Releases: ImagingDataCommons/idc-claude-skill
Releases · ImagingDataCommons/idc-claude-skill
v1.4.0
[1.4.0] - 2026-03-04
Added
- New "Identifying Tumor vs Normal Slides" section in digital pathology guide with two approaches:
- Structured DICOM tissue type via
primaryAnatomicStructureModifier_CodeMeaning(works across all SM collections) - TCGA barcode parsing via
ContainerIdentifier(TCGA collections only, catches metastatic edge cases)
- Structured DICOM tissue type via
- TCGA-BRCA worked examples showing tumor vs normal slide counts
- Documentation references to GDC TCGA barcode format and sample type codes
- Specimen preparation query examples: filtering by staining (H&E), embedding medium (FFPE vs frozen), and fixative, with note about array column syntax (
array_to_string,list_contains) - "Finding Pre-Computed Analysis Results" section: discovering derived datasets (nuclei segmentations, TIL maps) via
analysis_results_index, with example joining annotations back to source slides - Note about per-annotation measurements in DICOM ANN objects (extractable via highdicom after download), with link to microscopy_dicom_ann_intro tutorial
v1.3.1
[1.3.1] - 2026-02-11
Added
- Automatic idc-index package version check with upgrade prompt before any queries
- Version check compares installed version against
metadata.idc-indexin frontmatter and triggerspip install --upgradewhen outdated
Fixed
- Prevents "table not found" errors when using newer index tables (e.g.,
contrast_index) with older idc-index versions
v1.3.0
Added
- Digital pathology reference guide (
references/digital_pathology_guide.md) with SM, ANN, and SEG query patterns, join examples, and pathology tool recommendations seg_indexcoverage in digital pathology guide with cross-domain clarification (SEG used for both radiology and pathology) and query for finding pathology-specific segmentationsAnnotationGroupLabelfiltering examples for finding annotation groups by name- SM + ANN cross-reference queries showing how to find annotations on slide microscopy images
- Index discovery guidance before BigQuery section to ensure all local indices are checked first
- Documentation for new
ann_indexandann_group_indextables (Microscopy Bulk Simple Annotations) - Example queries for annotation series and annotation group metadata
- Explanation of downloaded DICOM file naming convention (
<crdc_instance_uuid>.dcm) - New reference guides extracted from SKILL.md:
references/index_tables_guide.md- Table schemas, DataFrame access, join column referencereferences/sql_patterns.md- Quick-reference SQL patterns for common queriesreferences/use_cases.md- End-to-end workflow examples
- Quick Navigation section in SKILL.md with decision triggers for when to load each reference
idc-data-versionfield in frontmatter metadata- Documentation for new
contrast_indextable (contrast bolus metadata for CT, MR, PT, XA, RF series)
Changed
- Updated to idc-index 0.11.9 (IDC data version v23)
- Reduced SKILL.md from 1,245 to 825 lines by extracting secondary content to reference files
- Core Capabilities sections remain inline to ensure correct API pattern usage
- Refactored detailed SM/ANN content from SKILL.md into
references/digital_pathology_guide.md, keeping brief summaries with pointers in main skill - Made IDC version (v23) more prominent in SKILL.md with verification guidance to prevent responses using older versions
- Clarified distinction between
index_tables_guide.md(structure/access) andsql_patterns.md(query examples)
Full Changelog: v1.2.0...v.1.3.0