Skip to content
DEXPRADEXPRA home

Interoperability

DICOM 101: The Standard That Runs Medical Imaging

DICOM answers almost every question an imaging or AI project has, and has done for thirty years. The answers are spread across roughly four thousand pages that nobody reads front to back. This paper is the orientation layer, written for engineers, data scientists and technical product people who need to be productive against real clinical imaging systems without a radiology background. It assumes you can read code and a network diagram, and nothing about medicine. It is also current: most DICOM introductions on the internet were written between 2010 and 2018 and have not been touched since, and a full section is devoted to what the standard has added in the meantime.

What’s inside

  • The four layers of the standard, and why modelling DICOM as a file format breaks a pipeline against a real PACS
  • The data model, the UID rules, data elements, modules, IODs and SOP classes
  • DIMSE association negotiation read from a real service log, and the DICOMweb operations that map onto it
  • From stored values to pixels: rescale, windowing, polarity, bit depth, geometry and frame of reference
  • What changed since 2020: HTJ2K, the rebuilt PS3.18, the retired TLS profiles, and the petabyte inventory service
  • DICOM and AI without the marketing: which objects exist for model output, and which do not
  • How to read a conformance statement, and nine expensive mistakes
  • A glossary, and a full source list checked against PS3 2026c

PDF · 32 pages

Request this resource

Working on one of these problems?

We’re publishing more than we’re consulting right now. If you have something specific in interoperability, data platforms, or clinical AI, tell us what it is and we’ll tell you what we think.

Get in touch