dicom index

Extracts DICOM information from files into a single sqlite file

Installation

pip install dicomindex

Usage

Index a DICOM in folder

To recursively read all DICOM files in /path/to/folder and add patient/study/series objects to index.

dicomindex index_full /tmp/myindex.sql /path/to/folder

Index faster

For big file collections over slow connections, indexing all files can easily take weeks. Annoying. You could use this to for a faster index:

dicomindex index_per_folder /tmp/myindex.sql /path/to/base_folder

This will find a single DICOM file to index in each folder recursively. This speeds up indexing significantly, but assumes base_folder contains sub-folders that are structured in a patient/series/study way.

Explore

I've indexed this folder. Now what?

What's in my index file? (quick overview)

dicomindex stats /path/to/folder

Detailed exploration - Datasette

• Install datasette
• From commandline, run datsette /tmp/myindex.sql

Detailed exploration - Python

Query the db file directly using dicomindex sqlalchemy ORM:

from dicomindex.orm import Study
from dicomindex.persistence import SQLiteSession

with SQLiteSession("/tmp/archive.sql") as session:
    studies = session.query(Study).all()
    print(f"found {len(studies)} studies")

Non-goals

dicomindex will not do the following:

• Comprehensive storage. Use a PACS like orthanc. If an ImageComments element contains many pages of data, dicomindex will only save the first 1024 characters. It maps many different DICOM Value Representations to 'flat string'. It is meant for indexing, not faithful reproduction.
• Exploration and browsing tools. Use datasette

Useful for

• DICOM data exploration and sanitizing legacy archives
• Messy, semi-structured data, 'archives' that are just collections of folders with DICOM files in them