How Radon's 1917 Transform Powers Modern CT Scanners

In 1917, Austrian mathematician Johann Radon introduced a pure mathematical theory known as the Radon transform, decades before the invention of digital computers or advanced medical imaging. This article provides an overview of how Radon’s abstract calculus—which proved that a two-dimensional object could be reconstructed entirely from an infinite number of one-dimensional projections—became the exact mathematical blueprint for the modern Computed Tomography (CT) scanner. We will explore the mechanics of the Radon transform, how X-ray beams physically execute this math, and how the inverse transform turns raw radiation data into life-saving images of the human body.

The Flaw of Traditional X-Rays

To understand why Radon’s math was necessary, you first have to understand the limitation of a standard X-ray. A traditional X-ray compresses a three-dimensional human body into a flat, two-dimensional shadow. Because everything in the path of the X-ray beam overlaps, dense bones easily hide soft tissues, tumors, or blood clots located behind them.

To see inside the body clearly without surgery, doctors needed a way to view a single, flat cross-section—or "slice"—of a patient. The challenge was figuring out how to map the interior of a slice simply by shooting X-rays through its edges from the outside.

What is the Radon Transform?

Johann Radon solved this exact problem strictly as a mathematical puzzle in 1917. He asked: if you have an unknown two-dimensional shape with varying internal densities, can you map its interior just by measuring the total density along straight lines passing through it?

Radon proved that you could. In mathematics, adding up values along a continuous straight path is called a "line integral." The Radon transform takes a 2D object and calculates the line integrals of all possible straight lines passing through that object from every possible angle (from 0 to 360 degrees).

If you collect all of these line measurements from every angle, you generate a complete set of projection data. More importantly, Radon proved that this process is reversible.

How a CT Scanner Executes Radon’s Math

A modern CT scanner is essentially a physical machine built to execute the Radon transform on the human body. Here is how the physics of the machine maps directly to Radon’s 1917 paper:

  • The Unknown Object: A thin cross-sectional slice of the patient's body.
  • The Straight Line: A narrow X-ray beam shot through the side of that body slice.
  • The Line Integral: As the X-ray travels through the body, tissues and bones absorb some of the radiation. A detector on the opposite side measures how much X-ray energy survived the trip. The amount of lost energy equals the total density of the tissues along that exact straight line. This single measurement is one of Radon's line integrals.
  • All Possible Angles: The CT scanner spins its X-ray tube and detectors in a full circle around the patient, taking thousands of these line measurements from every angle.

When the scan is finished, the computer does not yet have a picture of the patient's organs. Instead, it has a grid of raw data representing the Radon transform of the patient's body slice, known in radiology as a "sinogram."

The Inverse Radon Transform: Building the Image

Having the sinogram is only half the battle; doctors cannot diagnose a patient by looking at raw projection data. To turn the data into a recognizable picture of organs and bones, the CT scanner must perform the Inverse Radon Transform.

In his 1917 paper, Radon provided the exact formula to reverse his transform. If you know the values of the line integrals from all angles, you can calculate the exact density of any single point inside the object.

In modern CT scanners, computers apply Radon's inverse math using an algorithm called Filtered Back Projection:

  1. Back Projection: The computer takes the density measurement from a single X-ray line and smears that value back across a blank digital canvas along the exact angle the X-ray traveled.
  2. Repeating the Process: It repeats this smearing process for thousands of measurements across all 360 degrees. Where dense objects (like bone) exist, the smeared lines overlap and reinforce each other, creating a bright spot on the screen.
  3. Filtering: Simply overlapping the lines creates a blurry star-like artifact. Using calculus derived from Radon's inverse formula, the computer applies a mathematical filter that cancels out the blurriness, leaving behind a razor-sharp image of the internal anatomy.

From Abstract Theory to the Nobel Prize

When Johann Radon published his formula, he had no practical application in mind, and there were no computers capable of calculating the millions of equations required to reconstruct an image. His work sat largely unnoticed by the medical community for over fifty years.

In the late 1960s and early 1970s, physicist Allan Cormack and engineer Godfrey Hounsfield independently rediscovered the principles of Radon's math. Combining this calculus with newly invented minicomputers and X-ray technology, they built the first working CT scanners. They were awarded the 1979 Nobel Prize in Physiology or Medicine, cementing Johann Radon’s 1917 abstract math as one of the most important foundations in modern medical history.