(Product Development Insight) Virtual Product Development Tools—Innovation and Risk Management: Part 1 (MDDI archive, Nov 04)

	Email this Page
	Print this Page

Originally Published MDDI November 2004

Product Development Insight

Virtual Product Development Tools—Innovation and Risk Management: Part 1

Virtual processes and tools enable medical device design engineers to innovate and cut costs while they get products into the marketplace quickly.

Leslie Rickey
MSC.Software Corp.

Leslie Rickey

The migration of baby boomers into their sixties and seventies is accelerating demand for bio-medical devices critical to improving quality of life and lifesaving procedures. These devices require physical testing before certification, release for clinical trials, and ultimate use by doctors and patients. Physical tests are time-consuming, cost prohibitive, and often incapable of providing a complete understanding of a product’s performance, e.g., how it functions and responds in real-life applications.

Engineers of biomedical devices, including implantable orthopedic devices, prostheses, dental implants, and artificial limbs and organs, are discovering that virtual product development (VPD) processes and tools, including finite element analysis (FEA) simulation software, can provide a tremendous resource for understanding performance and moving biomedical devices through development, manufacture, and regulatory approval and into the hands of doctors and patients.

Through VPD, engineers gain a better understanding of product performance attributes and eliminate design problems. This, in turn, enables the development of cost-effective, innovative designs with higher levels of reliability in less time than with traditional processes validated by physical tests. The first installment of this article provides an overview of VPD and how it can help reduce product development risk and cost. The next installment will focus on using VPD to generate analytical data for support of the FDA approval process. It will also examine the role of VPD in identifying uncertainty in factors that affect the performance and function of a design.

Product Life Cycle Management

One bone can have multiple material properties. A femur has a trabecular bone at its ends and a cortical bone at its center.

VPD technologies are a subset of product life cycle management (PLM) technologies, which include all technologies related to product design and engineering, such as CAD, computer-aided engineering (CAE), data management, and others. Within the PLM technologies, VPD technologies are those that rely upon computer simulation to accurately predict product performance. With VPD, all phases of the product design process use an integrated combination of simulation software technology and traditional physical testing to simulate and correlate product performance.

The Limits of Physical Testing. The biomedical device industry spends hundreds of billions of dollars every year on research and development. Even so, products occasionally reach the market with something overlooked or an unfortunate combination of unchecked factors and circumstances leading to failure. Unfortunately, it is just too time-consuming and expensive to build hundreds or thousands of physical design iterations, test them all, and maintain schedules and cost requirements.
Depending on the types of materials and products, it may take as little as one day to make a prototype and the test process may take only a week. Of course, that time frame also assumes that soft tissue or bone specimens, if required, are available in time for the test. In situations where physical prototypes are easy to build, physical testing may be relatively inexpensive; however, VPD test methods can still deliver value. When product failures or surprises surface, one or more design iterations result, and repeat physical testing is required. Each iteration adds cost and ultimately delays marketing of the product.

Using virtual prototypes to detect problems or performance issues early in the product development process enables problems to be corrected quickly. The physical test process then becomes a validation phase, thereby reducing time to market.

While essential for certification, correlation, and validation, physical testing does not always lead to a thorough understanding of biomedical devices and their interaction with the human body. For example, the physical characteristics of soft tissue or bone structure from two human bodies are different, so when stresses and loads are applied during testing, the resulting information will not be repeatable from test to test. Therefore, studies to determine the best balance between conflicting design objectives (trade studies) and design of experiments (DOE) based solely on physical test results may not provide all the data necessary to reach a legitimate conclusion.

In addition, although physical testing of the interaction between soft tissue and devices such as stents is possible, it is very expensive. Because the physical test results are not repeatable from test to test, not enough information is generated about how the stent structure and its material properties react to different tissue types. Often, the time constraints of product development and the availability of appropriate soft-tissue specimens for physical testing are in conflict as well.

