In any industry, the time to bring a product to market costs a significant amount of capital. This is especially true in the medical device design world. With the advent of advanced computational modeling techniques, including the non-linear finite element analysis capabilities of Structural Integrity Associates (SI), medical device designers now possess a tool to bring their products to market much more quickly, resulting in lower costs and faster revenue generation.According to Makeover et. it costs an average of $31 Million to bring a 510k medical device, and $94 Million to bring a PMA medical device from its initial proof of concept through FDA clearance . Sources of this cost include spending significant time and resources on the concept development phases as well as the clinical study phases. Delays in these areas can be extremely costly. Sheldon et. estimated in 2014, that an 8 week delay can cost a 30 person company $1.8 Million in lost time and revenue . If an additional 20 animal clinical study is required or 100 patient clinical study is required, a company should expect to see an estimated cost of $5.5 Million and $10.8 Million respectively . SI uses computational modeling to supplement the design iteration process and help reduce these costs in three primary ways:
Though the design cycle and process of medical devices can vary significantly in time and complexity, design teams nearly always follow the same general blueprint for taking a device from an idea through to a finished product. The gray boxes shown in Figure 2 show the typical design cycle for a medical device design team. To begin, a team identifies a need in the industry, for example, surgeons constantly ask for “a separate tool to seal vessels in young children”. Once a need is identified, the team begins the ideation process. The ideation process involves brainstorming and establishing a potential solution to address the need developed in step one. Often many ideas are chosen and taken to the next step, prototyping the design. During the prototype design phase, the engineering team develops a design of their initial protype or prototypes often using Computer Aided Drafting (CAD). These prototypes are then fabricated and evaluated for their ability to address the need established in the initial step. From the protype tests, new needs or different needs often arise, or the prototype does not perform as expected and the whole process begins again. Once the prototype is deemed to address the need at hand, clinical tests are designed and conducted and FDA approval is sought. Often, over the life of a project, engineering teams may traverse this design loop tens or even hundreds of times, testing multiple prototypes and refining their devices over time. Depending on the complexity of the device, it can take companies weeks or months just to produce and test one prototype. SI seeks to significantly change this process by using computational modeling to speed up the design iteration process, catch potential errors early in the process ,and limit and improve experimental testing.
As mentioned earlier, the design and testing of one physical prototype can take a significant amount of time and be costly. Typically, companies must develop several physical prototypes to arrive at a solution and often could benefit from even more iteration if time and money allowed. Computational simulation allows for design teams to speed up the design process in two primary ways. First, it allows designers to test prototypes virtually, eliminating the need of producing costly physical prototypes and conducting expensive tests. Secondly, it allows designers to converge on the optimal design much more quickly by providing a platform for testing immense numbers of prototypes through optimization schemes and parametric analysis.
It can often take several weeks or months and significant financial resources to procure materials, machine parts, and assemble a prototype. Frequently, all of this is done just to discover the device does not work as intended or tweaks need to be made; thus, another device must be built. With computational modeling, engineers can replace the prototype production and physical testing phases with computational models that often take hours and at most days or a couple of weeks to set up and run. Thus, engineers can gain significant insight into their design in an order of magnitude less time than it would take to build and test a full device, saving them essential time and reaching market (and revenue) much more quickly.
The second benefit of computational modeling’s speed is that it allows for many more design iterations to be examined and improved. For example, if a client is iterating in the design process to determine the optimal carbon fiber orientation in a device, it takes them 6 weeks to produce and test a prototype. After 3 prototypes and 18 weeks, they have discovered a workable solution but have not found what they deem to be the optimal solution. They must move forward because they have run out of time and money. To develop a computational model of the device would have taken 2 weeks, and each simulation 3 hours to run. Thus, in the 18 weeks it took to examine 3 prototypes, 128 different simulations and carbon fiber orientations could have been examined, allowing for significant improvement of the device and convergence on an optimal solution.
Not only does “virtual prototyping” or computational modeling allow for increased design cycle time, it allows engineering teams to discover potential errors in their device design by allowing them to examine scenarios not possible to reproduce using bench testing. Thus, many errors in device performance are not discovered until in-vivo clinical testing which occurs late in the product development phase. Changes to a device during or after clinical testing are an order of magnitude more expensive than changes during the design iteration phase of product development. Computational models allow for simulations to be conducted representing in-situ conditions before a prototype has even been built, allowing engineers to discover errors that prototype testing would not have caught and doing so when design changes are inexpensive to make.
Expanding on the benefits laid out in the previous section, computational modeling enables teams to conduct fewer experimental tests and informs the teams of what types of clinical tests will provide the most useful information. For example, a team is looking to examine a vessel sealing device’s impact on surrounding tissues. It is not possible to examine the thermal spread to surrounding tissue and organs during an arterial fusion with ex-vivo bench tests, so they must conduct expensive in-vivo animal tests. Additionally, the team is unsure if the in-vivo test will even produce the phenomena they desire to examine. By conducting computational simulations of the tests before performing any experimental tests, the team is able to examine if any damage will occur to the surrounding tissues due to thermal spread, providing the team with an answer to their question without going through the expensive process of a full animal study. Secondly, by conducting the simulations before testing, the scientists will gain insight into what the tests will likely produce, allowing them to make changes to the test, or examine if an alternative testing method would be more beneficial. Therefore, the team can optimize their financial and temporal resources and ensure that their clinical testing produces the desired results.
In this work, three arguments have been laid out demonstrating SI’s ability to drastically impact a design team’s process through the use of finite element modeling and computational simulation. With SI’s capabilities, engineers can reduce the time to market of their product, catch unforeseen errors, and improve their experimental testing process, enabling them to get better products to market faster, less expensively. Additionally, the cutting-edge work and research being done at SI is at the forefront of a surge of what may soon become commonplace in the medical device world, for “The FDA also believes that computational modeling is poised to become a critical tool for accelerating regulatory decision making. Continued adoption will be essential for advancing the FDA’s mission” .
 Makower, J., Meer, A. Denend, L., “FDA Impact on U.S. Medical Technology Innovation”, 2010
 Sheldon, M. “Accelerating Medical Device Innovation in the U.S.” The National Acadamies Innovation Policy Forum, 2014
 Morrison, T.M., Pathmananthan, P., Adwan, M., Margerrison, E., “Advancing Regulatory Science with Computational Modeling for Medical Devices at the FDA’s Office of Science and Engineering Laboratories,”, Frontiers in Medicine, 2018