Brain strokes are one of the leading causes of human deaths and disabilities. They can occur in the form of an ischemic stroke (i-stroke, where blood flow is cut off—e.g., by a blood clot) or a hemorrhagic stroke (h-stroke, where bleeding in the brain is causing the damage). Traditional imaging methods for studying possible conditions for brain stroke include computed tomography (CT) and magnetic-resonance-imaging (MRI) systems, although the use of electromagnetic tomography (EMT) is rapidly gaining favor as a means of studying human brain conditions. As a device under test (DUT), the human brain poses many challenges for the use of EMT, due to the wide range of dielectric permittivity in the brain, from its fluids to the skull.
Undeterred by the challenges of performing electrical analysis on brain tissues and other materials found in the human brain, researchers from EMTensor GmbH helped to develop what might be considered a “massive-MIMO” analysis system with a large number of small antennas in a spherical configuration to surround a subject with high-frequency transmitters and receivers. Measurements at around 1 GHz were combined with their own direct solvers using finite-element (FE) and finite-difference-time-domain (FDTD) modeling techniques. This combination of 2D and 3D special analysis around the brain enabled the researchers to perform EMT imaging of the human brain and its collection of objects and materials with a wide range of dielectric constants as part of a new approach for monitoring human brain stroke.
The researchers’ BRIM G2 system uses 177 ceramic-loaded rectangular-waveguide antennas with dielectric constant of 60, aperture of 21 × 7 mm, and length of 53 mm. The antennas have a low-frequency cutoff of about 922 MHz and are well suited for L-band applications. They are mounted in a spherical configuration in a measurement chamber formed of medical-grade stainless steel. The antennas are distributed within the chamber in a ring formation, to allow for 2D and 3D image reconstruction. Every ring contains an even number of antennas, with a receiver for every transmitter.
The 2D measurements were made with the aid of a model ZVA8 vector network analyzer (VNA) from Rohde & Schwarz. To provide a transceiver mode of operation for the measurement system, each antenna was accompanied by a model P9402A single-pole, double-throw (SPDT) switch from Keysight Technologies, and each receive-path signal was boosted by a model ZRL-2150 low-noise amplifier (LNA) with 25-dB small-signal gain from Mini-Circuits.
The 3D measurements used the test chamber equipped with a ZNB4 VNA from Rohde & Schwarz, two switch matrices from Corry Micronics (www.cormic.com), and semirigid coaxial cables for interconnections. The two switch matrices provide time-division access to the antennas, using different rings of antennas at different times for transmit and receive functions.
For calibrating the many antennas and their different interconnection transmission lengths, a reference antenna with an almost omnidirectional radiation pattern is placed in the center of the measurement chamber and its transmissions are measured with the multiple smaller antennas. Similarly, differences among the transmission paths of the many smaller antennas can be identified when the reference antenna is operating in a receive mode. The researchers were able to develop a calibration coefficient for the measurement chamber and its many antennas, which could then be used as a factor for the raw data measured at the VNA interface with the test chamber.
Certainly, the setup and use of the BRIM G2 system is not simple and requires precise calibration and measurement practices. It also requires the reconstruction of an image for the subject (the brain) being studied through the analysis of the EM field distributions inside the test chamber, which is dependent upon the distribution of material permittivity within the chamber. The test chamber was modeled and analyzed with the aid of the High Frequency Structure Simulator (HFSS) electromagnetic (EM) simulation software from Ansys Corp. The simulated and actual results obtained by the researchers show great promise in this test approach as a monitoring method for learning more about human brain strokes, and possibly how to prevent them.
See “Electromagnetic Tomography for Detection, Differentiation, and Monitoring of Brain Strike,” IEEE Antennas & Propagation Magazine, Vol. 59, No. 5, October 2017, p. 86.