The situation is similar for bone: no two bones share identical geometry and material properties. Add to that the cost of test equipment, as well as the time for conducting experiments, such as heavy fatigue testing of joint-replacement products, and physical testing then becomes both an economic and time issue.

Virtual prototypes can detect problems early in the development process.

VPD can help balance the risks associated with new product development by providing the ability to simulate the reaction of complex materials to stress, temperature, fatigue, and other strain or loading scenarios. This insight provides engineers with a better understanding of device behavior and interaction. With this knowledge, engineers can make better decisions. For example, an engineer can determine whether a new material or concept is feasible and what, if anything, should be redesigned.

Simulation Accuracy. The accuracy of simulation is affected by several factors, including correlation with physical test results, physical or material properties, and convergence. Correlation of simulation is critical for medical device manufacturers—as it is for any manufacturer. Correlation can be done for any given test or series of tests. For example, if an intraocular lens manufacturer runs a compression force test on a lens, the simulation results can be validated by comparing them with physical test results, ensuring that the simulation model and the process correlate with reality. The entire physical test process can be simulated, including the device itself and the physical equipment used to perform the test. The virtual test results can be shared across the enterprise, made repeatable, and refined over time.

Precise material or mechanical properties are required for accurate bio-medical FEA analysis. The difficulties with determining these properties include variance between human and animal tissue properties, as well as differences between properties captured during ex vivo and in vivo tests. However, methods are being refined for hydrating soft tissue. This improves correlation between ex vivo and in vivo properties, leading to more accurate soft tissue properties.

Many different material models of soft tissues have been developed. These include models of simple linear elastic materials, hyperelastic materials, viscoelastic materials, and even poroelastic materials. As in any situation, trade-offs, such as cost versus benefit, must be considered. For example, a number of properties can be used for simulating the interaction between a stent and arterial tissue. A linear elastic representation of arterial tissue is simple and computationally less expensive than a hyperelastic model. Essentially, a linear elastic material model uses a single number for material properties.

In most situations, this model is inadequate for representing the response of an artery over the entire range of deformations that can occur during delivery of an interventional device. A hyperelastic model or a material model with plasticity may provide a better representation over the entire range of deformation to which an artery may be subjected during device delivery. However, using these types of material structure models is computationally more expensive. Increasing the complexity of the material model adds an additional cost for developing and running more complex experiments to define multiple material parameters or coefficients.

At the other end of the spectrum from soft tissue are bone structure material properties, which tend to be much simpler. However, different types of bone have different properties and one bone can have multiple material properties. For example, a femur has trabecular bone at its ends and cortical bone in the center. Trabecular bone consists of connected tubular shapes and is porous, while cortical bone is very dense.

In addition to the accuracy of material properties and correlation with physical tests, another factor in determining the accuracy of simulation is convergence. This is the use of internal mathematical tests to determine whether a solution is accurate. In theory, the more elements in a model, the closer it is to reality. For example, the control point of a model with a low number of elements is observed for changes as the number of elements is increased. Using the same model, the same load conditions are applied, and the control point is observed again. The process is repeated until the position of the control node no longer changes when the number of elements is increased. That is when the finite element model converges to a single value.

Innovative Product Development

Introducing innovative new products faster than competitors is one key to winning and capturing market share, especially in the competitive medical device market. The faster innovative designs can be evaluated, certified, and put into production, the greater the opportunity for beating competitors to market and gaining the edge in market share. Overall, VPD helps reduce product development time and costs by giving device manufacturers a range of simulation technologies with which to test products virtually for performance, durability, fatigue, drop, safety, and stress.

VPD includes the application of computer technology and simulation software techniques in all phases of the biomedical device design process. Biomedical applications range from the introduction of exotic materials subjected to unique environments to simple devices with extremely low tolerance for failure, including implantable orthopedic devices, prostheses, dental implants, and artificial limbs and organs. Using VPD tools, biomedical device engineers can virtually test almost every conceivable—and inconceivable—circumstance, enabling them to create innovative product designs while minimizing costs and the company’s risk of financial exposure caused by a product’s failure.

