CERN Large Hadron Collider instrumental in medical and automation developments

February 19th, 2014, Published in Articles: EngineerIT


It is almost unbelievable that systems like LabView, PXI and NI FlexRio hardware could become an important and integral part of a CERN-developed treatment for cancer.

CERN’s Dr Johannes Gutleber

Dr. Johannes Gutleber

CERN (the European Organisation for Nuclear Research) made history over the past few years when it fired up the Large Hadron Collider (LHC), probing the fundamental structure of the universe. The world’s largest and most complex scientific instruments are used to study the basic constituents of matter – the fundamental particles, colliding at close to the speed of light. The process gives physicists clues about how the particles interact, and provides insights into the fundamental laws of nature. The instruments used at CERN are purpose-built particle accelerators and detectors. Accelerators boost beams of particles to high energies before the beams are made to collide with each other or with stationary targets. Detectors observe and record the results of these collisions.

However, little is known about deployment of this technology in the treatment of cancer, or how through the use of reconfigurable I/O architecture and LabView, a medical instrument approximately the size of a rugby field,  is controlled to deliver the most precise pinpointed radiation therapy.

Delivering a keynote address at the National Instruments 2013 NI-Week held in Austin Texas, CERN’s Dr. Johannes Gutleber explained that the development of the medical accelerator  relied heavily on the facilities offered by LabView  and the availability of  advanced reconfigurable  I/O architecture such as FlexRio hardware.  This  combines high-performance modular I/O, powerful field-programmable gate arrays (FPGAs), and PC-based technologies into a platform ideally suited for on-board processing and real-time analysis.

Dr. Gutleber said that to develop such an advanced cancer treatment system required two separate approaches. “First we had to design the physics and then the technical  aspects of the system.   The physics design is an outflow of what we learned with the Large Hadron Collider. How many magnets would we need? How strong do the magnetic fields of the magnets have to be? What does the radio frequency scheme look like? What kind of iron sources would be used and what type?  What should the layout of the beam be? The beam in the machine works at the third harmonic so it oscillates in the synchrotron and this has to be understood. Controlling a third resonance is very tricky. These questions had to be resolved. After completing the physics design we did not immediately embark on designing a specific system, but continued our research on particle accelerators.”

When the Austrian government, a CERN partner, decided to build a medical accelerator to treat cancer, CERN  was well placed to become involved. “With the physics design completed we could almost immediately start on the technical design. This was in 2009. The particle accelerator was delivered at the end of August 2013.”

The rest of the system will be installed and tested by Austrian scientists and engineers under the direction of CERN personnel. Completion is expected in 2014.

There will be four treatment rooms, but only one patient can be treated at a time while the other three patients are being prepared. The patient is immobilised and placed in position so that the pinpoint radio therapy can be directed at the correct spot. This requires use the use of X-rays and a team of medical people who must ensure the treatment is delivered at the correct site.

“Controlling the system was one of the most challenging tasks. Patient position must be accurate within one tenth of a millimetre. In the technical design we had some major challenges. There are  many configuration settings such as  8000 energy levels which can be used with seven dfferent  ion settings. When extracting the beam we can manipulate the duration of the beam from anything of from 0 to one second to 20 seconds long. We have beam settings of 1 mm x 1 mm to 10 mm x 10 mm. We have three different  ion sources upgradable to five. In the end there are hundreds of thousands of different settings for the front end controller to drive the beam line elements.  To manage that, high input real-time systems are needed, which I believe are not very common at this time. Computing is still very much computation-bound. We do not require much of that,  we require big  data shifting facilities. From information on hard disks in the front end, a few hundred waveforms are selected  and cached  into the  memory. From there we play out  the wave forms into the FlexRios.  Thee FPGAs then play out the wave forms and feed them to the power converters of the magnets. We drive the magnets with DC power and as the wave form changes the magnetic field changes. This is how the acceleration process worksm – it is done in a highly synchronised manner. As the one cycle is played out , the system sets up the next cycle. This means, unlike older designs, there is no delay between beams, one follows the other, reducing the treatment cycle. To operate synchronously,  sophisticated timing systems are required. The 300 magnets all have to work at micro second precision.  Included in the system there are a large number of sensors which are all read out with PXI, a PC-based platform for measurement and automation systems  and visualised on LabView panels. The peer-to-peer streaming technology from NI enables direct data streaming among multiple FPGA modules, or between select PXI Express modular instruments and FPGA modules, without sending data back to the host processor with  streaming at rates exceeding 800 Mbps (aggregate) on up to 16 separate streams.

Many suppliers and sub-contractors were involved in the project, which required close collaboration between them and the design and construction teams. Undoubtedly a lot of lessons were learned and new techniques developed. It is with projects like the medical accelerator in Austria that thinking moves way out of the box and beyond;  and specific innovative solutions are developed that ultimately find their way into products and systems deployed in industry. It therefore remains a prerequisite that scientists, researchers and industry work together – even if in some instances  the profit motive has to be set aside.

Related Articles

  • AI tools to accelerate enterprise IoT solutions
  • How drones and artificial intelligence can help speed hurricane recovery
  • How Coriolis measurement helps keep the Scotch in whisky
  • Energy efficient digital front-end designs
  • IITPSA announces 2019 President’s Awards shortlist