diff --git a/README.md b/README.md index 7c81203..b9b61f0 100644 --- a/README.md +++ b/README.md @@ -3,7 +3,7 @@

aide-svrtk

- AI-driven, automated SVR reconstruction packaged as an AIDE Application, based on + AI-driven, automated version of SVR reconstruction packaged as an AIDE Application, based on the open-source MONAI Application Package (MAP) standard.
@@ -20,14 +20,14 @@ The slice-to-volume reconstruction toolkit ([SVRTK](https://github.com/SVRTK/SVRTK)) is an image-based registration framework for reconstruction of 3D volumes from multiple 2D image slices. SVRTK is used extensively for performing -motion-corrected reconstruction of MRI data. +motion-corrected reconstruction of MRI data. The [automated SVR](https://github.com/SVRTK/auto-proc-svrtk) version employs deep learning [MONAI](https://github.com/Project-MONAI/MONAI) networks for localisation and reorientation of the brain to the standard radiological space. Currently, this MAP implements SVRTK for reconstruction of 2D **fetal brain MRI** data into motion-corrected 3D volumes. Further MAPs for reconstruction of other fetal organs are in development and coming soon. [AIDE](https://www.aicentre.co.uk/platforms#view1) is an open-source platform for the deployment of AI applications in -healthcare settings. This repo, created and led by [Tom Roberts](https://github.com/tomaroberts), packages automated -3D fetal brain MRI reconstruction into a [MONAI Application Package (MAP)](https://github.com/Project-MONAI/monai-deploy) +healthcare settings. This repo, created and led by [Tom Roberts](https://github.com/tomaroberts), packages [automated +3D fetal brain MRI reconstruction](https://github.com/SVRTK/auto-proc-svrtk) into a [MONAI Application Package (MAP)](https://github.com/Project-MONAI/monai-deploy) for execution on AIDE, and other platforms compatible with the MAP standard. ## aide-svrtk MAP workflow @@ -38,7 +38,7 @@ MRI DICOM Series contained the motion-corrected reconstruction. The aide-svrtk MAP consists of three operators: 1. `dcm2nii_operator.py` – converts the input 2D MRI DICOM Series into NIfTI format required by SVRTK, using [dcm2niix](https://github.com/rordenlab/dcm2niix) -2. `fetal_mri_3d_brain_recon_operator.py` – runs automated fetal brain SVRTK reconstruction, in three main steps: +2. `fetal_mri_3d_brain_recon_operator.py` – runs [automated fetal brain SVRTK reconstruction](https://github.com/SVRTK/auto-proc-svrtk), in three main steps: a. AI-driven brain masking b. Slice-to-volume registration reconstruction c. AI-driven brain reorientation @@ -98,17 +98,17 @@ The final MAP is called `map` ```shell # Initial packaging of MAP -monai-deploy package app -t ghcr.io/svrtk/aide-svrtk/map-init:0.2.0 -r requirements.txt -l DEBUG +monai-deploy package app -t ghcr.io/svrtk/aide-svrtk/map-init:0.2.1 -r requirements.txt -l DEBUG # Build 3rd-party software on top of MAP -docker build -t ghcr.io/svrtk/aide-svrtk/map:0.2.0 app/ +docker build -t ghcr.io/svrtk/aide-svrtk/map:0.2.1 app/ # Test MAP with MONAI Deploy -monai-deploy run ghcr.io/svrtk/aide-svrtk/map:0.2.0 input/ output/ +monai-deploy run ghcr.io/svrtk/aide-svrtk/map:0.2.1 input/ output/ # Push initial MAP and final MAP to GHCR -docker push ghcr.io/svrtk/aide-svrtk/map-init:0.2.0 -docker push ghcr.io/svrtk/aide-svrtk/map:0.2.0 +docker push ghcr.io/svrtk/aide-svrtk/map-init:0.2.1 +docker push ghcr.io/svrtk/aide-svrtk/map:0.2.1 ``` ## Optional @@ -116,13 +116,13 @@ docker push ghcr.io/svrtk/aide-svrtk/map:0.2.0 Enter Docker container for testing ```shell -docker run --gpus all -it --rm -v local/path/to/aide-svrtk/input:/var/monai/input/ --entrypoint /bin/bash ghcr.io/svrtk/aide-svrtk/map:0.2.0 +docker run --gpus all -it --rm -v local/path/to/aide-svrtk/input:/var/monai/input/ --entrypoint /bin/bash ghcr.io/svrtk/aide-svrtk/map:0.2.1 ``` Run on specified GPU if machine has >1 available ```shell -CUDA_VISIBLE_DEVICES=2 monai-deploy run ghcr.io/svrtk/aide-svrtk/map:0.2.0 input/ output/ +CUDA_VISIBLE_DEVICES=2 monai-deploy run ghcr.io/svrtk/aide-svrtk/map:0.2.1 input/ output/ ``` ## Running on AIDE diff --git a/app/Dockerfile b/app/Dockerfile index b5be450..2127892 100644 --- a/app/Dockerfile +++ b/app/Dockerfile @@ -1,4 +1,4 @@ -FROM ghcr.io/svrtk/aide-svrtk/map-init:0.2.0 AS build +FROM ghcr.io/svrtk/aide-svrtk/map-init:0.2.1 AS build # Add dcm2niix to MAP WORKDIR /bin @@ -19,7 +19,7 @@ WORKDIR /var/monai # Add SVRTK to MAP -WORKDIR /home +WORKDIR /bin # Install general libraries RUN apt-get update && apt-get install --no-install-recommends --no-install-suggests -y \ @@ -37,32 +37,32 @@ RUN apt-get install -y \ # Install MIRTK/SVRTK RUN git clone https://github.com/SVRTK/MIRTK.git \ - && mkdir -p /home/MIRTK/Packages/SVRTK -RUN git clone https://github.com/SVRTK/SVRTK.git /home/MIRTK/Packages/SVRTK \ - && mkdir -p /home/MIRTK/build \ - && cd /home/MIRTK/build \ + && mkdir -p /bin/MIRTK/Packages/SVRTK +RUN git clone https://github.com/SVRTK/SVRTK.git /bin/MIRTK/Packages/SVRTK \ + && mkdir -p /bin/MIRTK/build \ + && cd /bin/MIRTK/build \ && cmake -D WITH_TBB="ON" -D MODULE_SVRTK="ON" .. \ && make -j -ENV PATH="$PATH:/home/MIRTK/build/bin:/home/MIRTK/build/lib/tools" +ENV PATH="$PATH:/bin/MIRTK/build/bin:/bin/MIRTK/build/lib/tools" WORKDIR /var/monai -# Add 3D UNet model weights & execution script +# Add MONAI 3D UNet model weights & execution script WORKDIR /home # Setup 3D UNet models -RUN git clone https://github.com/SVRTK/Segmentation_FetalMRI.git --branch svrtk-docker-gpu-0.10 --single-branch /home/Segmentation_FetalMRI +RUN git clone https://github.com/SVRTK/auto-proc-svrtk.git /home/auto-proc-svrtk -RUN wget https://gin.g-node.org/SVRTK/fetal_mri_network_weights/raw/master/checkpoints-brain-loc-2-labels/latest.ckpt -P /home/Segmentation_FetalMRI/trained-models/checkpoints-brain-loc-labels \ - && wget https://gin.g-node.org/SVRTK/fetal_mri_network_weights/raw/master/checkpoints-brain-loc-2-labels-cropped/latest.ckpt -P /home/Segmentation_FetalMRI/trained-models/checkpoints-brain-loc-labels-cropped \ - && wget https://gin.g-node.org/SVRTK/fetal_mri_network_weights/raw/master/checkpoints-brain-reo-5-labels/latest.ckpt -P /home/Segmentation_FetalMRI/trained-models/checkpoints-brain-reorientation \ - && wget https://gin.g-node.org/SVRTK/fetal_mri_network_weights/raw/master/checkpoints-brain-reo-5-labels-raw-stacks/latest.ckpt -P /home/Segmentation_FetalMRI/trained-models/checkpoints-brain-reorientation-stacks +RUN wget https://gin.g-node.org/SVRTK/fetal_mri_network_weights/raw/master/monai-checkpoints-atunet-brain_bet_all_degree_raw_stacks-1-lab/best_metric_model.pth -P /home/auto-proc-svrtk/trained_models/monai-checkpoints-atunet-brain_bet_all_degree_raw_stacks-1-lab \ + && wget https://gin.g-node.org/SVRTK/fetal_mri_network_weights/raw/master/monai-checkpoints-unet-global-loc-2-lab/best_metric_model.pth -P /home/auto-proc-svrtk/trained_models/monai-checkpoints-unet-global-loc-2-lab \ + && wget https://gin.g-node.org/SVRTK/fetal_mri_network_weights/raw/master/monai-checkpoints-unet-svr-brain-reo-5-lab/best_metric_model.pth -P /home/auto-proc-svrtk/trained_models/monai-checkpoints-unet-svr-brain-reo-5-lab \ + && wget https://gin.g-node.org/SVRTK/fetal_mri_network_weights/raw/master/monai-checkpoints-unet-svr-brain-reo-raw-stacks-5-lab/best_metric_model.pth -P /home/auto-proc-svrtk/trained_models/monai-checkpoints-unet-svr-brain-reo-raw-stacks-5-lab # Install Python packages -RUN python -m pip install -r /home/Segmentation_FetalMRI/requirements.txt +RUN python -m pip install -r /home/auto-proc-svrtk/config/requirements.txt # Bugfix: without below, cannot import torch within Python # Error: OSError: /opt/hpcx/ompi/lib/libmpi.so.40: undefined symbol: opal_hwloc201_hwloc_get_type_depth @@ -70,9 +70,13 @@ RUN python -m pip install -r /home/Segmentation_FetalMRI/requirements.txt ENV PATH="${PATH}:/opt/hpcx/ompi/bin" ENV LD_LIBRARY_PATH="${LD_LIBRARY_PATH}:/opt/hpcx/ompi/lib" -COPY docker-recon-brain-auto.bash /home/scripts/docker-recon-brain-auto.bash -RUN mkdir -p /home/recon \ - && mkdir -p /home/output \ - && chmod +x /home/scripts/* +#COPY /home/auto-proc-svrtk/scripts/auto-brain-reconstruction-aide.sh /home/scripts/auto-brain-reconstruction.sh + + +RUN mkdir -p /home/recon +RUN mkdir -p /home/output +RUN mkdir -p /home/scripts +RUN chmod +x /home/auto-proc-svrtk/scripts/* + WORKDIR /var/monai diff --git a/app/docker-recon-brain-auto.bash b/app/docker-recon-brain-auto.bash deleted file mode 100644 index 51d012c..0000000 --- a/app/docker-recon-brain-auto.bash +++ /dev/null @@ -1,497 +0,0 @@ -#!/bin/bash - - -########################################################################################################################## -###### SCRIPT FOR AUTOMATED SVR RECONSTRUCTION (incl. optional GPU-acceleration): docker-recon-brain-auto.bash -###### King's College London 2021 -###### https://github.com/SVRTK/svrtk-docker-gpu -########################################################################################################################## - -echo -echo "-----------------------------------------------------------------------------" -echo "RUNNING AUTOMATED SVR RECONSTRUCTION" -echo "-----------------------------------------------------------------------------" -echo - -########################################################################################################################## -###### INPUTS -########################################################################################################################## - -#folder with input files -default_recon_dir=$1 - -#output path -output_path=$2 - -#cnn mode: -1 cpu, 1 gpu -cnn_mode=$3 - -#severity of motion mode: -1 minor, 1 severe -motion_correction_mode=$4 - - -source ~/.bashrc -cd ${default_recon_dir} - -########################################################################################################################## -###### DEFAULT PARAMETERS AND PATHS -########################################################################################################################## - - -# software and network weight paths -mirtk_path=/home/MIRTK/build/bin -segm_path=/home/Segmentation_FetalMRI -template_path=/home/Segmentation_FetalMRI/reference-templates -check_path_brain=/home/Segmentation_FetalMRI/trained-models/checkpoints-brain-loc-labels -check_path_brain_cropped=/home/Segmentation_FetalMRI/trained-models/checkpoints-brain-loc-labels-cropped -check_path_roi_reo_4lab=/home/Segmentation_FetalMRI/trained-models/checkpoints-brain-reorientation -check_path_roi_reo_4lab_stacks=/home/Segmentation_FetalMRI/trained-models/checkpoints-brain-reorientation-stacks - -# CNN parameters -res=128 -all_num_lab=1 -all_num_lab_reo=4 -current_lab=1 - -# NOTE: 1 package is acceptable for low motion datasets / -1 for severe cases for automated detection -if [ $motion_correction_mode -gt 0 ]; then - num_packages=-1 -else - num_packages=1 -fi - -# TODO: implement check in case slice thickness inconsistent across stacks -if [[ -f "${default_recon_dir}/log_slice_thickness.txt" ]]; then - default_thickness=`cat "${default_recon_dir}/log_slice_thickness.txt" | awk -F' ' '{print $1}'` -else - echo "WARNING: Setting default_thickness = 2.5" - default_thickness=2.5 -fi -echo - -# NOTE: change output resolution to 0.75 - 0.8 ? -output_resolution=0.85 - -# Reconstruction ROI names -roi_recon=("SVR") -roi_names=("brain") -roi_ids=(1) -selected_recon_roi=0 -selected_recon_roi=0 - -########################################################################################################################## -###### COPY INPUT .NII FILES TO THE PROCESSING FOLDER -########################################################################################################################## - -# navigate to the processing folder and remove any files left from previous processing sessions -test_dir=${default_recon_dir}/svr_processing_files -if [[ -d ${test_dir} ]];then -rm -r ${default_recon_dir}/svr_processing_files/* -else -mkdir ${default_recon_dir}/svr_processing_files -fi - -test_file=${default_recon_dir}/error.txt -if [[ -f ${test_file} ]];then -rm ${default_recon_dir}/error.txt -fi - -test_file=${default_recon_dir}/SVR-output.nii.gz -if [[ -f ${test_file} ]];then -rm ${default_recon_dir}/SVR-output* -fi - -# check the total number of available .nii files -num_stacks=$(find . -name "*.nii*" | wc -l) -if [ $num_stacks -gt 1 ]; then - echo " - found " ${num_stacks} ".nii files" -else - echo - echo "-----------------------------------------------------------------------------" - echo "ERROR : NO .NII FILES FOUND - EXIT ..." - echo "-----------------------------------------------------------------------------" - echo - echo "ERROR: NO INPUT FILLES. " > ${default_recon_dir}/error.txt - exit -fi - -main_dir=${default_recon_dir}/svr_processing_files -mkdir ${default_recon_dir}/svr_processing_files/cnn-recon-org-files -#find ${default_recon_dir}/ -name "*.nii*" -exec cp {} ${main_dir}/cnn-recon-org-files \; -cp ${default_recon_dir}/*.nii* ${main_dir}/cnn-recon-org-files/ -cd ${default_recon_dir}/svr_processing_files - - -mkdir cnn-recon-org-files-packages -cp cnn-recon-org-files/* cnn-recon-org-files-packages - -# split into packages only if the default parameter number is not 1 -if [ $num_packages -ne 1 ]; then - echo - echo "-----------------------------------------------------------------------------" - echo "SPLITTING INTO PACKAGES ..." - echo "-----------------------------------------------------------------------------" - echo - - cd ${main_dir} - cd cnn-recon-org-files-packages - - stack_names=$(ls *.nii*) - IFS=$'\n' read -rd '' -a all_stacks <<<"$stack_names" - - echo "folder : " ${in_file_dir} - for ((i=0;i<${#all_stacks[@]};i++)); - do - ${mirtk_path}/mirtk extract-packages ${all_stacks[$i]} ${num_packages} - rm ${all_stacks[$i]} - rm package-template.nii.gz - echo " - " ${all_stacks[$i]} - done -fi - -#final input files -cd ${main_dir} -stack_names=$(ls cnn-recon-org-files-packages/*.nii*) -IFS=$'\n' read -rd '' -a all_stacks <<<"$stack_names" - -########################################################################################################################## -###### 3D UNET-BASED GLOBAL BRAIN LOCALISATION + LOCAL MASK REFINEMENT + REORIENTATION (FOR SEVERE MOTION CASES) -########################################################################################################################## - -echo -echo "-----------------------------------------------------------------------------" -echo "3D UNET SEGMENTATION ..." -echo "-----------------------------------------------------------------------------" -echo - -cd ${main_dir} -mkdir cnn-out-files - -echo -echo "RUNNING GLOBAL LOCALISATION ..." -echo - -# 3D UNet global localisaton of the brain in the original stacks/packages -number_of_stacks=$(ls cnn-recon-org-files-packages/*.nii* | wc -l) -stack_names=$(ls cnn-recon-org-files-packages/*.nii*) -all_num_lab=1 -${mirtk_path}/mirtk prepare-for-cnn cnn-recon-res-files stack-files run.csv run-info-summary.csv ${res} ${number_of_stacks} $(echo $stack_names) ${all_num_lab} 0 -main_dir=$(pwd) -PYTHONIOENCODING=utf-8 python ${segm_path}/run_cnn_loc_gpu_cpu.py ${segm_path}/ ${check_path_brain}/ ${main_dir}/ ${main_dir}/cnn-out-files/ run.csv ${res} ${all_num_lab} ${cnn_mode} -out_mask_names=$(ls cnn-out-files/*seg_pr*.nii*) -out_stack_names=$(ls stack-files/*.nii*) -IFS=$'\n' read -rd '' -a all_masks <<<"$out_mask_names" -IFS=$'\n' read -rd '' -a all_stacks <<<"$out_stack_names" - -test_file=cnn-out-files/cnn-recon-res-files_in-res-stack-1000_img-0.nii.gz -if [[ ! -f ${test_file} ]];then - echo - echo "-----------------------------------------------------------------------------" - echo "ERROR : GLOBAL 3D UNET LOCALISATION FAILED ..." - echo "-----------------------------------------------------------------------------" - echo - echo "ERROR : GLOBAL 3D UNET LOCALISATION FAILED. " > ${default_recon_dir}/error.txt - exit -fi - -echo -echo "RUNNING CROPPED REFINEMENT ..." -echo - -mkdir cropped-files -mkdir cropped-cnn-out-files -mkdir tmp-masks-global - -# cropping all stacks based on the global brain ROI masks -for ((i=0;i<${#all_stacks[@]};i++)); -do - jj=$((${i}+1000)) - ${mirtk_path}/mirtk extract-label cnn-out-files/*-${jj}_seg_pr*.nii* tmp-org-m.nii.gz 1 1 - ${mirtk_path}/mirtk extract-connected-components tmp-org-m.nii.gz tmp-org-m.nii.gz -max-size 1000000 - ${mirtk_path}/mirtk dilate-image tmp-org-m.nii.gz dl-m.nii.gz -iterations 6 - ${mirtk_path}/mirtk crop-image stack-files/stack-${jj}.nii.gz dl-m.nii.gz cropped-files/cropped-stack-${jj}.nii.gz -done - -# 3D UNet segmentation of the brain in the cropped stacks -number_of_stacks=$(ls cropped-files/*.nii* | wc -l) -stack_names=$(ls cropped-files/*.nii*) -all_num_lab=1 -${mirtk_path}/mirtk prepare-for-cnn cropped-cnn-recon-res-files cropped-cnn-recon-stack-files cropped-run.csv cropped-run-info-summary.csv ${res} ${number_of_stacks} $(echo $stack_names) ${all_num_lab} 0 -PYTHONIOENCODING=utf-8 python ${segm_path}/run_cnn_loc_gpu_cpu.py ${segm_path}/ ${check_path_brain_cropped}/ ${main_dir}/ ${main_dir}/cropped-cnn-out-files/ cropped-run.csv ${res} ${all_num_lab} ${cnn_mode} - -test_file=cropped-cnn-out-files/cropped-cnn-recon-res-files_in-res-stack-1000_img-0.nii.gz -if [[ ! -f ${test_file} ]];then - echo - echo "-----------------------------------------------------------------------------" - echo "ERROR : CROPPED 3D UNET LOCALISATION FAILED ..." - echo "-----------------------------------------------------------------------------" - echo - echo "ERROR : CROPPED 3D UNET LOCALISATION FAILED. " > ${default_recon_dir}/error.txt - exit -fi - -echo -echo "EXTRACTING ROI-SPECIFIC MASKS ..." -echo - -cd ${main_dir} -out_mask_names=$(ls cropped-cnn-out-files/*seg_pr*.nii*) -out_stack_names=$(ls stack-files/*.nii*) -IFS=$'\n' read -rd '' -a all_masks <<<"$out_mask_names" -IFS=$'\n' read -rd '' -a all_stacks <<<"$out_stack_names" - -cd ${main_dir} - -mkdir mask-files-${roi_ids[${selected_recon_roi}]} -current_lab=1 - -for ((i=0;i<${#all_masks[@]};i++)); -do - jj=$((${i}+1000)) - # extraction of brain masks - ${mirtk_path}/mirtk transform-image cropped-cnn-out-files/*-${jj}_seg_pr*.nii* mask-files-${roi_ids[${selected_recon_roi}]}/mask-${jj}.nii.gz -target cropped-files/*-${jj}.nii.gz -interp NN - ${mirtk_path}/mirtk extract-label mask-files-${roi_ids[${selected_recon_roi}]}/mask-${jj}.nii.gz mask-files-${roi_ids[${selected_recon_roi}]}/mask-${jj}.nii.gz ${current_lab} ${current_lab} - ${mirtk_path}/mirtk extract-connected-components mask-files-${roi_ids[${selected_recon_roi}]}/mask-${jj}.nii.gz mask-files-${roi_ids[${selected_recon_roi}]}/mask-${jj}.nii.gz -max-size 800000 - - # centering stacks vs. brain centre for severe motion cases - #if [ $motion_correction_mode -gt 0 ]; then - - ${mirtk_path}/mirtk transform-image mask-files-${roi_ids[$selected_recon_roi]}/mask-${jj}.nii.gz mask-files-${roi_ids[$selected_recon_roi]}/mask-${jj}.nii.gz -target cropped-files/cropped-stack-${jj}.nii.gz -interp NN - ${mirtk_path}/mirtk centre-volume cropped-files/cropped-stack-${jj}.nii.gz mask-files-${roi_ids[$selected_recon_roi]}/mask-${jj}.nii.gz cropped-files/cropped-stack-${jj}.nii.gz - ${mirtk_path}/mirtk centre-volume mask-files-${roi_ids[$selected_recon_roi]}/mask-${jj}.nii.gz mask-files-${roi_ids[$selected_recon_roi]}/mask-${jj}.nii.gz mask-files-${roi_ids[$selected_recon_roi]}/mask-${jj}.nii.gz - - #fi - - echo stack-files/stack-${jj}.nii.gz " - " cropped-cnn-out-files/*-${jj}_seg_pr*.nii* " - " mask-files-${roi_ids[${selected_recon_roi}]}/mask-${jj}.nii.gz -done - -# CNN landmark-based reorientation of stacks to the standard space for severe motion cases -if [ $motion_correction_mode -gt 0 ]; then - - mkdir cropped-cnn-out-files-4reo - mkdir stack-dofs-to-atl - mkdir stack-reo-mask-files - - number_of_stacks=$(ls cropped-files/*.nii* | wc -l) - stack_names=$(ls cropped-files/*.nii*) - - # 3D UNet segmentation of 4 brain landmark labels - all_num_lab_reo=4 - ${mirtk_path}/mirtk prepare-for-cnn cropped-res-files cropped-stack-files cropped-run-4reo.csv cropped-run-info-summary.csv ${res} ${number_of_stacks} $(echo $stack_names) ${all_num_lab_reo} 0 - PYTHONIOENCODING=utf-8 python ${segm_path}/run_cnn_loc_gpu_cpu.py ${segm_path}/ ${check_path_roi_reo_4lab_stacks}/ ${main_dir}/ ${main_dir}/cropped-cnn-out-files-4reo/ cropped-run-4reo.csv ${res} ${all_num_lab_reo} ${cnn_mode} - - reo_roi_ids=(1 2 3 4) - lab_start=(1 2 1 4) - lab_stop=(1 2 4 4) - lab_cc=(2 2 1 1) - - # 1 - front WM, 2 - back WM, 3 - bet, 4 - cerebellum - test_file=cropped-cnn-out-files-4reo/cropped-res-files_in-res-stack-1000_img-0.nii.gz - if [[ ! -f ${test_file} ]];then - echo - echo "-----------------------------------------------------------------------------" - echo "ERROR : CROPPED 3D UNET REORIENTATION FAILED ..." - echo "-----------------------------------------------------------------------------" - echo - echo "ERROR : CROPPED 3D UNET REORIENTATION FAILED. " > ${default_recon_dir}/error.txt - exit - fi - - out_mask_names=$(ls cropped-cnn-out-files-4reo/*seg_pr*.nii*) - IFS=$'\n' read -rd '' -a all_masks <<<"$out_mask_names" - - for ((i=0;i<${#all_masks[@]};i++)); - do - - jj=$((${i}+1000)) - echo " - " ${jj} " ... " - - # extracting landmark masks - for ((w=0;w<${#reo_roi_ids[@]};w++)); - do - current_lab=${reo_roi_ids[$w]} - s1=${lab_start[$w]} - s2=${lab_stop[$w]} - currenct_cc=${lab_cc[$w]} - ${mirtk_path}/mirtk extract-label ${main_dir}/cropped-cnn-out-files-4reo/*-${jj}_seg_pr*.nii* tmp-org-m.nii.gz ${s1} ${s2} - ${mirtk_path}/mirtk extract-connected-components tmp-org-m.nii.gz tmp-org-m.nii.gz -n ${currenct_cc} - cp tmp-org-m.nii.gz stack-reo-mask-files/mask-${jj}-${current_lab}.nii.gz - done - - ${mirtk_path}/mirtk