An officer of the Federal Food and Drug Administration and a mechanical engineer by training, Tina Morrison is well-versed in both FDA regulations and FEA simulation.
Morrison is the chair of the new FDA-wide working group on Modeling and Simulation, and a keynote speaker at the upcoming CAASE 2018 Conference. Her talk is titled, “Priorities Advancing Regulatory Science and In Silico Medicine at the FDA.”
As the regulatory advisor of computational modeling for the Center for Devices and Radiological Health (CDRH), Morrison leads the group that develops guidance documents on the use of modeling and simulation in the regulatory evaluation of medical devices. The research at CDRH covers, among others, the use of Computational Fluid Dynamics (CFD) to model medical devices that interact with the patient’s body.
The Human Geometry
The geometry creation and analysis tools in mainstream CAD and simulation packages reflect their roots in automotive, aerospace, and industrial machinery. But the medical device industry’s adoption of these tools broaden their scopes and test their capacity.
“Modeling [patient anatomy] is not a problem. We can now use a combination of CT scan data and MRI data to construct the STL file the body, then print a 3D model,” said Morrison. “But one of the challenges is uncertainty quantification — understanding the uncertainty involved in reconstructing the geometry of the patient’s body.”
When Metal Meets Flesh
Simulating the mechanical operations of the medical device and the stress loads on it are straightforward enough. But the analysis gets much more complicated when dealing with the device’s interaction with the human body, Morrison pointed out.
“Most firms are using digital tools to simulate the loads and how they affect the device itself,” she observed. “But some would go so far as to simulate how the device interacts with the patient’s body — for example, a medical stint or implant inside someone, or the kinematics of a replaced knee joint while the patient is performing some activities. So in those, the material science is important.”
Common manufacturing materials — such as metal, steel, and plastic — have known properties users can obtain from published literature. Many simulation software also comes preloaded with a library of standard materials, ready for the users to pick and choose from a drop-down menu.
But digital representation of soft tissues, bones, and human muscles is still under development. Therefore, to ensure accurate simulation results, most users dealing with such materials may need to run clinical tests on their own to obtain empirical data.
“People might be surprised to learn that FDA wants to advance simulation not just as a useful scientific tool but also as a regulatory tool,” Morrison said.
For more on Morrison’s upcoming talk at CAASE, visit the link here.