imaging-data-commons

Name: imaging-data-commons
Author: K-Dense-AI

31kstars3.1kforks

How to install imaging-data-commons?

k-dense-ai/scientific-agent-skills/imaging-data-commons

$npx -y skills add K-Dense-AI/scientific-agent-skills --skill imaging-data-commons

Installs into the current project.

Summary

Query and download public cancer imaging data from NCI Imaging Data Commons using idc-index. Use for accessing large-scale radiology (CT, MR, PET) and pathology datasets for AI training or research. No authentication required. Query by metadata, visualize in browser, check licens

Files · 1

View on GitHub

SKILL.md

1	# Imaging Data Commons
2
3	## Overview
4
5	Use the `idc-index` Python package to query and download public cancer imaging data from the National Cancer Institute Imaging Data Commons (IDC). No authentication required for data access.
6
7	Current IDC Data Version: v23 (always verify with `IDCClient().get_idc_version()`)
8
9	Primary tool: `idc-index` ([GitHub](https://github.com/imagingdatacommons/idc-index))
10
11	CRITICAL - Check package version and upgrade if needed (run this FIRST):
12
13	```python
14	import idc_index
15
16	REQUIRED_VERSION = "0.11.14" # Must match metadata.idc-index in this file
17	installed = idc_index.__version__
18
19	if installed < REQUIRED_VERSION:
20	print(f"Upgrading idc-index from {installed} to {REQUIRED_VERSION}...")
21	import subprocess
22	subprocess.run(["pip3", "install", "--upgrade", "--break-system-packages", "idc-index"], check=True)
23	print("Upgrade complete. Restart Python to use new version.")
24	else:
25	print(f"idc-index {installed} meets requirement ({REQUIRED_VERSION})")
26	```
27
28	Verify IDC data version and check current data scale:
29
30	```python
31	from idc_index import IDCClient
32	client = IDCClient()
33
34	# Verify IDC data version (should be "v23")
35	print(f"IDC data version: {client.get_idc_version()}")
36
37	# Get collection count and total series
38	stats = client.sql_query("""
39	SELECT
40	COUNT(DISTINCT collection_id) as collections,
41	COUNT(DISTINCT analysis_result_id) as analysis_results,
42	COUNT(DISTINCT PatientID) as patients,
43	COUNT(DISTINCT StudyInstanceUID) as studies,
44	COUNT(DISTINCT SeriesInstanceUID) as series,
45	SUM(instanceCount) as instances,
46	SUM(series_size_MB)/1000000 as size_TB
47	FROM index
48	""")
49	print(stats)
50	```
51
52	Core workflow:
53	1. Query metadata → `client.sql_query()`
54	2. Download DICOM files → `client.download_from_selection()`
55	3. Visualize in browser → `client.get_viewer_URL(seriesInstanceUID=...)`
56
57	## When to Use This Skill
58
59	- Finding publicly available radiology (CT, MR, PET) or pathology (slide microscopy) images
60	- Selecting image subsets by cancer type, modality, anatomical site, or other metadata
61	- Downloading DICOM data from IDC
62	- Checking data licenses before use in research or commercial applications
63	- Visualizing medical images in a browser without local DICOM viewer software
64
65	## Quick Navigation
66
67	Core Sections (inline):
68	- IDC Data Model - Collection and analysis result hierarchy
69	- Index Tables - Available tables and joining patterns
70	- Installation - Package setup and version verification
71	- Core Capabilities - Essential API patterns (query, download, visualize, license, citations, batch)
72	- Best Practices - Usage guidelines
73	- Troubleshooting - Common issues and solutions
74
75	Reference Guides (load on demand):
76
77	\| Guide \| When to Load \|
78	\|-------\|--------------\|
79	\| `index_tables_guide.md` \| Complex JOINs, schema discovery, DataFrame access \|
80	\| `use_cases.md` \| End-to-end workflow examples (training datasets, batch downloads) \|
81	\| `sql_patterns.md` \| Quick SQL patterns for filter discovery, annotations, size estimation \|
82	\| `clinical_data_guide.md` \| Clinical/tabular data, imaging+clinical joins, value mapping \|
83	\| `cloud_storage_guide.md` \| Direct S3/GCS access, versioning, UUID mapping \|
84	\| `dicomweb_guide.md` \| DICOMweb endpoints, PACS integration \|
85	\| `digital_pathology_guide.md` \| Slide microscopy (SM), annotations (ANN), pathology workflows \|
86	\| `bigquery_guide.md` \| Full DICOM metadata, private elements (requires GCP) \|
87	\| `cli_guide.md` \| Command-line tools (`idc download`, manifest files) \|
88	\| `parquet_access_guide.md` \| Direct Parquet queries via GCS (no idc-index install needed) \|
89
90	## IDC Data Model
91
92	IDC adds two grouping levels above the standard DICOM hierarchy (Patient → Study → Series → Instance):
93
94	- collection_id: Groups patients by disease, modality, or research focus (e.g., `tcga_luad`, `nlst`). A patient belongs to exactly one collection.
95	- analysis_result_id: Identifies derived objects (segmentations, annotations, radiomics features) across one or more original collections.
96
97	Use `collection_id` to find original imaging data, may include annotations deposited along with the images; use `analysis_result_id` to find AI-generated or expert annotations.
98
99	Key identifiers for queries:
100	\| Identi

