Skip to content

Commit 6ab7564

Browse files
authored
DOC: add documentation for clinical_index (#137)
1 parent 4bd2632 commit 6ab7564

File tree

1 file changed

+31
-0
lines changed

1 file changed

+31
-0
lines changed

docs/column_descriptions.md

Lines changed: 31 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -116,3 +116,34 @@ the attributes exist with suffix `_code_designator_value_str` and
116116
and CodeValue, or CodeMeaning. If this is new to you, a brief explanation on the
117117
three-value based coding scheme in DICOM can be found at
118118
https://learn.canceridc.dev/dicom/coding-schemes.
119+
120+
## `clinical_index`
121+
122+
Many of the image collections available in IDC are accompanied by clinical data.
123+
Such clinical data is organized in one or more tables that are shared alongside
124+
the images.
125+
126+
Each row in `clinical_index` corresponds to a column in a clinical table
127+
available in IDC. You can use this index to find collections that have a
128+
specific clinical attribute, compare availability of the clinical data across
129+
collections, identify patients that have specific clinical characteristics.
130+
131+
Note that IDC does not perform any harmonization of the clinical data across
132+
collections, or any validation of the content of the tables. We share clinical
133+
data as it was provided by the submitter.
134+
135+
provides the list of all of the columns across all of the clinical tables
136+
available in IDC. It contains the following items:
137+
138+
- `collection_id`: identifier of the collection where the given clinical data
139+
attribute is available
140+
- `short_table_name`: name of the clinical data table where the attribute is
141+
encountered; the referenced table can be loaded into a Pandas DataFrame using
142+
the `IDCClient.get_clinical_data()` call
143+
- `table_name`: fully resolved name of the table in IDC Google BigQuery public
144+
dataset (only relevant if you would like to search using BigQuery)
145+
- `column`: name of the column that is available in the given clinical table
146+
- `colum_label`: label of the column (this field may contain more extensive
147+
information describing a given column)
148+
- `values`: set of values defining the content of the column (relevant if the
149+
column contains fixed list of values and not free text)

0 commit comments

Comments
 (0)