A note on what a patent cannot protect

A patent is the thing we teach researchers to want. It is also the part of a position that is easiest to count, which is why it tends to become the part we manage. The CT scanner is the clearest case I know of what a patent does not buy.

What happened

On 1 October 1971, at Atkinson Morley's Hospital in Wimbledon, a machine took the first clinical CT scan of a living brain. It found a cystic mass in the patient's left frontal lobe.

The machine came out of EMI's Central Research Laboratories. Godfrey Hounsfield, who built it, had no university degree. He held a diploma from Faraday House and had taught radar in the RAF. In 1979 he shared the Nobel Prize in Physiology or Medicine with Allan Cormack, for the development of computer assisted tomography. Neither man held a medical degree.

Nobel Prize in 1979. Out of the industry in 1980. EMI accepted a takeover bid from Thorn, and the scanner business went to General Electric the year after.

Two things went wrong, and neither of them was the science.

The patent did not hold anyone out

US 3,778,614 was assigned to EMI, with a priority date of 23 August 1968 and a grant in December 1973. EMI sued Ohio-Nuclear and other entrants. Several of those challenges settled before trial in exchange for licences.

David Teece set out why in 1986. Once the product was on the market it could be taken apart and its essential features copied, and a patent on a machine of that kind describes the invention well enough to help the person copying it. EMI could extract terms. It could not extract exclusion.

EMI never owned the delivery

Teece's word for what EMI was missing is complementary assets: manufacturing at scale, distribution, field service, clinical training. A hospital does not buy a scanner. It buys a scanner that is installed, serviced within a day, and operated by staff somebody trained.

EMI had none of those and could not easily contract for them. Its competitors reached whole-body scanning first and built the service networks while EMI was still litigating.

Invention decides who gets the credit. Delivery decides who gets paid.

One correction, because it is repeated everywhere

The Beatles did not pay for the CT scanner. EMI was their label, and the story is charming enough that it survives without a source. EMI Music did not fund EMI Medical. The prototype programme ran on EMI's own money together with support from the UK Department of Health and Social Security.

Why I am writing this down

In a technology transfer office the patent is the artefact we can file, number, renew and report. It is therefore the thing that gets managed, and the thing a researcher is congratulated for. None of that makes it the thing that decides the outcome.

Before a licence, the question worth more than the claim chart is this one: in this market, who owns manufacturing, distribution, service and training, and what does the licensee still have to buy or build that the patent does not give them? If the answer is that a large incumbent already owns all four, a strong patent is a bargaining position and not a business.

The same question decides spin-outs. A team with a good patent and no route to those assets is a licensing candidate. We tend to make that call on enthusiasm rather than on the map, and then wonder why the company stalls two years in.

This is the second time in a month I have found the same habit in our field. We measure the part that is countable, and we let it stand in for the part that matters.

What I could not establish

Whether EMI ever lost a patent case. I could not locate a judgment; the challenges I traced settled. The word "lost" is repeated in secondary accounts and I would not put it in print.

The share of the prototype programme carried by public funding. The figures I found range from about forty thousand dollars to over six hundred thousand pounds depending on the source, so the proportion cannot honestly be stated.

---

Sources

• IEEE Milestone, First Computerized Tomography (CT) X-ray Scanner, 1971, plaque at Hayes, dedicated 26 October 2022.
• The Nobel Prize, Physiology or Medicine 1979, Allan M. Cormack and Godfrey N. Hounsfield.
• Hounsfield GN, Computerized transverse axial scanning (tomography): Part 1, British Journal of Radiology 46(552), 1973; Ambrose J, Part 2, same issue.
• US Patent 3,778,614, assignee EMI Limited, priority 23 August 1968, granted 11 December 1973.
• Teece DJ, Profiting from technological innovation, Research Policy 15(6), 1986.
• Harvard Business School working paper 20-004 on EMI and computed tomography.
• On the Beatles claim: Journal of Computer Assisted Tomography 36(2), 2012.