A multitalented group of engineers led by Professor Gordon Sarty is developing a compact Magnetic Resonance Imaging (MRI) scanner for spaceflight duty. The intent is to support space medicine research and astronaut health monitoring required for longer and more remote space missions. The first post of duty would be on the International Space Station (ISS), to monitor physiological changes occurring during long-duration missions. Sarty is Acting Chair of Biomedical Engineering at the University of Saskatchewan.
Space MedicineOur current knowledge of space medicine is largely the result of 40 years of experience with long-duration occupation of the ISS and earlier space stations. The ISS has been continuously occupied for 14 years, and thirty people have each spent more than a year (the record is 2.2 years) on a space station. Despite this, most data on the long-term medical effects of space travel comes from post-flight studies of returning astronauts. As the medical equipment aboard the ISS offers little more capability than a well-stocked first-aid kit, in-orbit capability for collecting or processing medical data is very limited.
Space travel causes a plethora of known medical problems, such as loss of as much as 25 percent of muscle mass, loss of bone mass of 1 to 2 percent per month, weakening of the immune system, and effects related to the very high stress environment of a space vehicle. Doubtless other negative effects will be encountered in longer missions.
During past space station missions, there have been three medically-driven evacuations of crew members to Earth. The expected rate for long-term missions involving the extremely healthy astronaut pool is about one evacuation per hundred person-years. Once you leave low-Earth orbit, however, depending on a quick return to Earth for emergency diagnosis and care quickly becomes impractical.
MRI ScannersA deep-space sickbay must be equipped to do research on the debilitating effects of long-term space travel as well as to provide diagnostic monitoring and treatment of the crew. One important part of these diagnostic capabilities is a viable medical imaging system. The need for frequent health monitoring of the astronauts points toward using a non-ionizing radiation technique such as MRI to avoid adding extra radiation exposure to that of the deep-space environment itself. The only other serious candidate is ultrasonic echo imaging, but this imaging method does not identify tissues nor penetrate bone, and makes whole-body scans a lengthy and difficult ordeal.
A three-Tesla conventional MRI scanner, the 3T Achieva by Philips (Photo: Kasuga Huang)
MRI works by combining two physical effects that allow us to measure the response of protons (the nuclei of hydrogen atoms) to an applied magnetic field, and to use gradient techniques to select a particular voxel (3D pixel) of a volume being scanned. The resulting image is essentially a plot of the density of hydrogen atoms as a function of position throughout the active volume of the imager. To carry out these processes, an MRI scanner requires a method of providing a large and highly uniform magnetic field (typically 1-3 Tesla – a Tesla is equal to 10,000 gauss, and the strength of the Earth's magnetic field is about half a gauss), and a method to isolate and measure the proton density in each voxel of the scanning volume.
Conventional MRI in SpaceThe paths taken to accomplish the above tasks in conventional MRI scanners are difficult and dangerous to use in a space-based whole-body MRI scanner. To begin with, the physical footprint of a conventional whole-body MRI scanner is generally in excess of the capability of even the largest available launch vehicles. In addition, the magnetic field for the scanner is provided by a superconducting magnet which usually weighs in excess of ten tons (some very-high magnetic field MRI scanners weigh as much as 250 tons.)
These magnets use recirculating liquid helium systems to cool the superconducting magnet windings. If the magnet quenches (spontaneously regains its resistance), both magnet damage and conversion of a large amount of liquid helium into helium gas will result, likely followed by sudden and catastrophic "disassembly" of the cryogenic system. Such an event would seriously disrupt the ecosystem of any reasonably sized spacecraft.
Voxel selection is carried out using rapidly changing gradient magnetic fields which are added to the uniform magnetic field of the main magnet. To switch the various gradient coils between gradient patterns, however, requires enormous high-voltage high-current pulses. This is the source of the banging noise you hear during an MRI scan – the coils themselves are twisting and warping with the effect of the magnetic field on their rapidly changing currents. Supplying such pulses also drives premature failure of the gradient coil electronics – quickly changing large currents and dealing with their associated transient voltage spikes is hard on electronics, and particularly hard on solid-state circuitry.
The Halbach cylinderFortunately, there are other designs for MRI scanners that are more suited to the constraints of spaceflight. The large, heavy, and dangerous superconducting magnets can be replaced by an arrangement of permanent magnets called a
The Halbach array was initially developed for focusing particle beams in high-energy accelerators. It is possible to create a cylindrical arrangement of permanent magnets that produce a highly uniform magnetic field inside the cylinder bore, while allowing only a very small portion of the magnetic field to exist outside the Halbach cylinder. an important consideration for a space-based MRI scanner. As modern magnetic materials are capable of generating fields of over one Tesla, such Halbach cylinders are suitable for use in MRI scanners.
