Structural model analyzing the von Mises stress helps find the optimum diaphragm thickness for wireless implantable actuator.Multiphysics software simulations are used by biomedical equipment developers to reliably design complex mechanisms for enhancing the human physical condition. These medical devices can include tools for treating cancers, enhancing hearing and treating chronic back pain.

Researchers at the Cross Cancer Institute, Edmonton, Alberta, Canada, for example, combined a linear particle accelerator (linac) for treating cancer tumors with a magnetic resonance imaging (MRI) scanner to precisely position the linac’s cancer-killing x-rays onto a target. However, the problem encountered previously with combining these two medical devices is that they interfere—the faint radio frequency (RF) signals created by the MRI used for imaging the tumor also create stray magnetic fields that deflect the linac’s beam. This is especially noticeable on tumors in the thorax or abdomen. Utilizing Comsol’s Multiphysics software for the past six years, the researchers first created a system for shielding the x-ray beam from the MRI. They then simulated and optimized the structure of the MRI scanner to incorporate a hole for the x-ray beam to pass through. In doing this, the research team had to incorporate angle-specific field heterogeneties due to the rotating path of the x-ray beam. They solved this by designing the linac and the MRI scanner as one complete system that moved together.

“Monte Carlo modeling of these radiation fields is the gold-standard,” says Gino Fallone, a prof. at the Univ. of Alberta, Edmonton, who established a task force to resolve these design issues. “There are thousands of publisher papers showing their accuracy (less than 1%) compared to standard measurement devices (i.e., diode, ion chamber). Some of this physical measurement uncertainty is due to the experimental setup uncertainties.”

The multiphysics modeling systems they used also allowed them to reduce the overall size of the linac and the resulting room housing the system as well. While undergoing final certification trials, the system is expected to be in full scale use by 2016.

Researchers at Cochlear Ltd., Sydney, Australia, have also used multiphysics software models to create a Direct Acoustic Cochlear Implant (DACI) that provides mechanical (acoustic) stimulations directly to the cochlea, especially targeted at people who suffer from severe to profound mixed hearing loss. This device covers the gap where conventional hearing aids do not have enough power. The Codacs system starts with a standard behind the ear device. It contains batteries, directional microphone and digital-signal processing circuitry. These signals are transmitted wirelessly to an actual implanted in the ear cavity right behind the external auditory canal. The Codacs actuator is not intended to amplify the sound (as in traditional hearing aids). Instead, it directly amplifies the pressure waves inside the cochlea.

Codacs designers were challenged with the extremely small size due to the limited space available in the mastoid cavity, the frequency characteristics similar to the human ear, low power consumption and the biocompatibility, or hermetic encapsulation, of device materials. What impressed the design team was the software’s capability to do a number of studies—structural, acoustic, electromagnetic and piezoelectric—in one unified package. This prevented time-consuming iterative processes and costly trial-and-error design approaches.

In another multiphysics-enabled medical study, researchers in the Neuromodulation Group at Lahey Clinic in Burlington, Mass., enabled doctors to learn how implanted electrodes treat back and leg pain. While a large number of processes over the years have involved surgically implanted electrodes that apply electric potential directly to the spine, there is still no precise understanding of its mode of action—these signals somehow interfere with the human pain signaling circuitry. Collecting a wealth of published information available, the researchers found that the scar tissue from implanting the electrodes alters the impedance seen by the implanted electrodes, which in turn alters the pattern of the electric field distribution. They theorized that a 3-D model could be used to accurately predict these changes and define corrective modification of the stimulus pattern. This could reverse the often-observed deterioration in the performance of the treatment. To study the electrical environment of the spinal cord, the research team used the software to create a finite element model of the gray and white matter in the cord, dura, cerebrospinal fluid, epidural tissue, scar tissue and stimulator electrodes.