mriqc

MRIQC image processing pipeline

---

MRIQC processes the participants and produces image quality metrics from T1w, T2w and BOLD data.

---

MRIQC

• Use run_mriqc.py to run MRIQC pipeline directly or wrap the script in an SGE/Slurm script to run on cluster

python run_mriqc.py --global_config CONFIG.JSON --subject_id 001 --output_dir OUTPUT_DIR_PATH

• Mandatory: Pass in the absolute path to the configuration containing the MRIQC container and data directory to global_config
• Mandatory: Pass in the subject id to participant_id
• Mandatory: Pass in the subject id to session_id
• Mandatory: Pass in the absolute path to the output directory to output_dir

Note

An example config is located here

Sample cmd:

python run_mriqc.py \
    --global_config GLOBAL_CONFIG \
    --participant_id SUBJECT_ID \
    --output_dir OUTPUT_DIR \
    --session_id SESSION_ID

Note

A run for a participant is considered successful when the participant's log file reads Participant level finished successfully

Evaluate MRIQC Results

• Use mriqc_tracker.py to determine how many subjects successfully passed through the MRIQC pipeline
  • Mandatory: Pass in the subject directory as an argument
• After a successful run of the script, a dictionary called tracker_configs is returned contained whether the subject passed through the pipeline successfully

Note

Multiple sessions can be evaluated, but each session will require a new job running this script

Sample cmd:

>>> results = {"pipeline_complete': mriqc_tracker.eval_mriqc(subject_dir, session_id)}
>>> results
 SUCCESS
>>> results = {"MRIQC_BOLD': mriqc_tracker.check_bold(subject_dir, session_id)}
>>> results
 FAIL