In designing an elliptical accommodative intraocular lens, simulation was an effective tool because it helped prove conceptual principles even before the prototyping stage.

Using simulation early in the design process, such as during the conceptual stage before the design path has been established, allows for virtual testing and validation of many design iterations to determine the best performance before building a single prototype. With more design iterations comes a better understanding of the design, enabling tighter bandwidths (rather than estimated safety factors) and optimization before a device is ever built. Early simulation enables engineers to model soft-tissue interaction, visualize the motion and stresses on the smallest mechanical and biologic structures, and model the latest breakthroughs in material science.

An Early Simulation Example. An innovative elliptical accommodative intraocular lens (EAIOL), developed by Mona F. Sarfarazi, MD, FICS, is a concept enabling the brain and eye muscles to focus a synthetic lens for close-up and distance vision. With the use of simulation, the principle behind the Sarfarazi EAIOL concept was proven before prototypes were made. Simulation enabled optimization of the EAIOL concept, providing a 50% reduction in time to market, and leading to a 50% reduction in manufacturing costs.

The design and function of the EAIOL dictated its basic shape: two lenses connected by a membrane. Intuition and initial studies indicated the lens shape could not have a significant effect on the motion of the EAIOL assembly. Compared with the haptics, the lenses are much stiffer and move as a rigid component. The challenge was to determine a design delivering the necessary motion of the lenses using available materials.

The focusing or accommodation process in the human eye occurs by a reshaping of the lens, occurring when the eye muscles relax and the fibers connecting the muscles to the lens, the zonula of Zinn, pull outwardly on the lens. The intraocular lens achieves the same result by a slightly different technique. Accommodation in the EAIOL is achieved by the pull on the connecting membrane between the lenses from the zonula of Zinn moving the lenses closer together.

Measurements indicated that the zonula pull outwardly on the human lens, and the value was applied to the finite element model as a prescribed displacement. Optics analyses indicated that an acceptable range of focal lengths could be achieved with approximately 2 mm of relative motion of the lenses. Given the specified boundary conditions, the geometry and material could be varied to achieve the desired objective. The geometry variation is somewhat limited, because the lens assembly must fit into the bag of the natural lens. The elliptical shape of the final cross section mimics the shape of the human capsular bag, where it will be inserted.

Also of interest during the initial analyses was the fact that if the cavity in between the two lenses was filled with a trapped fluid, the effective stiffness of the lens assembly increased significantly. This stiffness increase was detrimental to the performance of the IOL, so a vented design was applied, enabling internal fluid to migrate into and out of the cavity in the chamber in front of the anterior lens. Although stiffness was significantly reduced, results of the analyses indicated a level of accommodation much less than the desired 2 mm.

To reduce the stiffness and increase the accommodation, engineers simulated an interrupted solid of revolution where the haptics occupied only three 40 segments. This design resulted in an accommodation of 1.9 mm. The nonlinear nature of this response curve is characteristic of this type of system and allows a larger motion in the relative lens movement than in the zonula.

Numerous materials were evaluated for the purpose of this study. In general, softer materials were much more desirable because they provided a higher level of accommodation. However, the interrupted solid of revolution design allowed the use of polymethylmethacrylate, a relatively stiff but well-tested material proven for optics use. Additional analyses were run on the assembly, including stress, stress intensity, and tensile stress.

Conclusion

VPD gives device manufacturers a range of simulation technologies with which to test products for performance, durability, fatigue, safety, and other issues. By using virtual prototypes to detect problems or performance issues early in the product development process, problems can be corrected quickly. Making the physical test process a validation phase then reduces time to market.

VPD can be implemented easily, and it provides return-on-investment metrics in days and weeks. The software tools are scaleable and built for everyone from novice designers to PhD analysts to provide rapid process improvements and hard cost savings. Using physical testing for corroboration, VPD processes and tools enable medical device design engineers to innovate and move products through the approval cycle and into the marketplace as quickly and cost-effectively as possible.