---
title: "DICOM explained: the medical imaging standard | Quantlix"
description: "DICOM is the international standard for storing, sending and viewing medical images. Learn how it works, what DICOMweb adds and how it shapes imaging software."
canonical: "https://quantlix.com/en/standards/dicom/"
last-updated: 2026-11-16
---

Source: https://quantlix.com/en/standards/dicom/
Language: en
Description: DICOM is the international standard for storing, sending and viewing medical images. Learn how it works, what DICOMweb adds and how it shapes imaging software.

**DICOM** [Health and clinical data](https://quantlix.com/en/standards/#health)

# DICOM: the standard for medical images and imaging workflows

DICOM is the international standard for medical imaging that defines how images and related data are encoded, stored, exchanged and displayed, so scanners, archives and viewers from different vendors work together.

[Discuss your project](https://quantlix.com/en/contact/) [Read the key points](https://quantlix.com/en/standards/dicom/#standard-overview)

## At a glance

**Full name**

Digital Imaging and Communications in Medicine (DICOM), published as NEMA PS3 and recognized as ISO 12052

**Published by**

DICOM Standards Committee, with the National Electrical Manufacturers Association (NEMA) as secretariat

**Type**

International standard (data format and network protocol)

**Current version**

Maintained continuously and republished several times a year; edition 2026d was current when we checked in November 2026

**Applies to**

Imaging modalities, PACS, viewers, worklist and reporting systems and imaging AI tools in radiology, cardiology, pathology and other specialties

**Last reviewed**

November 16, 2026 (2026-11-16)

## In plain terms

### Key points

- A DICOM object carries pixel data and a structured header that describes the patient, study, series, equipment and acquisition.
- The standard defines both the file format and the network services that modalities, archives and viewers use to store, query and retrieve images.
- DICOMweb, defined in PS3.18, offers RESTful services for searching, retrieving and storing images over HTTP.
- DICOM is a living standard: edition 2026d was current in November 2026, and changes arrive through supplements and correction proposals.
- Each product declares its support in a DICOM Conformance Statement, the first document to read when you plan an integration.

### What is DICOM?

DICOM (Digital Imaging and Communications in Medicine) is **the international standard for medical imaging information**. It defines how an image and its context are encoded in a file, how imaging systems send and request images over a network and what each product must say about the features it supports. ISO recognizes it as ISO 12052, and the DICOM Standards Committee maintains it, with NEMA as secretariat.

You can see its value on every scan. A CT scanner from one vendor, a PACS (picture archiving and communication system) from another and a viewer from a third can share the same study because they all speak DICOM.

### What DICOM covers

The standard is published in more than 20 numbered parts, PS3.1 to PS3.22.

- **Information objects.** Definitions for each kind of image and related object, such as CT, MR, ultrasound, whole-slide pathology, structured reports and segmentations, in PS3.3.
- **Encoding and the data dictionary.** How attributes, tags and pixel data are encoded, including compressed transfer syntaxes, in PS3.5 and PS3.6.
- **Network services.** The DIMSE services for storing, finding and moving images and for modality worklists, in PS3.4, PS3.7 and PS3.8.
- **Files and media.** The DICOM file format and media profiles in PS3.10 to PS3.12.
- **Web services.** DICOMweb in PS3.18, with RESTful services to search, retrieve and store studies and to manage worklist items.
- **Security and display.** Security and confidentiality profiles in PS3.15 and the grayscale display function in PS3.14, so images look the same on calibrated screens.

### How DICOM is organized and released

Every DICOM object carries a header of tagged attributes arranged in a patient, study, series and instance hierarchy, with unique identifiers (UIDs) that let systems match objects across sites. A service-object pair (SOP) class combines an object type with a service, such as storing a CT image, and is how products state precisely what they can do.

**DICOM changes continuously rather than in big releases.** Working groups develop supplements for new capabilities and correction proposals for fixes, and the whole standard is republished several times a year as a dated edition. Each product publishes a Conformance Statement, structured by PS3.2, listing the SOP classes, transfer syntaxes and network roles it supports. Comparing two conformance statements tells you most of what you need to know before connecting two systems.

### How it shows up in software projects

- **Conformance review first.** Integrations start by comparing the conformance statements of the systems involved, so gaps in SOP classes or transfer syntaxes surface before code is written.
- **UIDs as keys.** Study, series and instance UIDs become the keys for indexing, deduplication and linking images to orders and reports.
- **DICOMweb for new apps.** Web viewers, cloud archives and AI pipelines usually use QIDO-RS, WADO-RS and STOW-RS, while DIMSE stays in place for modalities and older archives.
- **De-identification.** Research and AI datasets apply PS3.15 confidentiality profiles to headers, check pixels for burned-in text and keep a record of every change.
- **Storage and performance.** Large studies need streaming, sensible compression choices and storage tiers planned against real image volumes.

