A DICOM image is a medical image file that follows a universal standard called Digital Imaging and Communications in Medicine. Unlike a regular photo saved as a JPEG or PNG, a DICOM file packages the image together with detailed information about the patient, the scan, and the equipment that captured it. Every CT scan, MRI, X-ray, and ultrasound produced by modern imaging equipment is saved in this format, making it the backbone of how hospitals store, share, and display medical images.
Where DICOM Came From
In 1982, the American College of Radiology and the National Electrical Manufacturers Association formed a joint committee to solve a practical problem: imaging devices from different manufacturers couldn’t talk to each other. A CT scanner made by one company produced files that a workstation from another company couldn’t read. Version 3.0 of their standard, released in the early 1990s, became what we now call DICOM. It established a common language so that any compliant device, regardless of manufacturer, can produce images that any other compliant device can open and display. The standard is still maintained today by the Medical Imaging & Technology Alliance, a division of NEMA.
What Makes It Different From a JPEG or PNG
Consumer image formats like JPEG and PNG were designed for photos and web graphics. They typically store 8 bits of data per pixel, giving you 256 shades of gray. That’s fine for a vacation photo but not for a radiologist trying to distinguish subtle tissue differences. Medical imaging sensors can capture 12 or even 16 bits per pixel, producing thousands of shades of gray. DICOM preserves that full range of data without stripping it away.
The difference shows up in measurable image quality. A 2023 study in the Brazilian Dental Journal compared DICOM files against TIFF, BMP, PNG, and JPEG versions of the same dental radiographs. The DICOM files had lower image noise, higher brightness, and greater image uniformity than every other format tested. The non-DICOM formats performed about the same as each other, but none matched DICOM’s quality on any of those measures.
Beyond image quality, DICOM files carry something consumer formats simply don’t: a rich set of structured medical data embedded right alongside the pixel information.
What’s Inside a DICOM File
Every DICOM file has two main parts: a header and the image data itself. The header begins with a fixed 128-byte preamble, followed by four bytes containing the characters “DICM,” which is how software confirms it’s looking at a real DICOM file. After that come the metadata elements, organized into standardized tags.
Those tags carry a surprising amount of information, grouped into categories:
- Patient information: name, date of birth, ID number
- Study details: the date and time of the exam, the referring physician, the reason for the scan
- Series information: which body part was examined, the anatomical plane (axial, sagittal, or coronal), and a text description of the scan sequence
- Equipment data: the manufacturer, model, and technical settings used during image capture
- Pixel data characteristics: whether the image is grayscale or color, its dimensions, and how the pixel values should be interpreted
Each tag is identified by a pair of numbers. For example, tag (0018,0015) stores which body part was examined, and tag (0020,0037) records the orientation of the patient in the image. This structured tagging system is what allows hospital software to automatically sort, search, and organize thousands of images across patients and exams. It’s worth noting that these tags aren’t always filled in perfectly. One large study of brain MRI data found that the “Body Part Examined” field was incorrectly filled about 15% of the time, which is why radiologists don’t rely on metadata alone.
Which Types of Scans Use DICOM
Virtually every modern medical imaging modality produces DICOM files. The standard covers CT, MRI, conventional X-ray (computed radiography), ultrasound, nuclear medicine, X-ray angiography, and fluoroscopy. It has since expanded to include PET scans, radiation therapy planning, visible light imaging (such as endoscopy and pathology photos), and structured reporting. If a hospital acquires a new imaging device today, it is expected to output DICOM-compliant files.
How Hospitals Store and Manage DICOM Images
The standard defines the file format, but hospitals need a system to actually store and retrieve those files at scale. That system is called PACS, short for Picture Archiving and Communication System. Think of DICOM as the language and PACS as the library. When a technologist performs a CT scan, the scanner formats the images as DICOM files, then sends them over the hospital network to PACS, where they’re archived. When a radiologist or surgeon needs to review the images, they pull them from PACS to a viewing workstation.
DICOM also defines how images travel between systems. Files can be sent over traditional network connections using a protocol called DIMSE, or through web-based services similar to how your browser loads a webpage. This flexibility means images can move between departments, hospitals, or even across the country as long as both ends speak DICOM.
How DICOM Protects Patient Privacy
Because every DICOM file contains patient names, birth dates, and other identifying information, the standard includes built-in privacy and security features. Since 1999, DICOM has supported encryption for data moving over network connections using TLS, the same technology that secures online banking. For web-based transfers, it uses HTTPS.
The standard also allows sensitive portions of a DICOM file, specifically the protected health information, to be encrypted within the file itself. This means a DICOM image can be stored or emailed with its patient data locked behind encryption while the pixel data remains accessible. When the goal is to encrypt the entire file (not just the patient data), DICOM defers to external encryption tools but is designed to work smoothly alongside them. For research purposes, DICOM files are often “de-identified,” meaning the patient-specific tags are stripped out or replaced before sharing.
How to Open DICOM Files
If you’ve received a CD or USB drive after a medical scan, the images on it are almost certainly in DICOM format. You can’t open them with your computer’s default photo viewer. You need a DICOM viewer.
Several free options exist. MicroDicom is a widely used free viewer for Windows that includes standard tools for adjusting brightness, contrast, and zoom. It’s designed to run directly from a CD or USB drive without installation, which is useful because many imaging centers include a portable viewer right on the patient disc. You insert the disc, the viewer launches automatically, and your images appear.
Other popular free viewers include Horos (for Mac), RadiAnt (Windows, free for non-commercial use), and 3D Slicer (cross-platform, often used for research). Most of these let you scroll through slices of a CT or MRI, adjust window and level settings to highlight different tissues, and measure distances within the image. For clinical use, hospitals rely on professional-grade viewers integrated into their PACS, which offer more advanced tools like 3D reconstruction and side-by-side comparison with prior exams.
Compression and File Size
DICOM files can be large. A single CT scan of the chest might contain 300 to 500 individual slices, each its own DICOM file, and the total study can reach several hundred megabytes. To manage storage and transmission, the standard supports multiple compression methods. The default transfer format is uncompressed, and every DICOM-compliant system is required to support it. Beyond that, systems can negotiate lossless compression (which shrinks the file without losing any data) or lossy compression (which sacrifices some detail for smaller file sizes). Lossy compression is occasionally used for teaching or referral copies, but diagnostic-quality images are typically stored lossless or uncompressed to preserve every bit of information the scanner captured.