How a Göttingen MRI trick became the Max Planck Society's most lucrative invention
In 1985 a small team in Göttingen made MRI a hundred times faster. Then they spent years fighting GE, Siemens and Philips in court. The result: about 155 million euros and a technique in every scanner.
The first magnetic resonance image of a human body, taken in 1977, needed 4 hours and 45 minutes. Patients had to lie motionless for longer than a feature film, and moving organs such as the heart were out of the question. Less than a decade later a group of physicists and chemists in Göttingen found a way to make the same kind of measurement roughly a hundred times faster. Their method, FLASH, became part of practically every MRI scanner on Earth and the single most profitable patent the Max Planck Society has ever held.
It is also a case study in something Europe is often accused of doing badly: turning publicly funded basic research into money. The story has a clever idea, a fast patent filing, a refusal by the world's biggest medical technology companies to pay, and a court fight that a non-profit research body won.
The problem: waiting for protons
MRI works by placing the body in a strong magnetic field, which lines up the tiny magnetic moments of hydrogen nuclei in water and fat. A radio pulse tips them out of alignment; as they swing back they emit a faint radio signal that, with the help of magnetic field gradients, can be located in space and turned into an image.
The catch lies in the tipping. Classic sequences used large flip angles, typically 90 degrees, which use up most of the available magnetisation in one go. Before the next measurement the scanner had to wait for the nuclei to relax back, a process governed by the tissue's T1 relaxation time, often a large fraction of a second. An image needs many such measurements, so the waiting dominated everything.
The idea: tip them only a little
In the early 1980s Jens Frahm, a physical chemist born in Oldenburg in 1951 who led the Biomedical NMR research group at the Max Planck Institute for Biophysical Chemistry from 1982, worked on the problem with Axel Haase, Dieter Matthaei, Wolfgang Hänicke and Klaus-Dietmar Merboldt. Their solution was counterintuitive: use a small flip angle, so each pulse disturbs only a fraction of the magnetisation, and read the signal as a gradient echo rather than a spin echo.
Because most of the magnetisation is left untouched, there is no need to wait. The next pulse can follow after a few milliseconds. The signal from each measurement is weaker, but there are vastly more of them in the same time. The team named the method Fast Low Angle SHot, FLASH. Scan times that had been measured in minutes fell to seconds, a speed up the Max Planck Society describes as at least a factor of 100.
The first clinical images appeared in The Lancet in 1985 and the main methods paper in the Journal of Magnetic Resonance in 1986. Crucially, the team did not publish first and think about patents later. German patent application P 35 04 734.8 was filed on 12 February 1985. The US patent was granted in 1987 and the European patent in 1989.
Adopted in months, paid for in years
The effect on the industry was immediate. Moving images of the beating heart, 3D datasets and fast routine scans suddenly became practical. One researcher later put it bluntly in the society's own account: from then on, no manufacturer could live without it. Leading manufacturers introduced FLASH within a few months.
Paying for it was another matter. The society's technology transfer company, then called Garching Instrumente and later Garching Innovation, asked for a royalty of about one per cent of turnover. General Electric, Siemens and Philips all filed oppositions against the patents and continued to use the technique. Philips also sought to have the US patent declared invalid.
What followed, according to the Max Planck Society's magazine, was about seven years of preparation for litigation led on the transfer side by Bernhard Hertel, who travelled to the United States dozens of times. General Electric was the most stubborn opponent. The opposing side fielded large teams of lawyers and hired a Nobel laureate in chemistry as an expert witness. Negotiations were transcribed in such detail, because a careless remark could have endangered the patent, that the minutes reportedly recorded a barking dog being silenced by Frahm's attorney.
The turning point came in 1992, after the European patent was upheld in full. Siemens took a licence, reportedly at 0.7 per cent of turnover with a down payment of 20 million US dollars. General Electric then gave way, and Toshiba, Hitachi, Philips and other manufacturers also signed. A 1993 judgment required the companies to pay licence fees retroactively. The society says it had spent close to 1.5 million euros on the proceedings up to that point; its magazine puts the total legal cost at about three million Deutschmarks.
155 million euros
The return dwarfed the cost. According to the Max Planck Society, the FLASH patent generated about 155 million euros in licence income for the society, the Göttingen institute and the inventors, the largest sum in the history of its transfer company. To put that in context, Max Planck Innovation reports total licensing revenue of more than 585 million euros from all inventions since 1970. One MRI patent accounts for more than a quarter of everything.
The money did not simply vanish into the general budget. Since 1993 Frahm has directed Biomedizinische NMR Forschungs GmbH, a non-profit company set up within the Göttingen institute, which gave his group a degree of independence to keep developing the method.
The sequel: MRI as video
The patents have long since expired, but Frahm's group kept going. In 2010 it presented FLASH2, which combines heavily undersampled radial data with iterative image reconstruction, a computationally demanding approach that became feasible on modern graphics processors. The result is real-time MRI at up to 100 images per second, with a temporal resolution of around 20 milliseconds in published work.
That turns MRI into something closer to video. Clinicians can watch joints move, a heart beat without synchronising to an ECG, or a patient swallow and speak. In one collaboration with the Hanover University of Music, Drama and Media and players from the Berlin Philharmonic, the group filmed brass musicians and found that tongue movements largely shape the notes. Real-time MRI has been tested for routine use at the University Medical Center Göttingen and at universities in Germany, Britain and the United States, including Oxford and Johns Hopkins. Frahm is named as inventor on further European patents covering the newer methods.
Recognition
Frahm's awards track the story: the Karl Heinz Beckurts Award in 1993, the gold medal of the International Society for Magnetic Resonance in Medicine in 1991, election to the Hall of Fame of German Research in 2016, the European Inventor Award of the European Patent Office in the research category in 2018, and later the Werner von Siemens Ring, one of Germany's most prestigious technology prizes. It is a nice irony that a ring named after the founder of one of the companies that once refused to pay should honour the inventor.
What the FLASH story teaches
Three lessons stand out for anyone thinking about research and industry in Germany.
- File before you publish. The Göttingen team protected the idea early. Without the February 1985 application, the method would simply have become free knowledge for every manufacturer.
- A licence is only worth what you will defend. The world's leading device makers adopted FLASH immediately and paid only after years of pressure. A research organisation without the patience and funds to litigate would have collected nothing.
- Royalties compound. A modest share of turnover across an entire global industry, over the life of a patent, produced a sum that still dominates the Max Planck Society's licensing history.
The Max Planck Society estimates that around 100 million MRI examinations a year use FLASH. If you have had a scan, there is a good chance part of it was taken with a sequence that a small group in Göttingen worked out in the early 1980s and then fought for.
Sources
- Max Planck Society: European Inventor Award for fast MRI in medical diagnostics
- Max Planck Society: Success stories of Max Planck Innovation
- MaxPlanckResearch: Focus feature on technology transfer and FLASH
- Max Planck Society: From invention to innovation, yearbook 2017
- European Patent Office: Jens Frahm, European Inventor Award
- Max Planck Institute for Multidisciplinary Sciences: European Inventor Award press release
- Wikipedia: Fast low angle shot magnetic resonance imaging
- Ärzte Zeitung: Germany's most important technology prize for FLASH MRI