init-dof init.dof - z1=1; z2=2; z3=3; z4=4 ; - - # running landmark-based reorientation - ${mirtk_path}/mirtk register-landmarks ${template_path}/brain-ref-atlas-2021/new-brain-templ.nii.gz stack-reo-mask-files/mask-${jj}-0.nii.gz init.dof stack-dofs-to-atl/dof-to-atl-${jj}.dof 4 4 ${template_path}/brain-ref-atlas-2021/mask-${z1}.nii.gz ${template_path}/brain-ref-atlas-2021/mask-${z2}.nii.gz ${template_path}/brain-ref-atlas-2021/mask-${z3}.nii.gz ${template_path}/brain-ref-atlas-2021/mask-${z4}.nii.gz stack-reo-mask-files/mask-${jj}-${z1}.nii.gz stack-reo-mask-files/mask-${jj}-${z2}.nii.gz stack-reo-mask-files/mask-${jj}-${z3}.nii.gz stack-reo-mask-files/mask-${jj}-${z4}.nii.gz > tmp.txt - - # changing stack header orientation based on the transformations - ${mirtk_path}/mirtk edit-image cropped-files/cropped-stack-${jj}.nii.gz cropped-files/cropped-stack-${jj}.nii.gz -dofin_i stack-dofs-to-atl/dof-to-atl-${jj}.dof - ${mirtk_path}/mirtk edit-image mask-files-${roi_ids[$selected_recon_roi]}/mask-${jj}.nii.gz mask-files-${roi_ids[$selected_recon_roi]}/mask-${jj}.nii.gz -dofin_i stack-dofs-to-atl/dof-to-atl-${jj}.dof - - done - -fi - -########################################################################################################################## -###### STACK SELECTION + SVR RECONSTRUCTION -########################################################################################################################## - -echo -echo "-----------------------------------------------------------------------------" -echo "RUNNING STACK SELECTION ..." -echo "-----------------------------------------------------------------------------" -echo - -cd ${main_dir} - -mkdir out-proc-${roi_names[${selected_recon_roi}]} -cd out-proc-${roi_names[${selected_recon_roi}]} - -number_of_stacks=$(ls ../cropped-files/*.nii* | wc -l) -stack_names=$(ls ../cropped-files/*.nii*) -mask_names=$(ls ../mask-files-${roi_ids[$selected_recon_roi]}/*.nii*) - -# selection/exclusion of stacks based on similarity metrics and generation of the average mask -mkdir proc-stacks -${mirtk_path}/mirtk stacks-and-masks-selection ${number_of_stacks} $(echo $stack_names) $(echo $mask_names) proc-stacks 15 1 - -test_file=average_mask_cnn.nii.gz -if [[ ! -f ${test_file} ]];then - echo - echo "-----------------------------------------------------------------------------" - echo "ERROR : STACK SELECTION FAILED ..." - echo "-----------------------------------------------------------------------------" - echo - echo "ERROR: STACK SELECTION FAILED. " > ${default_recon_dir}/error.txt - exit -fi - -# generation of the average template -${mirtk_path}/mirtk average-images average_volume.nii.gz proc-stacks/*nii* -${mirtk_path}/mirtk resample-image average_volume.nii.gz average_volume.nii.gz -size 1 1 1 -${mirtk_path}/mirtk average-images average_volume.nii.gz proc-stacks/*nii* -target average_volume.nii.gz -${mirtk_path}/mirtk transform-image average_mask_cnn.nii.gz average_mask_cnn.nii.gz -target average_volume.nii.gz -interp NN -${mirtk_path}/mirtk erode-image average_mask_cnn.nii.gz average_mask_cnn.nii.gz - -echo -echo "-----------------------------------------------------------------------------" -echo "RUNNING RECONSTRUCTION ..." -echo "-----------------------------------------------------------------------------" -echo - -# SVR reconstruction -nStacks=$(ls proc-stacks/*.nii* | wc -l) -${mirtk_path}/mirtk reconstruct ${main_dir}/${roi_recon[${selected_recon_roi}]}-output.nii.gz ${nStacks} proc-stacks/*.nii* -mask average_mask_cnn.nii.gz -template average_volume.nii.gz -default_thickness ${default_thickness} -svr_only -iterations 3 -structural -resolution ${output_resolution} - -test_file=${main_dir}/${roi_recon[${selected_recon_roi}]}-output.nii.gz -if [[ ! -f ${test_file} ]];then - echo - echo "-----------------------------------------------------------------------------" - echo "ERROR : SVR RECONSTRUCTION FAILED ..." - echo "-----------------------------------------------------------------------------" - echo - echo "ERROR: SVR RECONSTRUCTION FAILED. " > ${default_recon_dir}/error.txt - exit -fi - -########################################################################################################################## -###### CNN-BASED REORIENTATION -########################################################################################################################## - -echo -echo "-----------------------------------------------------------------------------" -echo "RUNNING LANDMARK-BASED REORIENTATION ..." -echo "-----------------------------------------------------------------------------" -echo - -cd ${main_dir} - -# 1 - front WM, 2 - back WM, 3 - bet, 4 - cerebellum -reo_roi_ids=(1 2 3 4); lab_start=(1 2 1 4); lab_stop=(1 2 4 4); lab_cc=(2 2 1 1); - -mkdir dofs-to-atl -mkdir in-recon-file -mkdir final-reo-mask-files - -cp ${main_dir}/${roi_recon[${selected_recon_roi}]}-output.nii.gz in-recon-file - -# 3D UNet 4 landmark label localisation for the reconstructed image -res=128 -all_num_lab_reo=4 -cnn_out_mode=1234 -${mirtk_path}/mirtk prepare-for-cnn recon-res-files recon-stack-files run.csv run-info-summary.csv ${res} 1 in-recon-file/*nii* ${all_num_lab_reo} 0 -mkdir ${main_dir}/reo-${cnn_out_mode}-cnn-out-files -PYTHONIOENCODING=utf-8 python ${segm_path}/run_cnn_loc_gpu_cpu.py ${segm_path}/ ${check_path_roi_reo_4lab} ${main_dir}/ ${main_dir}/reo-${cnn_out_mode}-cnn-out-files/ run.csv ${res} ${all_num_lab_reo} ${cnn_mode} - -test_file=${main_dir}/reo-${cnn_out_mode}-cnn-out-files/recon-res-files_in-res-stack-1000_img-0.nii.gz -if [[ ! -f ${test_file} ]];then - echo - echo "-----------------------------------------------------------------------------" - echo "ERROR : 3D UNET REORIENTATION OF THE SVR-RECONSTRUCTED IMAGE FAILED ..." - echo "-----------------------------------------------------------------------------" - echo - echo "ERROR : 3D UNET REORIENTATION OF THE SVR-RECONSTRUCTED IMAGE FAILED. " > ${default_recon_dir}/error.txt - exit -fi - -# extration of landmark labels -jj=1000 -for ((w=0;w<${#reo_roi_ids[@]};w++)); -do - current_lab=${reo_roi_ids[$w]} - s1=${lab_start[$w]} - s2=${lab_stop[$w]} - currenct_cc=${lab_cc[$w]} - ${mirtk_path}/mirtk extract-label ${main_dir}/reo-${cnn_out_mode}-cnn-out-files/*-${jj}_seg_pr*.nii* tmp-org-m.nii.gz ${s1} ${s2} - ${mirtk_path}/mirtk extract-connected-components tmp-org-m.nii.gz tmp-org-m.nii.gz -n ${currenct_cc} - cp tmp-org-m.nii.gz final-reo-mask-files/mask-${jj}-${current_lab}.nii.gz -done - -${mirtk_path}/mirtk init-dof init.dof -z1=1; z2=2; z3=3; z4=4; - -# landmark-based registration to the atlas space -${mirtk_path}/mirtk register-landmarks ${template_path}/brain-ref-atlas-2021/new-brain-templ.nii.gz final-reo-mask-files/mask-${jj}-0.nii.gz init.dof dofs-to-atl/dof-to-atl-${jj}.dof 4 4 ${template_path}/brain-ref-atlas-2021/mask-${z1}.nii.gz ${template_path}/brain-ref-atlas-2021/mask-${z2}.nii.gz ${template_path}/brain-ref-atlas-2021/mask-${z3}.nii.gz ${template_path}/brain-ref-atlas-2021/mask-${z4}.nii.gz final-reo-mask-files/mask-${jj}-${z1}.nii.gz final-reo-mask-files/mask-${jj}-${z2}.nii.gz final-reo-mask-files/mask-${jj}-${z3}.nii.gz final-reo-mask-files/mask-${jj}-${z4}.nii.gz > tmp.txt - -# transformation