// blog

DICOM Photometric Interpretation: MONOCHROME1 vs MONOCHROME2

Photometric Interpretation (0028,0004) tells a viewer how to read the pixel values. MONOCHROME2 means low values are black and high values white; MONOCHROME1 means the opposite, and is common on computed radiography X-rays. Color images use RGB, YBR_FULL, YBR_FULL_422 (typical for JPEG) or PALETTE COLOR. A viewer that ignores MONOCHROME1 shows X-rays inverted; the inversion belongs after the window.

Last updated 2 min read

MONOCHROME1 vs MONOCHROME2: the same chest X-ray shown correctly and shown inverted because the tag was ignored

Open some chest X-rays in a naive viewer and the bones come out black and the lungs white, like a film negative. The pixels are fine. The viewer just ignored Photometric Interpretation.

What the tag says

(0028,0004) Photometric Interpretation is a short code string. For gray images there are two:

Value Lowest value shows as Where you’ll meet it
MONOCHROME2 Black CT, MR, most images
MONOCHROME1 White Many computed radiography (CR) X-rays

Our two chest X-ray samples show both: the CR one is stored as MONOCHROME1, the DX one as MONOCHROME2. The image at the top is the CR X-ray shown correctly, and shown as a viewer that ignores the tag would.

Color values

Value Meaning
RGB Red, green, blue
YBR_FULL Brightness plus two color-difference channels; our ultrasound sample uses it
YBR_FULL_422 The same with half the color resolution; typical inside JPEG-compressed files
PALETTE COLOR Each value is an index into a color table stored in the file

Color images also have Samples per Pixel (0028,0002) = 3, and Planar Configuration (0028,0006) says whether the channels are interleaved (RGBRGB…) or stored one plane after another.

Where inversion goes

For gray images the order is:

  1. Stored value → rescale (to HU on CT).
  2. Window center and width → a gray level from black to white.
  3. If MONOCHROME1, invert that gray level.

Inverting the stored values at read time instead looks the same at first, but then the window values from the file (Window Center and Width) no longer match the data. Our viewer builds one lookup table per window that includes the inversion, so it costs nothing per pixel.

And the Invert button?

That’s for the person looking, not for the file. A user may want to see an image the other way round; the viewer applies that on top of what the tag says.

Key takeaways

  • (0028,0004): MONOCHROME2 low = black, MONOCHROME1 low = white.
  • CR X-rays are often MONOCHROME1; handle it or they look inverted.
  • Invert after rescale and window, not when reading.
  • Color: RGB, YBR_FULL, YBR_FULL_422, PALETTE COLOR; check Samples per Pixel and Planar Configuration.

Compare the two X-rays in the browser →

We’re glad to share what we know. When you want a viewer or portal of your own, we’ll put all of it to work on your project, and because we build with AI-assisted development, it costs far less than you might expect. Let’s talk.

Frequently asked questions

Why would anyone store an image as MONOCHROME1?

It matches how film and some detectors work, where more exposure means darker. The standard allows both, so viewers have to handle both.

Should I invert the pixel values when reading the file?

Better not. Keep the values as stored, apply rescale and window, and invert the displayed gray at the end. That way window values from the file still mean what they say.

What's YBR_FULL?

A color encoding with brightness (Y) and two color-difference channels (Cb, Cr), like many video and JPEG formats. A viewer converts it to RGB for the screen.

Sources

  1. DICOM PS3.3: Image Pixel Description (Photometric Interpretation)