The History of the CT Scanner: The Cut Without a Knife
They strap her into the chair and turn it over.
She is twenty-five. She blacked out at the top of a stairway, and the question is a brain tumor. To answer it they put a needle in near her spine, draw off the fluid the brain floats in, and push air into the space. Air shows on a photographic plate. Fluid does not. The chair turns her so the bubble travels where they need it.
The headache arrives when she moves her head. It arrives every time she moves her head, for a week, while she lies flat and waits for the telephone.
The answer is no tumor. Probably encephalitis. He cannot be certain.
The skull is a closed box. Bone stops X-rays. Brain does not. On a plain film the brain is fog with a wall around it.
So the profession worked by inference. Walter Dandy described pneumoencephalography in 1918: take out the fluid, put in air, photograph the shape of the space. It does not show the brain. It shows the brain’s furniture, and the radiologist reads the tumor from how the furniture was pushed aside.
The air study was standard into the 1970s. A neuroradiologist who trained on it called it the ultimate example of man’s inhumanity to man. Recovery ran to days or weeks, because the fluid returns only as fast as the body makes it. And often, as for the woman in the chair, the answer was a maybe.
In 1956 the only nuclear physicist within reach of Groote Schuur Hospital in Cape Town was a lecturer named Allan Cormack, who came a day and a half a week to supervise the isotopes. Watching radiotherapists plan doses for a head they treated as uniform, which it is not, he saw that the inside of an object might be recoverable from measurements taken outside.
He published the mathematics in 1963 and again in 1964. He held no doctorate in anything. There was practically no response. The most interesting request for a reprint, he said later, came from a Swiss avalanche research centre, which had noticed the method would work for snow on a mountain.
He was not the only one ignored. William Oldendorf, a neurologist at the Los Angeles VA, had a prototype in 1960 and a patent by 1963, and quit in 1964 after a manufacturer wrote that it could not imagine a market for so expensive an apparatus.
Godfrey Hounsfield left school in his mid-teens, learned radar in the RAF, and took a diploma at Faraday House that stayed his highest qualification. At EMI he had led the team that built Britain’s first all-transistor computer. In 1967 it came to him that with enough readings through one slice of an object, from enough angles, a computer could solve for what was inside.
EMI was not enthusiastic. It sold no medical products, and the laboratory director would not spend its money unless somebody else spent some too. The Department of Health put in £2,500 and EMI matched it. Published totals cannot be reconciled across the sources. The shape holds: small sums, matched, from a government department and a record company with no idea what it had.
The first rig was built on the bed of an old lathe. A gamma source crawled across, the object turned one degree, and it crawled back. Twenty-eight thousand readings, nine days to collect them, two hours to compute the picture. He scanned Perspex, then bullocks’ brains from an abattoir, then a preserved human brain. An X-ray tube in place of the gamma source brought nine days down to nine hours.
On 1 October 1971, at Atkinson Morley’s Hospital in Wimbledon, a woman in her early forties with a suspected tumor of the frontal lobe lay down with her head in a rubber sock inside a box of water. A pencil of X-rays crossed her skull, stepped one degree, and crossed again, a hundred and eighty times.
No computer was attached to the machine. The readings went onto tape, the tape went across London by car to EMI’s laboratories at Hayes, and the picture came back on a Polaroid. James Ambrose, the radiologist, waited two days.
The image was eighty squares by eighty, and by any standard since it is close to unreadable. The cyst was there, and surgery found it where the picture said.
Nobody defended the old way. There was nothing to defend. Within eight years the scanner was in something like a thousand hospitals; by 1980 the United States alone had 1,471.
Americans had about ninety-three million CT examinations in 2023. A study published last year modelled the radiation from that one year and projected on the order of a hundred thousand cancers over the lifetimes of the people scanned.
That figure deserves more caution than it gets: it is a projection, not a count. It extrapolates from the atomic bomb survivors, assuming risk stays proportional to dose all the way down, and serious radiation scientists say in public that the assumption is unproven at the low end. Others point to work that followed children scanned decades ago and found the excess about where the model put it. I do not know which side is right. I am confident the number is not zero, and that almost nobody scanned this year will hear a word about it.
The dose is the smaller problem. The larger one is that the machine sees everything in the plane, including everything nobody was looking for. Pulmonary nodules turn up in as many as one chest CT in four. Most are nothing. Some are not, and the only way to know is to go and take a piece.
Among nearly sixteen thousand adults who had a needle put through the chest wall to sample one, fifteen percent had a lung collapse. One in fifteen — of everyone biopsied, not of everyone who collapsed — needed a tube between the ribs to get it open again. One percent bled, and some ended on a ventilator.
Those people had walked in for a picture.
There is a stranger cost still. CT can resolve a clot two or three millimetres across in the lung’s smallest branches. Clots that small were about one percent of positive scans under the old test and about fifteen percent under the new one. Either we had been missing them, or small clots off the leg veins are ordinary, catching them before they reach the brain is among the things a lung is for, and the body dissolves them unasked. If that is right, the machine has photographed the lung doing its work, and we have called it a disease and started the blood thinners. Not everyone agrees; a clinician wrote to the same journal to argue that every embolism should be treated, however small.
Hounsfield and Cormack shared the Nobel Prize in 1979. They had never met. Cormack had written to Hounsfield years before, after learning what EMI was building, and Hounsfield never answered. At the banquet in Stockholm he asked Cormack to speak for them both.
EMI was out of the scanner business within the year, sold to General Electric. The story that the Beatles paid for the machine appears nowhere in the record until Hounsfield’s obituary in 2004.
What endures is a picture of the inside of a living person, made without opening them, at a price almost never quoted to the person. Nobody is strapped into a chair and turned upside down anymore. That is all of it, and it is enormous.
It cannot be undone, and most of us would not undo it. That is the bargain medicine made.
The next machine would show the soft tissue this one could not, and without a single X-ray — and its prize would leave out the man who believed he had earned it.
* * * * *
Charles C. Jett is an author, civic educator, and Professional Certified Coach based in Chicago. A graduate of the U.S. Naval Academy (Class of 1964) and Harvard Business School, he served during the Cold War aboard the nuclear submarine USS Ray (SSN 653), where his tactical innovations helped inspire Tom Clancy’s Jack Ryan character. He is the author of six books, including Super Nuke!, hosts four podcasts, and writes across his Critical Skills Blog platform on history, leadership, and the health of the American republic. In his writing he employs AI tools in a limited, supporting role for research, occasional image creation, and editing, while the prose and judgment remain entirely his own. He and his wife, Dr. Nancy Church, live and co-host the Chicago Salons at Water Tower Residences.
Leave a Reply