to the atlas space -${mirtk_path}/mirtk transform-image ${main_dir}/${roi_recon[${selected_recon_roi}]}-output.nii.gz ${main_dir}/reo-${roi_recon[${selected_recon_roi}]}-output.nii.gz -dofin dofs-to-atl/dof-to-atl-${jj}.dof -target ${template_path}/brain-ref-atlas-2021/ref-space-brain.nii.gz -interp BSpline - -# cropping any remaining black background -${mirtk_path}/mirtk crop-image ${main_dir}/reo-${roi_recon[${selected_recon_roi}]}-output.nii.gz ${main_dir}/reo-${roi_recon[${selected_recon_roi}]}-output.nii.gz ${main_dir}/reo-${roi_recon[${selected_recon_roi}]}-output.nii.gz - -echo -echo "-----------------------------------------------------------------------------" -echo "ADJUSTING & RENAMING FINAL FILES" -echo "-----------------------------------------------------------------------------" -echo - -test_file=${main_dir}/reo-${roi_recon[${selected_recon_roi}]}-output.nii.gz -if [[ -f ${test_file} ]];then - - # TAR - rename files - mv ${main_dir}/${roi_recon[${selected_recon_roi}]}-output.nii.gz ${main_dir}/${roi_recon[${selected_recon_roi}]}-output-withoutReorientation.nii.gz - cp ${main_dir}/reo-${roi_recon[${selected_recon_roi}]}-output.nii.gz ${main_dir}/${roi_recon[${selected_recon_roi}]}-output.nii.gz - - # TAR - pad in-plane so square FOV - ${mirtk_path}/mirtk resample-to-iso-grid ${main_dir}/${roi_recon[${selected_recon_roi}]}-output.nii.gz ${main_dir}/${roi_recon[${selected_recon_roi}]}-output.nii.gz 2 - - # TAR - move final files to /recon dir - cp ${main_dir}/${roi_recon[${selected_recon_roi}]}-output-withoutReorientation.nii.gz ${default_recon_dir}/${roi_recon[${selected_recon_roi}]}-output-withoutReorientation.nii.gz - cp ${main_dir}/${roi_recon[${selected_recon_roi}]}-output.nii.gz ${default_recon_dir}/${roi_recon[${selected_recon_roi}]}-output.nii.gz - - cp ${main_dir}/${roi_recon[${selected_recon_roi}]}-output.nii.gz ${output_path}/${roi_recon[${selected_recon_roi}]}-output.nii.gz - - echo "Reconstruction was successful: " ${output_path}/${roi_recon[${selected_recon_roi}]}-output.nii.gz - -else - echo "Reconstruction failed ... " - echo "ERROR : RECONSTRUCTION PIPELINE FAILED (REORIENTATION OR SAVING STEP). " > ${default_recon_dir}/error.txt -fi - - - -echo -echo "-----------------------------------------------------------------------------" -echo "-----------------------------------------------------------------------------" -echo - diff --git a/app/operators/docker-recon-brain-auto.bash b/app/operators/docker-recon-brain-auto.bash deleted file mode 100644 index 51d012c..0000000 --- a/app/operators/docker-recon-brain-auto.bash +++ /dev/null @@ -1,497 +0,0 @@ -#!/bin/bash - - -########################################################################################################################## -###### SCRIPT FOR AUTOMATED SVR RECONSTRUCTION (incl. optional GPU-acceleration): docker-recon-brain-auto.bash -###### King's College London 2021 -###### https://github.com/SVRTK/svrtk-docker-gpu -########################################################################################################################## - -echo -echo "-----------------------------------------------------------------------------" -echo "RUNNING AUTOMATED SVR RECONSTRUCTION" -echo "-----------------------------------------------------------------------------" -echo - -########################################################################################################################## -###### INPUTS -########################################################################################################################## - -#folder with input files -default_recon_dir=$1 - -#output path -output_path=$2 - -#cnn mode: -1 cpu, 1 gpu -cnn_mode=$3 - -#severity of motion mode: -1 minor, 1 severe -motion_correction_mode=$4 - - -source ~/.bashrc -cd ${default_recon_dir} - -########################################################################################################################## -###### DEFAULT PARAMETERS AND PATHS -########################################################################################################################## - - -# software and network weight paths -mirtk_path=/home/MIRTK/build/bin -segm_path=/home/Segmentation_FetalMRI -template_path=/home/Segmentation_FetalMRI/reference-templates -check_path_brain=/home/Segmentation_FetalMRI/trained-models/checkpoints-brain-loc-labels -check_path_brain_cropped=/home/Segmentation_FetalMRI/trained-models/checkpoints-brain-loc-labels-cropped -check_path_roi_reo_4lab=/home/Segmentation_FetalMRI/trained-models/checkpoints-brain-reorientation -check_path_roi_reo_4lab_stacks=/home/Segmentation_FetalMRI/trained-models/checkpoints-brain-reorientation-stacks - -# CNN parameters -res=128 -all_num_lab=1 -all_num_lab_reo=4 -current_lab=1 - -# NOTE: 1 package is acceptable for low motion datasets / -1 for severe cases for automated detection -if [ $motion_correction_mode -gt 0 ]; then - num_packages=-1 -else - num_packages=1 -fi - -# TODO: implement check in case slice thickness inconsistent across stacks -if [[ -f "${default_recon_dir}/log_slice_thickness.txt" ]]; then - default_thickness=`cat "${default_recon_dir}/log_slice_thickness.txt" | awk -F' ' '{print $1}'` -else - echo "WARNING: Setting default_thickness = 2.5" - default_thickness=2.5 -fi -echo - -# NOTE: change output resolution to 0.75 - 0.8 ? -output_resolution=0.85 - -# Reconstruction ROI names -roi_recon=("SVR") -roi_names=("brain") -roi_ids=(1) -selected_recon_roi=0 -selected_recon_roi=0 - -########################################################################################################################## -###### COPY INPUT .NII FILES TO THE PROCESSING FOLDER -########################################################################################################################## - -# navigate to the processing folder and remove any files left from previous processing sessions -test_dir=${default_recon_dir}/svr_processing_files -if [[ -d ${test_dir} ]];then -rm -r ${default_recon_dir}/svr_processing_files/* -else -mkdir ${default_recon_dir}/svr_processing_files -fi - -test_file=${default_recon_dir}/error.txt -if [[ -f ${test_file} ]];then -rm ${default_recon_dir}/error.txt -fi - -test_file=${default_recon_dir}/SVR-output.nii.gz -if [[ -f ${test_file} ]];then -rm ${default_recon_dir}/SVR-output* -fi - -# check the total number of available .nii files -num_stacks=$(find . -name "*.nii*" | wc -l) -if [ $num_stacks -gt 1 ]; then - echo " - found " ${num_stacks} ".nii files" -else - echo - echo "-----------------------------------------------------------------------------" - echo "ERROR : NO .NII FILES FOUND - EXIT ..." - echo "-----------------------------------------------------------------------------" - echo - echo "ERROR: NO INPUT FILLES. " > ${default_recon_dir}/error.txt - exit -fi - -main_dir=${default_recon_dir}/svr_processing_files -mkdir ${default_recon_dir}/svr_processing_files/cnn-recon-org-files -#find ${default_recon_dir}/ -name "*.nii*" -exec cp {} ${main_dir}/cnn-recon-org-files \; -cp ${default_recon_dir}/*.nii* ${main_dir}/cnn-recon-org-files/ -cd ${default_recon_dir}/svr_processing_files - - -mkdir cnn-recon-org-files-packages -cp cnn-recon-org-files/* cnn-recon-org-files-packages - -# split