Schematic of a refrigerator magnet, showing the rotating direction of the magnetic domains and the one-sided magnetic field which is produced therefrom
We are all familiar with Halbach arrays in an unlikely arena – the kitchen. A refrigerator magnet is a one-dimensional Halbach array. These are magnetized using a rotating magnetic field to produce rotating magnetization directions in the magnet strip. The result is as if the field lines from one side were transferred to the other side, where the larger field line density is equivalent to a larger magnetic field. This is sometimes called a one-sided magnet. You can test this yourself by trying to stick two refrigerator magnets back to back – they will barely attract each other. You may also notice that when the two magnets are oriented front to front, they may resist being placed right atop each other, and will stick with an offset on their long axis. The offset distance is half the spacing of a complete rotation of the magnetization directions.
Computer simulation of the magnetic field of a Halbach cylinder composed of four pairs of permanent magnets, each pair having different magnetic orientations than the other pairs (Image: Zureks)
As shown above, the magnitude of the local magnetic field on a Halbach cylinder is constant everywhere on the cylinder. However, the orientation changes so as to produce a reinforced, uniform magnetic field inside the cylinder bore, while only small magnetic fields appear external to the cylinder. This latter property is important for space-based MRI scanners, as most spacecraft designs require a clean electromagnetic environment. Although the magnetic field in the cylinder bore is not completely uniform in the above example, the uniformity can be greatly increased using magnetic shims to move the field lines around.
The Sarty group's prototype Halbach cylinder, used mainly to study shimming techniques for improving the homogeneity of the magnetic field in the cylinder bore (Photo: University of Saskatchewan)
The Halbach cylinder constructed by the Compact MRI project for laboratory testing is made of rare-earth permanent magnets. The cylinder has a volume of 0.75 m3, compared to volumes of about five to ten cubic meters for conventional MRI magnets.
The cylinder bore field strength is 0.15 tesla, though the magnetic field outside the curved outer surface of the magnetic cylinder is only 0.2 percent of that value (about 60 percent of the Earth's field) at a distance of 7 cm (2.75 in) from the cylinder surface. The magnetic field in the bore could be increased to about one tesla through appropriate choice of the magnet materials and their structure within the cylinder. With an optimized structure, a similar magnet for space MRI applications is projected to weigh less than 700 kg (1540 lbs) – a far cry from the 10 or more tons of conventional MRI magnets.
MRI images of knee and wrist joints taken using the TRASE RF phase gradient method (Photo: University of Saskatchewan)
A substitute for the magnetic field gradient approach toward measuring the proton density of a single voxel would also be of great benefit to a space-based MRI scanner. Sarty and his group are using a recently-developed approach to voxel measurement called TRansmit Array Spatial Encoding (TRASE).
The TRASE method offers an alternative to gradient-based MRI imaging. A gradient is still applied to the body in the imaging volume, but it is an RF phase gradient rather than a gradient of the magnetic field itself. Remeber, the phase of an RF voltage tells you where and when the voltage reaches its peak in each cycle. In the TRASE method, the flipping RF coil is replaced with an RF coil that produces a field with a spatially varying phase. Vastly less power is required to shift the phase gradient of the RF flipping radiation than to shift large-scale spatial gradients.
The basic idea of TRASE is that the scanning volume is subjected to RF pulses whose phase changes with time. These pulses sequentially excite a given voxel of the scanning volume, whereas those same pulses average out to produce zero signal from the other voxels in the scanning volume. There is a large degree of parallel operation in TRASE, unsorted using tomography, but the end result is to determine the proton density for each voxel in the scanning volume. The procedure is far from perfect and isn't all that simple to implement, but TRASE has been successfully used for MRI research scanners by several groups over the past decade.
MRI in space?
Canada has an ISS allotment of 50 kg (110 lbs) in return for their contribution of the Canadarm2. It is Professor Sarty's goal to use this allotment to install on the ISS a Compact MRI for arm and leg study. This would enable detailed studies of joint and bone degenerative changes, of fluid transfer between portions of the body, and other studies which cannot currently be carried out during exposure to the zero-g spaceflight environment. The outcome of his efforts is not yet decided, but it seems clear that in upcoming years the human race will need much more medical data relevant for long-duration spaceflight. Why not start now?