Illustration for the Image Acquisition ARRT category

Image Production · ARRT 2025

Digital Image Processing

The seven-stage processing chain, window width vs window level, histogram analysis and LUTs, the deviation index scale, and the PACS to MIMPS rename on the 2027 ARRT exam.

39 lessons 16 sections 8 key terms

Overview

Nothing on the monitor came off the detector looking like that. Between the exposure and the picture sits a chain of operations, some automatic and some yours, and the ARRT exam mostly tests whether you know which is which. The chapter runs on one sequence: readout, digitizing, pre-processing, histogram analysis, rescaling and the look-up table, post-processing, presentation. Pre-processing is the machine; post-processing is the person.

Digitizing is two processes, sampling and quantization. A 14-bit digital radiograph records 16,384 shades of gray while the human eye separates about 30, and windowing hands those 30 visible shades to whichever slice of the recorded range you choose. Window width sets contrast, narrow meaning short scale and high contrast; window level sets brightness.

Histogram analysis is automatic but runs on what you gave it. It wants four clearly collimated margins and rarely tolerates fewer than three, and the wrong anatomical menu applies the wrong reference histogram and look-up table. Rescaling then drags the anatomy onto the display range whether you used 4 mAs or 40, which is why exposure is judged from the deviation index: 0 is on target, and the scale is logarithmic, so +3 is roughly twice the target exposure and -3 roughly half.

The chapter ends at the archive. Lossless compression up to 3:1 is generally considered acceptable; lossy can pass 100:1 but discards data, which is why it is unacceptable for archiving mammography. DICOM is the standard that lets different manufacturers' systems exchange images, not a storage device, and the FDA renamed PACS to MIMPS in 2021, the name the 2027 specifications use.

What you’ll learn in this chapter

The 39 lessons in this chapter break down as follows. The full lesson content is unlocked when you start a free account.

Image Production

  1. Digital Image Processing

The pipeline

  1. The Processing Chain

The image grid

  1. Pixels, Matrix and the DEL
  2. Quiz

Digitizing

  1. Analog to Digital Conversion
  2. Quiz

Pre-processing

  1. Pre-processing the Detector
  2. Quiz

Raw data

  1. Raw, For-Processing, For-Presentation
  2. Quiz

Histogram analysis

  1. The Image Histogram
  2. Quiz
  3. Rescaling and the Look-Up Table
  4. How Histogram Analysis Gets Broken
  5. Quiz

Post-processing

  1. The Post-Processing Menu
  2. Quiz

Windowing

  1. Window Width and Window Level
  2. Quiz

Post-processing

  1. Edge Enhancement and Smoothing
  2. Quiz
  3. Cropping, Masking and Stitching
  4. Quiz

Annotation & ROI

  1. Annotation and the Region of Interest
  2. Quiz

Technique feedback

  1. Exposure Indicator and Deviation Index
  2. Quiz

Storage

  1. Compression and the Archive
  2. Quiz

Critique

  1. Critiquing a Processed Image

Informatics

  1. PACS, DICOM and the 2027 Rename
  2. The Systems Around the Archive
  3. Quiz

Recap

  1. What to Carry Out of This Chapter
  2. Quiz
  3. Quiz
  4. Quiz
  5. Quiz
  6. Quiz

Key terms in this chapter

These are the 8 terms most likely to appear on the ARRT registry from this chapter. Use them as a flashcard pre-quiz.

Window Width
The number of gray levels assigned to the displayed image, which sets image contrast. A narrow width gives short scale, high contrast; a wide width gives long scale, low contrast.
Window Level
The location on the digital number scale where the grays are assigned. It regulates the brightness of the displayed image and selects which tissue range you are looking at.
Look-Up Table (LUT)
A stored mathematical mapping that converts each image input value to a different output value. There is a different LUT for each anatomical projection.
Rescaling
Shifting the recorded values of interest onto the display range. It compensates for over- and underexposure, which is why a technique error can still produce a normal-looking image.
Deviation Index
How far the exposure landed from the target for that examination. Zero is on target, and the scale is logarithmic: +3 is roughly twice the target exposure, -3 roughly half.
Flat Fielding
An automatic pre-processing correction that equalizes the response of every pixel to a uniform x-ray beam, canceling effects like the heel effect gradient on the receptor.
DICOM
Digital Imaging and Communications in Medicine, the ACR and NEMA standard that lets different manufacturers' equipment exchange images. It stores nothing; the archive does.
MIMPS
Medical Image Management and Processing System, the FDA's 2021 rename of the PACS category. The 2027 ARRT specifications use the new name for the same system.

Sample practice question: Image Acquisition

One free sample from the 98-question Image Acquisition bank. See the format, the rationale style, and the difficulty before you sign up.

An adult chest x-ray is performed at 80 kVp and 4 mAs at 72 inches SID. To maintain density at 40 inches SID, what new mAs is required?

  1. A. 1 mAs
  2. B. 2 mAs
  3. C. 1.2 mAs
  4. D. 13 mAs
Show answer and rationale

A, Incorrect: 1 mAs would result in significant underexposure. Apply the density maintenance formula.

B, Incorrect: Distance is decreasing, so mAs should decrease, but not by half.

C, Correct: Correct. Density maintenance formula: mAs₂ = mAs₁ × (SID₂² / SID₁²) = 4 × (40² / 72²) = 4 × (1600/5184) = 4 × 0.309 ≈ 1.2 mAs. Closer distance means more intensity, so less mAs is needed.

D, Incorrect: 13 mAs would more than triple the exposure. Reversed formula error, increasing distance requires more mAs, not less.

See more Image Acquisition questions →

Read 3 chapters free.

The free tier unlocks 3 complete chapters, 50 practice questions, and 1 sample timed exam. No credit card required.

Frequently asked questions

What does the ARRT Radiography Image Production category cover?

Nothing on the monitor came off the detector looking like that. Between the exposure and the picture sits a chain of operations, some automatic and some yours, and the ARRT exam mostly tests whether you know which is which. The chapter runs on one sequence: readout, digitizing, pre-processing, histogram analysis, rescaling and the look-up table, post-processing, presentation. Pre-processing is the machine; post-processing is the person.

How many lessons are in the Digital Image Processing chapter?

This chapter contains 39 lessons across 16 sections, plus a knowledge-check quiz at the end. The full lesson content is unlocked with a Premium subscription. The free tier includes 3 complete chapters.

Is this chapter aligned with the ARRT 2025 Content Specifications?

Yes. Every chapter on this site maps directly to the ARRT Radiography Content Specifications effective 2025. This chapter falls under the Image Production domain of the official ARRT exam blueprint.

Supporting tools

Report a bug

Tell us what's wrong. We'll take a look.