into packages only if the default parameter number is not 1 -if [ $num_packages -ne 1 ]; then - echo - echo "-----------------------------------------------------------------------------" - echo "SPLITTING INTO PACKAGES ..." - echo "-----------------------------------------------------------------------------" - echo - - cd ${main_dir} - cd cnn-recon-org-files-packages - - stack_names=$(ls *.nii*) - IFS=$'\n' read -rd '' -a all_stacks <<<"$stack_names" - - echo "folder : " ${in_file_dir} - for ((i=0;i<${#all_stacks[@]};i++)); - do - ${mirtk_path}/mirtk extract-packages ${all_stacks[$i]} ${num_packages} - rm ${all_stacks[$i]} - rm package-template.nii.gz - echo " - " ${all_stacks[$i]} - done -fi - -#final input files -cd ${main_dir} -stack_names=$(ls cnn-recon-org-files-packages/*.nii*) -IFS=$'\n' read -rd '' -a all_stacks <<<"$stack_names" - -########################################################################################################################## -###### 3D UNET-BASED GLOBAL BRAIN LOCALISATION + LOCAL MASK REFINEMENT + REORIENTATION (FOR SEVERE MOTION CASES) -########################################################################################################################## - -echo -echo "-----------------------------------------------------------------------------" -echo "3D UNET SEGMENTATION ..." -echo "-----------------------------------------------------------------------------" -echo - -cd ${main_dir} -mkdir cnn-out-files - -echo -echo "RUNNING GLOBAL LOCALISATION ..." -echo - -# 3D UNet global localisaton of the brain in the original stacks/packages -number_of_stacks=$(ls cnn-recon-org-files-packages/*.nii* | wc -l) -stack_names=$(ls cnn-recon-org-files-packages/*.nii*) -all_num_lab=1 -${mirtk_path}/mirtk prepare-for-cnn cnn-recon-res-files stack-files run.csv run-info-summary.csv ${res} ${number_of_stacks} $(echo $stack_names) ${all_num_lab} 0 -main_dir=$(pwd) -PYTHONIOENCODING=utf-8 python ${segm_path}/run_cnn_loc_gpu_cpu.py ${segm_path}/ ${check_path_brain}/ ${main_dir}/ ${main_dir}/cnn-out-files/ run.csv ${res} ${all_num_lab} ${cnn_mode} -out_mask_names=$(ls cnn-out-files/*seg_pr*.nii*) -out_stack_names=$(ls stack-files/*.nii*) -IFS=$'\n' read -rd '' -a all_masks <<<"$out_mask_names" -IFS=$'\n' read -rd '' -a all_stacks <<<"$out_stack_names" - -test_file=cnn-out-files/cnn-recon-res-files_in-res-stack-1000_img-0.nii.gz -if [[ ! -f ${test_file} ]];then - echo - echo "-----------------------------------------------------------------------------" - echo "ERROR : GLOBAL 3D UNET LOCALISATION FAILED ..." - echo "-----------------------------------------------------------------------------" - echo - echo "ERROR : GLOBAL 3D UNET LOCALISATION FAILED. " > ${default_recon_dir}/error.txt - exit -fi - -echo -echo "RUNNING CROPPED REFINEMENT ..." -echo - -mkdir cropped-files -mkdir cropped-cnn-out-files -mkdir tmp-masks-global - -# cropping all stacks based on the global brain ROI masks -for ((i=0;i<${#all_stacks[@]};i++)); -do - jj=$((${i}+1000)) - ${mirtk_path}/mirtk extract-label cnn-out-files/*-${jj}_seg_pr*.nii* tmp-org-m.nii.gz 1 1 - ${mirtk_path}/mirtk extract-connected-components tmp-org-m.nii.gz tmp-org-m.nii.gz -max-size 1000000 - ${mirtk_path}/mirtk dilate-image tmp-org-m.nii.gz dl-m.nii.gz -iterations 6 - ${mirtk_path}/mirtk crop-image stack-files/stack-${jj}.nii.gz dl-m.nii.gz cropped-files/cropped-stack-${jj}.nii.gz -done - -# 3D UNet segmentation of the brain in the cropped stacks -number_of_stacks=$(ls cropped-files/*.nii* | wc -l) -stack_names=$(ls cropped-files/*.nii*) -all_num_lab=1 -${mirtk_path}/mirtk prepare-for-cnn cropped-cnn-recon-res-files cropped-cnn-recon-stack-files cropped-run.csv cropped-run-info-summary.csv ${res} ${number_of_stacks} $(echo $stack_names) ${all_num_lab} 0 -PYTHONIOENCODING=utf-8 python ${segm_path}/run_cnn_loc_gpu_cpu.py ${segm_path}/ ${check_path_brain_cropped}/ ${main_dir}/ ${main_dir}/cropped-cnn-out-files/ cropped-run.csv ${res} ${all_num_lab} ${cnn_mode} - -test_file=cropped-cnn-out-files/cropped-cnn-recon-res-files_in-res-stack-1000_img-0.nii.gz -if [[ ! -f ${test_file} ]];then - echo - echo "-----------------------------------------------------------------------------" - echo "ERROR : CROPPED 3D UNET LOCALISATION FAILED ..." - echo "-----------------------------------------------------------------------------" - echo - echo "ERROR : CROPPED 3D UNET LOCALISATION FAILED. " > ${default_recon_dir}/error.txt - exit -fi - -echo -echo "EXTRACTING ROI-SPECIFIC MASKS ..." -echo - -cd ${main_dir} -out_mask_names=$(ls cropped-cnn-out-files/*seg_pr*.nii*) -out_stack_names=$(ls stack-files/*.nii*) -IFS=$'\n' read -rd '' -a all_masks <<<"$out_mask_names" -IFS=$'\n' read -rd '' -a all_stacks <<<"$out_stack_names" - -cd ${main_dir} - -mkdir mask-files-${roi_ids[${selected_recon_roi}]} -current_lab=1 - -for ((i=0;i<${#all_masks[@]};i++)); -do - jj=$((${i}+1000)) - # extraction of brain masks - ${mirtk_path}/mirtk transform-image cropped-cnn-out-files/*-${jj}_seg_pr*.nii* mask-files-${roi_ids[${selected_recon_roi}]}/mask-${jj}.nii.gz -target cropped-files/*-${jj}.nii.gz -interp NN - ${mirtk_path}/mirtk extract-label mask-files-${roi_ids[${selected_recon_roi}]}/mask-${jj}.nii.gz mask-files-${roi_ids[${selected_recon_roi}]}/mask-${jj}.nii.gz ${current_lab} ${current_lab} - ${mirtk_path}/mirtk extract-connected-components mask-files-${roi_ids[${selected_recon_roi}]}/mask-${jj}.nii.gz mask-files-${roi_ids[${selected_recon_roi}]}/mask-${jj}.nii.gz -max-size 800000 - - # centering stacks vs. brain centre for severe motion cases - #if [ $motion_correction_mode -gt 0 ]; then - - ${mirtk_path}/mirtk transform-image mask-files-${roi_ids[$selected_recon_roi]}/mask-${jj}.nii.gz mask-files-${roi_ids[$selected_recon_roi]}/mask-${jj}.nii.gz -target cropped-files/cropped-stack-${jj}.nii.gz -interp NN - ${mirtk_path}/mirtk centre-volume cropped-files/cropped-stack-${jj}.nii.gz mask-files-${roi_ids[$selected_recon_roi]}/mask-${jj}.nii.gz cropped-files/cropped-stack-${jj}.nii.gz - ${mirtk_path}/mirtk centre-volume mask-files-${roi_ids[$selected_recon_roi]}/mask-${jj}.nii.gz mask-files-${roi_ids[$selected_recon_roi]}/mask-${jj}.nii.gz mask-files-${roi_ids[$selected_recon_roi]}/mask-${jj}.nii.gz - - #fi - - echo stack-files/stack-${jj}.nii.gz " - " cropped-cnn-out-files/*-${jj}_seg_pr*.nii* " - " mask-files-${roi_ids[${selected_recon_roi}]}/mask-${jj}.nii.gz -done - -# CNN landmark-based reorientation of stacks to the standard space for severe motion cases -if [ $motion_correction_mode -gt 0 ]; then - - mkdir cropped-cnn-out-files-4reo - mkdir stack-dofs-to-atl - mkdir stack-reo-mask-files - - number_of_stacks=$(ls cropped-files/*.nii* | wc -l) - stack_names=$(ls cropped-files/*.nii*) - - # 3D UNet segmentation of 4 brain landmark labels - all_num_lab_reo=4 - ${mirtk_path}/mirtk prepare-for-cnn cropped-res-files cropped-stack-files cropped-run-4reo.csv cropped-run-info-summary.csv ${res} ${number_of_stacks} $(echo $stack_names) ${all_num_lab_reo} 0 - PYTHONIOENCODING=utf-8 python ${segm_path}/run_cnn_loc_gpu_cpu.py ${segm_path}/ ${check_path_roi_reo_4lab_stacks}/ ${main_dir}/ ${main_dir}/cropped-cnn-out-files-4reo/ cropped-run-4reo.csv ${res} ${all_num_lab_reo} ${cnn_mode} - - reo_roi_ids=(1 