It is common when developing a new energy-based medical device platform. The platform is a system of interrelated parts, each with its own design team. The energy delivery team (i.e. “hardware”) is tasked with designing a system that delivers the right amount of energy at the right time and incorporates the required sensing and control systems to do it consistently and safely. The hand piece team (i.e. “disposable”) must deliver the energy to the tissue while positioning the sensors to provide the desired information back to the hardware.

Reminiscent of the chicken and the egg, developing an entire platform often runs into a problem. The hardware side wants to know what the energy and sensing requirements are, but the disposable side wants hardware that can deliver energy before evaluating the prototype that was designed to provide the energy requirements.

Problem: the traditional medical device system prototype/test model requires system research tools to be built before evaluating a device prototype.

That is why platform level medical device design has an early stage where “research tools” are developed on the hardware side. To add to the complexity of the problem, the requirements range can be vast, making it difficult to build this first generation research tool. Delays in the development of the research tool to provide energy as desired with the level of control needed for the device itself are an early project challenge which has long-term effects on the product’s actual launch into the marketplace.

Once the research tool is decided upon and prototypes are being evaluated, there is always a question whether the testing results are skewed because of the disposable design or the hardware research tool isn’t delivering energy in the desired manner. So, all the testing must include measuring the research tool outputs as well as the end result effect on the tissue, requiring troubleshooting of the entire system at every stage.

Result: Delays in the research tools often extend timelines for prototype evaluation, which can delay the project by months.

Using computational modeling and simulation (CM&S) early in the platform development process can alleviate this problem. Drawing on knowledge of the desired end tissue effect and the physical mechanisms that create it, CM&S can be used to quickly narrow the energy delivery requirements for a hardware research tool. Virtual prototyping can also iterate disposable designs though the evaluation of the end result, allowing faster and cheaper design changes while still meeting the project’s timeline.

Using CM&S, early stage research and development for a medical device platform can benefit by:

• reducing the requirements range for the research tool, resulting in easier design and less expensive components.
• facilitating the disposable design iterations through virtual prototyping, allowing for intelligent design decisions before the research tool is available.
• significantly fewer physical prototype iterations leads to a shorter time to market, positively impacting the ROI for the project.

These benefits stem from using computational modeling and simulation to look at the end tissue effect and how it is affected by the system design. By including the hardware and disposable impacts on the desired result earlier in the design decision making process, the entire development process is streamlined.

Solution: Use tissue effect simulation to tighten the requirements and make progress on the device while simplifying the research tool, reducing lengthy and costly requirements for both. This translates to faster time to market, an earlier platform launch, and more revenue from the completed system.

Arlen Ward, PhD, PE is a Modeling and Simulation Principal for System Insight Engineering, LLC. SIE provides computational modeling and simulation services for medical device companies with a focus on the interaction between energy and tissue. Follow SIE on Twitter or visit the company website at systeminsightengineering.com for more information.

What makes computational modeling and simulation of medical devices so unique? It is the same thing that makes them medical— the interaction with biology. When you are running a stress analysis on a device mechanism or a thermal analysis on enclosed electronics, the modeling is the same for medical devices as for other products. When your questions involve the interaction with tissue the simulations start to take on elements that are unique to medical devices.

Consider the following analyses:

• What laser wavelength and power will provide the desired surgical result for our new device concept?
• How much more power do we have to supply at the electrosurgery generator if we increase the size of our deployable electrode by 20%?
• What are the performance expectations in human trials for our device that has been developed using porcine models?
• Is our sensor placement optimized for the physiological parameter we want to measure?
• If we change our design/manufacturing method/algorithm to improve the device manufacturability, will it change the surgical effect in unintended ways?

Every one of these scenarios requires a computational model for biological materials in order to analyze with simulation tools. These tissue models range from simple to complex: from static and isotropic defined materials to nonlinear and partially reversible multiscale linked physical mechanisms accounting for anisotropy in tissue layers and induced biological response (like heat-shock or healing). Understanding the material science of biological tissue becomes key as the need for accurate and timely simulations comes to the forefront of device design.

