About

This repository includes scripts to process and analyze images from all 1714 acute individuals with suspected stroke (with 1449 having stroke confirmed) in the Stroke Outcome Optimization Project (SOOP) , as well as the resulting images derived from the processing. The goal of this repository is to provide a minimal starting point for analyzing the SOOP. This acute dataset complements the Aphasia Recovery Cohort (ARC) from individuals with chronic impairments. This educational resource illustrates how to process clinical datasets stored in the BIDS format. Our hope is that more sophisticated methods can improve clinical lesion mapping, spatial processing and prediction.

Scripts

The Matlab Scripts require Matlab, SPM12, and the clinical toolbox installed. All scripts require the raw data provided from the Stroke Outcome Optimization Project (SOOP). Users will need to adjust the paths for the scripts to match their file system.

The Matlab scripts must be run first:

• acuteNormalize.m : this Matlab script spatially normalizes lesions and images using the clinical toolbox for SPM. For each indvidual, it assumes a FLAIR scan, a TRACE scan and a lesion drawn on the TRACE. It first coregisters the TRACE image to the FLAIR image, transforming the lesion to FLAIR space. The FLAIR scan is then normalized into standard space using the FLAIR template. This transform is used to warp the lesion into standard space. The T1w image is also brain extracted based on estimates of gray and white matter from the unified segmentation-normalization. One should specify the input folder (inDir, which is the outDir for bidsHarvest), and output folder (outDir) as well as a path for temporarily storing intermediate images (tempDir).

The Python scripts assume the Matlab scripts have been run. Each file will require the user to specify paths for their file system.

• nii2meanLesion.py : Creates an average lesion map (showing incidence of injury).
• nii2meanFLAIR.py : Creates an average FLAIR image.
• bids_bitmaps.py : Creates one PNG image per participant allowing visual inspection to determine that the normalization performed well.
• lesion2artery.py : creates a tab-separated value file named artery.tsv to identify the proportion of each region of the Arterial Atlas that has been damaged for each individual.
• clean_artery_tsv.py : given the artery.tsv file, create a new tsv file named artery_cleaned.tsv that only contains columns with variability. By default, it uses a 5% sufficient affection threshold.
• clean_participant_tsv.py : given a BIDS format participants.tsv file, create a new file participants_cleaned.tsv that only contains the variables of interest. Without modification, this script will preserve the first, third and fifth columns (subject_id, age and nihss) from participants.tsv in the new file participants_cleaned.tsv.
• merge_artery_tsv.py concatenates the files participants_cleaned.tsv and artery_cleaned.tsv to create merged_artery_participants.tsv. Note that some participants do not have lesions, so only rows where the participant is named in both input files are preserved.
• deep_learn.py : uses lesion and patient age to predict patient's impairment on the NIH Stroke Scale (NIHSS). This script reads the participants_cleaned.tsv file. You can edit this file to set the features that are included and excluded.

By default, the deep_learn.py should report both neural network and support vector regression predictions:

Neural Network - Correlation (R): 0.5142377941206763, p-value: 2.6454265340519674e-62
SVR - Correlation (R): 0.5501800789686954, p-value: 8.097513252055928e-73

Note the performance is numerically somewhat better than what is achieved by uncommenting the line columns_to_keep = ['age_at_stroke', 'lesion_volume', rv]. This constrained model ignores regional injury, and reflects the potency of lesion volume to predict generalized impairment measures. From first principles, one might expect additional features to improve performance as the dataset size is increased and more specific behavioral measures are provided. The performance of this simpler model is:

Neural Network - Correlation (R): 0.49739806569287565, p-value: 8.537857290618983e-58
SVR - Correlation (R): 0.4785154590654683, p-value: 4.973518502919079e-53

NIFTI

These are the derived NIfTI format images created by the lesionNorm script. These can be viewed with many neuroimaging tools, including our web-based drag-and-drop NiiVue.

• wsub-*_FLAIR.nii.gz : anatomical T2-weighted FLAIR MRI scan from each participant warped (w) to standard space and SynthStrip brain-extracted. ** Due to the large file size, these images are not provided on Github. These are available from OSF**
• bwsrsub-*_lesion : map of injury from each individual warped to standard space.
• ArterialAtlas136.nii.gz : arterial atlas of brain regions, in same space as individual images.
• ArterialAtlas136.txt : Text file which provides name and index number for brain regions in the atlas.
• FLAIR_mean_1714.nii.gz : Average FLAIR image for all participants.
• lesion_mean_1449.nii.gz : Average lesion map for all participants.

PNG

These bitmap images are created by the bids_bitmaps.py and aid quality assurance for the image processing.

• wsub-M*.png provide orthogonal slices (axial, sagittal, coronal) and volume rendered images for the warped and brain extracted T1-weighted MRI scan with the lesion shown as translucent red. The cross-hair is positioned at the center of mass for the lesion. Since brain-extraction is derived from the unified normalization-segmentation algorithm, an accurate brain extraction and cortical surface rendering is consistent with an accurate normalization to standard space.

How to Acknowledge

If you use SOOP, please cite our work: update when manuscript reviewed