2 3 4) - lab_start=(1 2 1 4) - lab_stop=(1 2 4 4) - lab_cc=(2 2 1 1) - - # 1 - front WM, 2 - back WM, 3 - bet, 4 - cerebellum - test_file=cropped-cnn-out-files-4reo/cropped-res-files_in-res-stack-1000_img-0.nii.gz - if [[ ! -f ${test_file} ]];then - echo - echo "-----------------------------------------------------------------------------" - echo "ERROR : CROPPED 3D UNET REORIENTATION FAILED ..." - echo "-----------------------------------------------------------------------------" - echo - echo "ERROR : CROPPED 3D UNET REORIENTATION FAILED. " > ${default_recon_dir}/error.txt - exit - fi - - out_mask_names=$(ls cropped-cnn-out-files-4reo/*seg_pr*.nii*) - IFS=$'\n' read -rd '' -a all_masks <<<"$out_mask_names" - - for ((i=0;i<${#all_masks[@]};i++)); - do - - jj=$((${i}+1000)) - echo " - " ${jj} " ... " - - # extracting landmark masks - for ((w=0;w<${#reo_roi_ids[@]};w++)); - do - current_lab=${reo_roi_ids[$w]} - s1=${lab_start[$w]} - s2=${lab_stop[$w]} - currenct_cc=${lab_cc[$w]} - ${mirtk_path}/mirtk extract-label ${main_dir}/cropped-cnn-out-files-4reo/*-${jj}_seg_pr*.nii* tmp-org-m.nii.gz ${s1} ${s2} - ${mirtk_path}/mirtk extract-connected-components tmp-org-m.nii.gz tmp-org-m.nii.gz -n ${currenct_cc} - cp tmp-org-m.nii.gz stack-reo-mask-files/mask-${jj}-${current_lab}.nii.gz - done - - ${mirtk_path}/mirtk init-dof init.dof - z1=1; z2=2; z3=3; z4=4 ; - - # running landmark-based reorientation - ${mirtk_path}/mirtk register-landmarks ${template_path}/brain-ref-atlas-2021/new-brain-templ.nii.gz stack-reo-mask-files/mask-${jj}-0.nii.gz init.dof stack-dofs-to-atl/dof-to-atl-${jj}.dof 4 4 ${template_path}/brain-ref-atlas-2021/mask-${z1}.nii.gz ${template_path}/brain-ref-atlas-2021/mask-${z2}.nii.gz ${template_path}/brain-ref-atlas-2021/mask-${z3}.nii.gz ${template_path}/brain-ref-atlas-2021/mask-${z4}.nii.gz stack-reo-mask-files/mask-${jj}-${z1}.nii.gz stack-reo-mask-files/mask-${jj}-${z2}.nii.gz stack-reo-mask-files/mask-${jj}-${z3}.nii.gz stack-reo-mask-files/mask-${jj}-${z4}.nii.gz > tmp.txt - - # changing stack header orientation based on the transformations - ${mirtk_path}/mirtk edit-image cropped-files/cropped-stack-${jj}.nii.gz cropped-files/cropped-stack-${jj}.nii.gz -dofin_i stack-dofs-to-atl/dof-to-atl-${jj}.dof - ${mirtk_path}/mirtk edit-image mask-files-${roi_ids[$selected_recon_roi]}/mask-${jj}.nii.gz mask-files-${roi_ids[$selected_recon_roi]}/mask-${jj}.nii.gz -dofin_i stack-dofs-to-atl/dof-to-atl-${jj}.dof - - done - -fi - -########################################################################################################################## -###### STACK SELECTION + SVR RECONSTRUCTION -########################################################################################################################## - -echo -echo "-----------------------------------------------------------------------------" -echo "RUNNING STACK SELECTION ..." -echo "-----------------------------------------------------------------------------" -echo - -cd ${main_dir} - -mkdir out-proc-${roi_names[${selected_recon_roi}]} -cd out-proc-${roi_names[${selected_recon_roi}]} - -number_of_stacks=$(ls ../cropped-files/*.nii* | wc -l) -stack_names=$(ls ../cropped-files/*.nii*) -mask_names=$(ls ../mask-files-${roi_ids[$selected_recon_roi]}/*.nii*) - -# selection/exclusion of stacks based on similarity metrics and generation of the average mask -mkdir proc-stacks -${mirtk_path}/mirtk stacks-and-masks-selection ${number_of_stacks} $(echo $stack_names) $(echo $mask_names) proc-stacks 15 1 - -test_file=average_mask_cnn.nii.gz -if [[ ! -f ${test_file} ]];then - echo - echo "-----------------------------------------------------------------------------" - echo "ERROR : STACK SELECTION FAILED ..." - echo "-----------------------------------------------------------------------------" - echo - echo "ERROR: STACK SELECTION FAILED. " > ${default_recon_dir}/error.txt - exit -fi - -# generation of the average template -${mirtk_path}/mirtk average-images average_volume.nii.gz proc-stacks/*nii* -${mirtk_path}/mirtk resample-image average_volume.nii.gz average_volume.nii.gz -size 1 1 1 -${mirtk_path}/mirtk average-images average_volume.nii.gz proc-stacks/*nii* -target average_volume.nii.gz -${mirtk_path}/mirtk transform-image average_mask_cnn.nii.gz average_mask_cnn.nii.gz -target average_volume.nii.gz -interp NN -${mirtk_path}/mirtk erode-image average_mask_cnn.nii.gz average_mask_cnn.nii.gz - -echo -echo "-----------------------------------------------------------------------------" -echo "RUNNING RECONSTRUCTION ..." -echo "-----------------------------------------------------------------------------" -echo - -# SVR reconstruction -nStacks=$(ls proc-stacks/*.nii* | wc -l) -${mirtk_path}/mirtk reconstruct ${main_dir}/${roi_recon[${selected_recon_roi}]}-output.nii.gz ${nStacks} proc-stacks/*.nii* -mask average_mask_cnn.nii.gz -template average_volume.nii.gz -default_thickness ${default_thickness} -svr_only -iterations 3 -structural -resolution ${output_resolution} - -test_file=${main_dir}/${roi_recon[${selected_recon_roi}]}-output.nii.gz -if [[ ! -f ${test_file} ]];then - echo - echo "-----------------------------------------------------------------------------" - echo "ERROR : SVR RECONSTRUCTION FAILED ..." - echo "-----------------------------------------------------------------------------" - echo - echo "ERROR: SVR RECONSTRUCTION FAILED. " > ${default_recon_dir}/error.txt - exit -fi - -########################################################################################################################## -###### CNN-BASED REORIENTATION -########################################################################################################################## - -echo -echo "-----------------------------------------------------------------------------" -echo "RUNNING LANDMARK-BASED REORIENTATION ..." -echo "-----------------------------------------------------------------------------" -echo - -cd ${main_dir} - -# 1 - front WM, 2 - back WM, 3 - bet, 4 - cerebellum -reo_roi_ids=(1 2 3 4); lab_start=(1 2 1 4); lab_stop=(1 2 4 4); lab_cc=(2 2 1 1); - -mkdir dofs-to-atl -mkdir in-recon-file -mkdir final-reo-mask-files - -cp ${main_dir}/${roi_recon[${selected_recon_roi}]}-output.nii.gz in-recon-file - -# 3D UNet 4 landmark label localisation for the reconstructed image -res=128 -all_num_lab_reo=4 -cnn_out_mode=1234 -${mirtk_path}/mirtk prepare-for-cnn recon-res-files recon-stack-files run.csv run-info-summary.csv ${res} 1 in-recon-file/*nii* ${all_num_lab_reo} 0 -mkdir ${main_dir}/reo-${cnn_out_mode}-cnn-out-files -PYTHONIOENCODING=utf-8 python ${segm_path}/run_cnn_loc_gpu_cpu.py ${segm_path}/ ${check_path_roi_reo_4lab} ${main_dir}/ ${main_dir}/reo-${cnn_out_mode}-cnn-out-files/ run.csv ${res} ${all_num_lab_reo} ${cnn_mode} - -test_file=${main_dir}/reo-${cnn_out_mode}-cnn-out-files/recon-res-files_in-res-stack-1000_img-0.nii.gz -if [[ ! -f ${test_file} ]];then - echo - echo "-----------------------------------------------------------------------------" - echo "ERROR : 3D UNET REORIENTATION OF THE SVR-RECONSTRUCTED IMAGE FAILED ..." - echo "-----------------------------------------------------------------------------" - echo - echo "ERROR : 3D UNET REORIENTATION OF THE