### How we work with DICOM

We build imaging software for [healthcare](https://quantlix.com/en/industries/healthcare/) teams: DICOMweb services and viewers, routing between modalities and archives, de-identification pipelines and data preparation for imaging AI, with cloud archives designed by our [cloud architecture](https://quantlix.com/en/services/cloud-architecture/) team. Where images meet the wider patient record, we connect them to [HL7 FHIR](https://quantlix.com/en/standards/hl7-fhir/) resources, and we design storage and access controls with [HIPAA](https://quantlix.com/en/standards/hipaa/) and [GDPR](https://quantlix.com/en/standards/gdpr/) in mind.

We describe what we deliver plainly. Software we build follows the DICOM objects and services you need, and we test it against the conformance statements of your systems. Regulatory clearance of a medical device or AI tool, where it applies, belongs to its manufacturer and the relevant regulator.

## Where it applies

Industries where this standard often shapes the work.

- ### [Healthcare](https://quantlix.com/en/industries/healthcare/)

  Hospital and clinic workflows, digital health and medical technology

## Related standards

[All standards](https://quantlix.com/en/standards/)

- HL7 FHIR

  ### [Health data exchange over web APIs](https://quantlix.com/en/standards/hl7-fhir/)

  HL7 FHIR is the HL7 International standard for exchanging health data as small, linked resources over web APIs, used by EHRs, patient apps, payers and research platforms.
- HIPAA

  ### [US health information privacy and security](https://quantlix.com/en/standards/hipaa/)

  HIPAA is the US federal law whose Privacy, Security and Breach Notification Rules govern how health plans, providers and their business associates use, disclose and protect patient health information.
- GDPR

  ### [EU personal data protection law](https://quantlix.com/en/standards/gdpr/)

  GDPR is the European Union regulation that sets how organizations collect, use, share and protect personal data about people in the EU, giving individuals clear rights and every controller and processor clear duties.
- EU AI Act

  ### [EU rules for artificial intelligence](https://quantlix.com/en/standards/eu-ai-act/)

  EU AI Act is the common name for Regulation (EU) 2024/1689, the EU law on artificial intelligence. It bans certain practices and sets duties for high-risk systems, general-purpose AI models and transparency.

## Services that support this work

The services we usually bring to projects that involve this standard.

[View all services](https://quantlix.com/en/services/)

### [Enterprise applications](https://quantlix.com/en/services/enterprise-applications/)

Quantlix designs and builds enterprise applications for internal workflows, business rules and system integration, with clear access controls and maintainable architecture.

### [Cloud architecture](https://quantlix.com/en/services/cloud-architecture/)

Quantlix designs cloud infrastructure around workload requirements, security boundaries, operational visibility and cost, with an explicit path for deployment and recovery.

### [Artificial intelligence](https://quantlix.com/en/services/artificial-intelligence/)

Industrial AI for inspection, maintenance and operational decision support. Quantlix scopes data, integration, evaluation and human review around one workflow.

## Questions, answered

What teams usually ask us about this standard.

**What is the latest version of DICOM?**

DICOM is not versioned like most standards. It is republished several times a year as a dated edition, and edition 2026d was current when we checked dicomstandard.org in November 2026. Products are described by the object types and services they support rather than by an edition.

**What is the difference between DICOM and DICOMweb?**

DICOMweb is part of DICOM. Traditional DICOM networking uses its own DIMSE protocol over TCP/IP, while DICOMweb, defined in PS3.18, provides RESTful HTTP services: QIDO-RS to search, WADO-RS to retrieve and STOW-RS to store, plus UPS-RS for worklist items.

**How does DICOM relate to HL7 FHIR?**

They complement each other. DICOM handles images and imaging metadata, while FHIR carries orders, patient records and results, and the FHIR ImagingStudy resource can point to studies held in a DICOM archive.

**Does DICOM cover de-identification?**

Yes. PS3.15 defines attribute confidentiality profiles that list which header elements to remove or replace when images leave clinical care, for example for research or AI training. Text burned into the pixels needs separate checks.

**Can software be DICOM certified?**

There is no official DICOM certification. Vendors publish a Conformance Statement, following PS3.2, that describes what their product supports, and interoperability is shown by testing real systems together, for example at IHE Connectathons.

## Official sources

Read the official text before relying on any detail. What applies to you depends on your role, your location and the version in force.

- [DICOM Standard, current edition](https://www.dicomstandard.org/current)
- [DICOM PS3.1: Introduction and Overview](https://dicom.nema.org/medical/dicom/current/output/html/part01.html)
- [DICOMweb overview](https://www.dicomstandard.org/using/dicomweb)

These guides explain each standard in plain terms. They are not legal advice, and they do not state that any product or organization conforms to a standard.

## Planning software around this standard?

Tell us which systems are involved and what you need to show. We will help you build the requirements in from the start.

[Discuss your project](https://quantlix.com/en/contact/)