People who liked this also viewed

6 picks

Type

imaging-data-commons

31kstars3.1kforks

View on GitHub

How to install imaging-data-commons?

k-dense-ai/scientific-agent-skills/imaging-data-commons

$npx -y skills add K-Dense-AI/scientific-agent-skills --skill imaging-data-commons

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Imaging Data Commons
2
3	## Overview
4
5	Use the `idc-index` Python package to query and download public cancer imaging data from the National Cancer Institute Imaging Data Commons (IDC). No authentication required for data access.
6
7	Current IDC Data Version: v23 (always verify with `IDCClient().get_idc_version()`)
8
9	Primary tool: `idc-index` ([GitHub](https://github.com/imagingdatacommons/idc-index))
10
11	CRITICAL - Check package version and upgrade if needed (run this FIRST):
12
13	```python
14	import idc_index
15
16	REQUIRED_VERSION = "0.11.14" # Must match metadata.idc-index in this file
17	installed = idc_index.__version__
18
19	if installed < REQUIRED_VERSION:
20	print(f"Upgrading idc-index from {installed} to {REQUIRED_VERSION}...")
21	import subprocess
22	subprocess.run(["pip3", "install", "--upgrade", "--break-system-packages", "idc-index"], check=True)
23	print("Upgrade complete. Restart Python to use new version.")
24	else:
25	print(f"idc-index {installed} meets requirement ({REQUIRED_VERSION})")
26	```
27
28	Verify IDC data version and check current data scale:
29
30	```python
31	from idc_index import IDCClient
32	client = IDCClient()
33
34	# Verify IDC data version (should be "v23")
35	print(f"IDC data version: {client.get_idc_version()}")
36
37	# Get collection count and total series
38	stats = client.sql_query("""
39	SELECT
40	COUNT(DISTINCT collection_id) as collections,
41	COUNT(DISTINCT analysis_result_id) as analysis_results,
42	COUNT(DISTINCT PatientID) as patients,
43	COUNT(DISTINCT StudyInstanceUID) as studies,
44	COUNT(DISTINCT SeriesInstanceUID) as series,
45	SUM(instanceCount) as instances,
46	SUM(series_size_MB)/1000000 as size_TB
47	FROM index
48	""")
49	print(stats)
50	```
51
52	Core workflow:
53	1. Query metadata → `client.sql_query()`
54	2. Download DICOM files → `client.download_from_selection()`
55	3. Visualize in browser → `client.get_viewer_URL(seriesInstanceUID=...)`
56
57	## When to Use This Skill
58
59	- Finding publicly available radiology (CT, MR, PET) or pathology (slide microscopy) images
60	- Selecting image subsets by cancer type, modality, anatomical site, or other metadata
61	- Downloading DICOM data from IDC
62	- Checking data licenses before use in research or commercial applications
63	- Visualizing medical images in a browser without local DICOM viewer software
64
65	## Quick Navigation
66
67	Core Sections (inline):
68	- IDC Data Model - Collection and analysis result hierarchy
69	- Index Tables - Available tables and joining patterns
70	- Installation - Package setup and version verification
71	- Core Capabilities - Essential API patterns (query, download, visualize, license, citations, batch)
72	- Best Practices - Usage guidelines
73	- Troubleshooting - Common issues and solutions
74
75	Reference Guides (load on demand):
76
77	\| Guide \| When to Load \|
78	\|-------\|--------------\|
79	\| `index_tables_guide.md` \| Complex JOINs, schema discovery, DataFrame access \|
80	\| `use_cases.md` \| End-to-end workflow examples (training datasets, batch downloads) \|
81	\| `sql_patterns.md` \| Quick SQL patterns for filter discovery, annotations, size estimation \|
82	\| `clinical_data_guide.md` \| Clinical/tabular data, imaging+clinical joins, value mapping \|
83	\| `cloud_storage_guide.md` \| Direct S3/GCS access, versioning, UUID mapping \|
84	\| `dicomweb_guide.md` \| DICOMweb endpoints, PACS integration \|
85	\| `digital_pathology_guide.md` \| Slide microscopy (SM), annotations (ANN), pathology workflows \|
86	\| `bigquery_guide.md` \| Full DICOM metadata, private elements (requires GCP) \|
87	\| `cli_guide.md` \| Command-line tools (`idc download`, manifest files) \|
88	\| `parquet_access_guide.md` \| Direct Parquet queries via GCS (no idc-index install needed) \|
89
90	## IDC Data Model
91
92	IDC adds two grouping levels above the standard DICOM hierarchy (Patient → Study → Series → Instance):
93
94	- collection_id: Groups patients by disease, modality, or research focus (e.g., `tcga_luad`, `nlst`). A patient belongs to exactly one collection.
95	- analysis_result_id: Identifies derived objects (segmentations, annotations, radiomics features) across one or more original collections.
96
97	Use `collection_id` to find original imaging data, may include annotations deposited along with the images; use `analysis_result_id` to find AI-generated or expert annotations.
98
99	Key identifiers for queries:
100	\| Identi

People who liked this also viewed

6 picks

Type