SVR-RECONSTRUCTED IMAGE FAILED. " > ${default_recon_dir}/error.txt - exit -fi - -# extration of landmark labels -jj=1000 -for ((w=0;w<${#reo_roi_ids[@]};w++)); -do - current_lab=${reo_roi_ids[$w]} - s1=${lab_start[$w]} - s2=${lab_stop[$w]} - currenct_cc=${lab_cc[$w]} - ${mirtk_path}/mirtk extract-label ${main_dir}/reo-${cnn_out_mode}-cnn-out-files/*-${jj}_seg_pr*.nii* tmp-org-m.nii.gz ${s1} ${s2} - ${mirtk_path}/mirtk extract-connected-components tmp-org-m.nii.gz tmp-org-m.nii.gz -n ${currenct_cc} - cp tmp-org-m.nii.gz final-reo-mask-files/mask-${jj}-${current_lab}.nii.gz -done - -${mirtk_path}/mirtk init-dof init.dof -z1=1; z2=2; z3=3; z4=4; - -# landmark-based registration to the atlas space -${mirtk_path}/mirtk register-landmarks ${template_path}/brain-ref-atlas-2021/new-brain-templ.nii.gz final-reo-mask-files/mask-${jj}-0.nii.gz init.dof dofs-to-atl/dof-to-atl-${jj}.dof 4 4 ${template_path}/brain-ref-atlas-2021/mask-${z1}.nii.gz ${template_path}/brain-ref-atlas-2021/mask-${z2}.nii.gz ${template_path}/brain-ref-atlas-2021/mask-${z3}.nii.gz ${template_path}/brain-ref-atlas-2021/mask-${z4}.nii.gz final-reo-mask-files/mask-${jj}-${z1}.nii.gz final-reo-mask-files/mask-${jj}-${z2}.nii.gz final-reo-mask-files/mask-${jj}-${z3}.nii.gz final-reo-mask-files/mask-${jj}-${z4}.nii.gz > tmp.txt - -# transformation to the atlas space -${mirtk_path}/mirtk transform-image ${main_dir}/${roi_recon[${selected_recon_roi}]}-output.nii.gz ${main_dir}/reo-${roi_recon[${selected_recon_roi}]}-output.nii.gz -dofin dofs-to-atl/dof-to-atl-${jj}.dof -target ${template_path}/brain-ref-atlas-2021/ref-space-brain.nii.gz -interp BSpline - -# cropping any remaining black background -${mirtk_path}/mirtk crop-image ${main_dir}/reo-${roi_recon[${selected_recon_roi}]}-output.nii.gz ${main_dir}/reo-${roi_recon[${selected_recon_roi}]}-output.nii.gz ${main_dir}/reo-${roi_recon[${selected_recon_roi}]}-output.nii.gz - -echo -echo "-----------------------------------------------------------------------------" -echo "ADJUSTING & RENAMING FINAL FILES" -echo "-----------------------------------------------------------------------------" -echo - -test_file=${main_dir}/reo-${roi_recon[${selected_recon_roi}]}-output.nii.gz -if [[ -f ${test_file} ]];then - - # TAR - rename files - mv ${main_dir}/${roi_recon[${selected_recon_roi}]}-output.nii.gz ${main_dir}/${roi_recon[${selected_recon_roi}]}-output-withoutReorientation.nii.gz - cp ${main_dir}/reo-${roi_recon[${selected_recon_roi}]}-output.nii.gz ${main_dir}/${roi_recon[${selected_recon_roi}]}-output.nii.gz - - # TAR - pad in-plane so square FOV - ${mirtk_path}/mirtk resample-to-iso-grid ${main_dir}/${roi_recon[${selected_recon_roi}]}-output.nii.gz ${main_dir}/${roi_recon[${selected_recon_roi}]}-output.nii.gz 2 - - # TAR - move final files to /recon dir - cp ${main_dir}/${roi_recon[${selected_recon_roi}]}-output-withoutReorientation.nii.gz ${default_recon_dir}/${roi_recon[${selected_recon_roi}]}-output-withoutReorientation.nii.gz - cp ${main_dir}/${roi_recon[${selected_recon_roi}]}-output.nii.gz ${default_recon_dir}/${roi_recon[${selected_recon_roi}]}-output.nii.gz - - cp ${main_dir}/${roi_recon[${selected_recon_roi}]}-output.nii.gz ${output_path}/${roi_recon[${selected_recon_roi}]}-output.nii.gz - - echo "Reconstruction was successful: " ${output_path}/${roi_recon[${selected_recon_roi}]}-output.nii.gz - -else - echo "Reconstruction failed ... " - echo "ERROR : RECONSTRUCTION PIPELINE FAILED (REORIENTATION OR SAVING STEP). " > ${default_recon_dir}/error.txt -fi - - - -echo -echo "-----------------------------------------------------------------------------" -echo "-----------------------------------------------------------------------------" -echo - diff --git a/app/operators/fetal_mri_3d_brain_recon_operator.py b/app/operators/fetal_mri_3d_brain_recon_operator.py index 6af3a1b..08bddfb 100644 --- a/app/operators/fetal_mri_3d_brain_recon_operator.py +++ b/app/operators/fetal_mri_3d_brain_recon_operator.py @@ -33,28 +33,26 @@ def compute(self, op_input: InputContext, op_output: OutputContext, context: Exe # Run motion corrected reconstruction if not is_local_testing: - if torch.cuda.is_available(): - cnn_mode = "1" - logging.info("SVRTK reconstruction using GPU mode ...") - if not torch.cuda.is_available(): - cnn_mode = "-1" - logging.info("SVRTK reconstruction using CPU mode ...") - - motion_correction_mode = "-1" # -1 minor, 1 severe - logging.info("SVRTK reconstruction using Minor motion correction mode ...") +# if torch.cuda.is_available(): +# cnn_mode = "1" +# logging.info("SVRTK reconstruction using GPU mode ...") +# if not torch.cuda.is_available(): +# cnn_mode = "-1" +# logging.info("SVRTK reconstruction using CPU mode ...") +# +# motion_correction_mode = "-1" # -1 minor, 1 severe +# logging.info("SVRTK reconstruction ...") subprocess.run([ - "/home/scripts/docker-recon-brain-auto.bash", + "/home/auto-proc-svrtk/scripts/auto-brain-reconstruction-aide.sh", nii_stacks_path, - operator_workdir, - cnn_mode, - motion_correction_mode + operator_workdir ]) # Local testing: # create dummy SVR-output.nii.gz file in same location as output from docker-recon-brain-auto.bash if is_local_testing: - subprocess.run(["cp", "/path/to/local/SVR-output.nii.gz", operator_workdir]) + subprocess.run(["cp", "/path/to/local/reo-SVR-output-brain.nii.gz", operator_workdir]) logging.info("Completed SVRTK reconstruction ...") diff --git a/app/operators/nii2dcm_operator.py b/app/operators/nii2dcm_operator.py index 4719a7a..d11feed 100644 --- a/app/operators/nii2dcm_operator.py +++ b/app/operators/nii2dcm_operator.py @@ -55,7 +55,7 @@ def compute(self, op_input: InputContext, op_output: OutputContext, context: Exe dcm_output_path = op_output.get().path # SVRTK output - svrtk_output_filename = 'SVR-output.nii.gz' + svrtk_output_filename = 'reo-SVR-output-brain.nii.gz' svrtk_output_path = op_input.get("svrtk_output").path svrtk_nii_path = Path(svrtk_output_path / svrtk_output_filename) logging.info(f"SVRTK output NIfTI found: {svrtk_nii_path}") diff --git a/app/workflows/fetal-3d-brain-mri.json b/app/workflows/fetal-3d-brain-mri.json index 8979af9..42119c7 100644 --- a/app/workflows/fetal-3d-brain-mri.json +++ b/app/workflows/fetal-3d-brain-mri.json @@ -1,9 +1,9 @@ { "name": "fetal-brain-mri", - "version": "1.0.0", + "version": "0.2.1", "description": "Clinical Workflow for 3D fetal brain MRI reconstruction", "informatics_gateway": { - "ae_title": "MONAI", + "ae_title": "AIDE_QA", "data_origins": [], "export_destinations": [ "PACS" @@ -53,8 +53,8 @@ "workflow_name": "fetal-brain-mri", "reviewed_task_id": "brain-3d-recon", "application_name": "fetal-brain-mri", - "application_version": "1.0.0", - "mode": "QA", + "application_version": "0.2.1", + "mode": "AIDE_QA", "reviewer_roles": "clinician" }, "task_destinations": [ diff --git a/app/workflows/fetal-brain-3d-recon-argo-template.yaml b/app/workflows/fetal-brain-3d-recon-argo-template.yaml index 9fabda8..e493108 100644 --- a/app/workflows/fetal-brain-3d-recon-argo-template.yaml +++ b/app/workflows/fetal-brain-3d-recon-argo-template.yaml @@ -32,7 +32,7 @@ spec: metadata: {} container: name: '' - image: ghcr.io/svrtk/aide-svrtk/map:0.2.0 + image: ghcr.io/svrtk/aide-svrtk/map:0.2.1 command: - /bin/bash - '-c'