Skip to contents

Checks a project created with sci_create_project() and returns the plan of processing steps, each with the function that performs it:

  1. DICOM metadata extraction (extract_dicom_metadata())

  2. Anonymization, if requested (anonymize_dicom())

  3. DICOM to NIfTI conversion (convert_batch())

  4. Spinal cord segmentation (sct_segment_sc())

  5. Lesion segmentation with SCIsegV2 (sct_segment_lesion())

  6. Vertebral labeling (sct_label_vertebrae())

  7. Parameter extraction (sct_analyze_lesion(), sct_process_segmentation())

  8. Merge & export (save_table(), save_rds())

Usage

sci_run_pipeline(
  project_dir,
  steps = 1:8,
  anonymize = FALSE,
  overwrite = FALSE,
  verbose = TRUE
)

Arguments

project_dir

Character. Path to workflowr project root.

steps

Integer vector. Which steps to run. Default 1:8.

anonymize

Logical. Run anonymization step. Default FALSE.

overwrite

Logical. Kept for compatibility; not used while the runner only plans steps. Default FALSE.

verbose

Logical. Print progress. Default TRUE.

Value

Tibble with one row per requested step: step, name, status ("pending" or "skipped") and message (the function to run, or the reason for skipping).

Details

sci_run_pipeline() does not run the external tools itself: every requested step is reported as "pending" until you run its function. Use sci_pipeline_status() to see which outputs already exist.

Examples

proj <- file.path(tempdir(), "pipeline-study")
sci_create_project(proj, open = FALSE)
#> 
#> ── Creating SCI project: SCI Imaging Analysis ──────────────────────────────────
#> ✔ Directory structure created.
#> ✔ Templates created.
#> ✔ Project created at /tmp/Rtmp5ydOUY/pipeline-study
#> ℹ Next steps:
#>   1. Place DICOM files in data/raw/dicom/
#>   2. Fill in data/metadata/imaging_registry.csv
#>   3. Fill in data/metadata/clinical_data.csv
#>   4. Plan the processing steps with `sci_run_pipeline()`
sci_run_pipeline(proj, steps = 1:3)
#> 
#> ── Step 1: DICOM metadata extraction ──
#> 
#> ℹ Run `extract_dicom_metadata()` for this step.
#> Step 2 (Anonymization): skipped (anonymize = FALSE)
#> 
#> ── Step 3: DICOM to NIfTI conversion ──
#> 
#> ℹ Run `convert_batch()` for this step.
#> 
#> ── Pipeline Summary ──
#> 
#> 2 pending, 1 skipped
#> # A tibble: 3 × 4
#>    step name                      status  message                     
#>   <int> <chr>                     <chr>   <chr>                       
#> 1     1 DICOM metadata extraction pending run extract_dicom_metadata()
#> 2     2 Anonymization             skipped anonymize = FALSE           
#> 3     3 DICOM to NIfTI conversion pending run convert_batch()         
unlink(proj, recursive = TRUE)