Another complicating factor for device design is the variability of material parameters seen not only between but also within patients. Things like age, hydration, disease state, or scar tissue all play a role in tissue variability and quickly outstrip the relevance of static material properties. The ability of computational modeling and simulation to provide insight into the sensitivity of the result to the variability of the multitude of system inputs provides a powerful tool to know if the new design is likely to be effective across a wide range of patient populations. As simulation data is included more and more in 510(k) and PMA submissions to the FDA, knowing the key sensitive inputs to the device system will be the new expectation.

Arlen Ward, PhD, PE is a Modeling and Simulation Principal for System Insight Engineering, LLC. SIE provides computational modeling and simulation services for medical device companies with a focus on the interaction between energy and tissue. Follow SIE on Twitter or visit the company website at systeminsightengineering.com for more information.

Every once in a while I see a news report about someone hitting a first responder’s vehicle while it is stopped alongside the road. Given the number of flashing lights on an emergency vehicle these days, I can not imagine the driver doesn’t see them. It seems another factor is at play- it is the result of the driver’s focus. All the lights and activity has the focus of the driver, and their vehicle just follows their eyes. Bam.

“When a measurement becomes a target,
it ceases to be a good measure.”     –Goodhart’s Law

For difficult-to-articulate goals a common practice is to replace the goal with some sort of summary measurement.  Some value or statistic that gives the decision maker a view of the overall system.  The problem is while that approach may work for simple systems, as they become more complex we lose fidelity in our model if we rely too much on summary statistics.  One measurement doesn’t cut it anymore.  That measurement can hide complexity, hide problems, and obscure visibility into the system.  This leads to bad decisions being made off a poorly chosen measurement.

Model Fidelity: the degree of exactness with
which something is copied or reproduced

In his 2016 TED talk “The Surprising Habits of Original Thinkers” Adam Grant mentioned (9:50) that a lot can be determined about your job performance and work commitment based on the internet browser  you use.  This doesn’t mean to get better at your job you need to download a new browser. The measurement is not the target.  Another example was a nonprofit I learned about a few years ago.  They were collecting children’s books in order to distribute them to families that didn’t have any.  The basis of the charity was a study that determined that kids that grew up in homes with books in them did better in school that those that didn’t.  Clearly that meant we needed to put books in all the homes.  Starting a conversation about correlation and causation was not terribly welcome in the middle of a book drive.

Project and portfolio management are areas where this happens quite a bit.  Give the fuzziness of the forecast inputs, using net present value (NPV) as the metric for project decisions reduces a very complex system down to a single value is a prime example of decisions based on too little model fidelity.  It doesn’t take long before every project seems to have the requisite NPV.  Is that because the quality of all the proposed projects improved over that time?  That seems unlikely. I think Goodhart would agree.

If the decisions are important and your system is complex, it becomes extremely important to pick your metrics carefully and weigh them against your knowledge of the system.

Arlen Ward, PhD, PE is a Modeling and Simulation Principal for System Insight Engineering, LLC. SIE provides computational modeling and simulation services for medical device companies with a focus on the interaction between energy and tissue. Follow SIE on Twitter or visit the company website at systeminsightengineering.com for more information.

Of all the TED Talks I have seen over the years, a particular one has resurfaced over and over again. Benjamin Zander’s talk “The transformative power of classical music” has had a lot to teach me. And not always about music, this time it was about leadership.

Now, I had an amazing experience. I was 45 years old, I’d been conducting for 20 years, and I suddenly had a realization. The conductor of an orchestra doesn’t make a sound. But the conductor doesn’t make a sound. He depends, for his power, on his ability to make other people powerful. And that changed everything for me.

The lessons of a conductor translate easily into any leadership role, echoing Robert Greenleaf or Simon Sinek. The power of a leader comes from their ability to make other people powerful.

I realized my job was to awaken possibility in other people.

What a great job to have! To awaken the possibility in other people. To see their potential and work with them to realize it. But how do you know if you are succeeding?

How do you find out? You look at their eyes. If their eyes are shining, you know you’re doing it.

If the eyes are not shining, you get to ask a question. And this is the question: who am I being that my players’ eyes are not shining? We can do that with our children, too. Who am I being, that my children’s eyes are not shining? That’s a totally different world.

As conductors. As parents. As leaders. Who am I being that they are not realizing their potential? Self awareness and appreciation for those you serve are a powerful potential in all of us. A totally different world indeed.