras 0 -10 0 \
  --cc \
  --viewport axial \
  --colorscale \
  --screenshot ./cox1.miba.vt.axial.mean.png 2 true

freeview \
  --volume \
    ${FREESURFER_HOME}/subjects/fsaverage/mri/orig.mgz \
    cox1.vt.fsaverage.nii.gz:colormap=heat:opacity=0.5:heatscale=1.8,2.3,2.8:heatscale_options=truncate \
  --ras 0 -10 0 \
  --cc \
  --viewport sagittal \
  --colorscale \
  --screenshot ./cox1.miba.vt.sagittal.mean.png 2 true

freeview \
  --surface \
    ${FREESURFER_HOME}/subjects/fsaverage/surf/lh.white:overlay=cox1.vt.fsaverage.lh.nii.gz:overlay_threshold=1.80,2.80:overlay_color=heat,truncate:annot=${FREESURFER_HOME}/subjects/fsaverage/label/lh.aparc.annot:annot_outline=1 \
  --viewport 3d \
  --colorscale \
  --screenshot ./cox1.miba.vt.lh.mean.white.png 2 true

freeview \
  --surface \
    ${FREESURFER_HOME}/subjects/fsaverage/surf/lh.inflated:overlay=cox1.vt.fsaverage.lh.nii.gz:overlay_threshold=1.80,2.80:overlay_color=heat,truncate:annot=${FREESURFER_HOME}/subjects/fsaverage/label/lh.aparc.annot:annot_outline=1 \
  --viewport 3d \
  --colorscale \
  --screenshot ./cox1.miba.vt.lh.mean.inflated.png 2 true

Figures of the average Vs across unblocked/blocked subjects

freeview \
  --volume \
    ${FREESURFER_HOME}/subjects/fsaverage/mri/orig.mgz \
    cox1.miba.vnd.fsaverage.nii.gz:colormap=heat:opacity=0.5:heatscale=0.0,1.9,3.8:heatscale_options=truncate \
  --ras 0 -10 0 \
  --cc \
  --viewport axial \
  --colorscale \
  --screenshot ./cox1.miba.vs.axial.meanpng 2 true

freeview \
  --volume \
    ${FREESURFER_HOME}/subjects/fsaverage/mri/orig.mgz \
    cox1.miba.vnd.fsaverage.nii.gz:colormap=heat:opacity=0.5:heatscale=0.0,1.9,3.8:heatscale_options=truncate \
  --ras 0 -10 0 \
  --cc \
  --viewport sagittal \
  --colorscale \
  --screenshot ./cox1.miba.vs.sagittal.mean.png 2 true

freeview \
  --surface \
    ${FREESURFER_HOME}/subjects/fsaverage/surf/lh.white:overlay=cox1.miba.vnd.fsaverage.lh.nii.gz:overlay_threshold=0.00,2.80:overlay_color=heat,truncate:annot=${FREESURFER_HOME}/subjects/fsaverage/label/lh.aparc.annot:annot_outline=1 \
  --viewport 3d \
  --colorscale \
  --screenshot ./cox1.miba.vs.lh.white.png 2 true

freeview \
  --surface \
    ${FREESURFER_HOME}/subjects/fsaverage/surf/lh.inflated:overlay=cox1.miba.vnd.fsaverage.lh.nii.gz:overlay_threshold=0.00,2.80:overlay_color=heat,truncate:annot=${FREESURFER_HOME}/subjects/fsaverage/label/lh.aparc.annot:annot_outline=1 \
  --viewport 3d \
  --colorscale \
  --screenshot ./cox1.miba.vs.lh.inflated.png 2 true

Figures of the average BPND across unblocked/blocked subjects

freeview \
  --volume \
    ${FREESURFER_HOME}/subjects/fsaverage/mri/orig.mgz:grayscale=30,150 \
    cox1.miba.bpnd.fsaverage.nii.gz:colormap=nih:colorscale=0.0,4.0 \
  --ras 0 -10 0 \
  --cc \
  --viewport axial \
  --colorscale \
  --screenshot ./cox1.miba.bpnd.axial.mean.png 2 true

freeview \
  --volume \
    ${FREESURFER_HOME}/subjects/fsaverage/mri/orig.mgz:grayscale=30,150 \
    cox1.miba.vnd.fsaverage.nii.gz:colormap=heat:opacity=0.5:heatscale=0.0,1.3,2.6:heatscale_options=truncate \
  --ras 0 -10 0 \
  --cc \
  --viewport sagittal \
  --colorscale \
  --screenshot ./cox1.miba.bpnd.sagittal.mean.png 2 true

freeview \
  --surface \
    ${FREESURFER_HOME}/subjects/fsaverage/surf/lh.white:overlay=cox1.miba.bpnd.fsaverage.lh.nii.gz:overlay_threshold=0.50,1.50:overlay_color=heat,truncate:annot=${FREESURFER_HOME}/subjects/fsaverage/label/lh.aparc.annot:annot_outline=1 \
  --viewport 3d \
  --colorscale \
  --screenshot ./cox1.miba.bpnd.lh.white.png 2 true

freeview \
  --surface \
    ${FREESURFER_HOME}/subjects/fsaverage/surf/lh.inflated:overlay=cox1.miba.bpnd.fsaverage.lh.nii.gz:overlay_threshold=0.50,1.50:overlay_color=heat,truncate:annot=${FREESURFER_HOME}/subjects/fsaverage/label/lh.aparc.annot:annot_outline=1 \
  --viewport 3d \
  --colorscale \
  --screenshot ./cox1.miba.bpnd.lh.inflated.png 2 true

References

OHBM 2023 abstract: https://ww6.aievolution.com/hbm2301/index.cfm?do=abs.viewAbs&abs=2785
PetSurfer: https://surfer.nmr.mgh.harvard.edu/fswiki/PetSurfer
Bloodstream walkthrough: https://www.youtube.com/watch?v=Kud6MWYPKxg
Logan method: https://www.pmod.com/files/download/v31/doc/pkin/2329.htm

pwighton / miba

Steps to create a MIBA

1) Run `petprep_hmc`

2) Merge datasets

3) Run `bloodstream`

4) Run `FreeSurfer 7.4.1`

5) Run `cox1-preproc.bash`

5) Run `cox1-proc`

6) Compute Vnd via Lassen plot

7) Run `miba_gen.py`

To generate the average across the first session of all subjects

To generate Vs (Vt - Vnd) for each baseline/blocked subject

To generate BPND (Vs / Vnd) for each baseline/blocked subject

8) Generate figures

Figures of the average across the first session of all subjects

Figures of the average Vs across unblocked/blocked subjects

Figures of the average BPND across unblocked/blocked subjects

References

About

Languages

Steps to create a MIBA

1) Run petprep_hmc

2) Merge datasets

3) Run bloodstream

4) Run FreeSurfer 7.4.1

5) Run cox1-preproc.bash

5) Run cox1-proc

6) Compute Vnd via Lassen plot

7) Run miba_gen.py

To generate the average across the first session of all subjects

To generate Vs (Vt - Vnd) for each baseline/blocked subject

To generate BPND (Vs / Vnd) for each baseline/blocked subject

8) Generate figures

Figures of the average across the first session of all subjects

Figures of the average Vs across unblocked/blocked subjects

Figures of the average BPND across unblocked/blocked subjects

References

About

Languages

1) Run `petprep_hmc`

3) Run `bloodstream`

4) Run `FreeSurfer 7.4.1`

5) Run `cox1-preproc.bash`

5) Run `cox1-proc`

7) Run `miba_gen.py`