What “first 3D scanner” can mean
No single machine was ever announced as the first 3D medical scanner, so the title needs unpacking. The history is really a shift from images that flatten the body into one shadow to images built from many thin slices, which software can stack into a volume. Several people and several decades are involved, and different histories choose different starting points.
The problem with flat images
A conventional X-ray shows everything along the beam path superimposed. A rib, a lung nodule and the spine can all land on the same spot of film. Radiologists learned to infer depth from experience and by taking views from different angles, but the picture remained a projection.
Early tomography: sharpening one plane
Between the 1920s and 1930s, several investigators in Europe and elsewhere independently proposed moving the X-ray tube and film in opposite directions during exposure. Structures in one chosen plane stayed sharp while others blurred. This was called by various names, including tomography, planigraphy and stratigraphy, and credit is shared among people such as Alessandro Vallebona and Bernard Ziedses des Plantes. It was an analog trick, with no computer involved. The blurred layers were still present, so image quality was limited.
The mathematical key
A 1917 paper by the Austrian mathematician Johann Radon showed that a function in a plane can in principle be recovered from its line integrals, meaning the sums along all lines through it. That is exactly the situation when X-rays pass through a slice from many directions. The result was little used in medicine at the time. Later, Allan Cormack in the early 1960s and others rediscovered or developed related methods, and William Oldendorf built an experimental apparatus in 1961 that demonstrated the idea on a model.
Computed tomography changes the picture
In the early 1970s Godfrey Hounsfield at EMI built a working scanner that collected many transmission readings around the head and let a computer calculate a cross-sectional image. Each picture was a single slice built from a grid of small elements. Scanning several slices in sequence let doctors look through an organ one level at a time. This is the strongest candidate for the “first” clinical slice-based scanner, though it was a 2D slice device rather than a true volume imager.
Why computing was the bottleneck
Early reconstructions were slow. The first scanners took minutes to collect data for each slice, and the calculation then needed substantial processing time on the minicomputers of the day. Memory was small, and images were coarse by modern standards. Improvements came from faster detectors, better algorithms such as filtered back-projection, and cheaper and faster electronic computers. Around 1990 spiral scanning, in which the table moves continuously while the tube rotates, made it practical to collect an entire volume in one pass. Multi-detector designs later added more rows and faster coverage.
From slices to volumes
Once data exist as a stack of slices, software can resample them to show other planes, such as side views, or draw surfaces and see-through volumes. This took off in the 1980s and 1990s as graphics hardware improved, which is why graphics cards and medical imaging have a shared history. MRI naturally produces data in multiple planes, using mathematical transformations on magnetic-resonance signals. Ultrasound and PET have also developed 3D forms.
Beyond viewing
Three-dimensional data also support making physical objects. Some hospitals use scan data to produce anatomical models through additive manufacturing, helping surgeons plan or teach, and early researchers applied similar thinking in the age of rapid prototyping. These models are aids, not replacements for the scan itself.
Where the credit sits
If you want one name, Hounsfield and Cormack are the usual answers for slice-based reconstruction. But tomography pioneers of the 1930s, Radon’s mathematics and the computer engineers who made reconstruction fast enough to be useful are all part of the story. Treat any single “first” with caution.