Virtual ExperienceOctober 25, 2018

Academic Project of the Year: 2018 SIMULIA Winner

‘Project of the Year’ is an annual contest held by Dassault Systèmes where students…
header
Avatar Katha Sheth

‘Project of the Year’ is an annual contest held by Dassault Systèmes where students from around the world are invited to submit projects created with Dassault Systèmes products and software.  Winners are chosen by votes from Facebook users and an overall winner is chosen by a jury at Dassault Systèmes headquarters.

This interview with SIMULIA winner, Lance Frazer, was conducted by SIMULIA Academic Programs Specialist, Katha Sheth. Lance used Abaqus  for his project, “Subchondral Bone Cysts.”


Please describe why you chose this topic for your project. Why does this workflow need to be analyzed? 

So often, problems like these (subchondral bone cysts) are ignored by industry largely because they have their resources tied up in other projects that may be more lucrative or beneficial to the company. However, these problems plague thousands and thousands of people, and animals alike. Who’s going to look out for them? After dozens of years of being aware of the problem, we’re still unsure of how exactly these cysts develop, and how we’re supposed to treat them. If a child athlete develops one of these holes, their competitive career is over, as they’ll be sidelined for a year. If a military recruit develops one of these holes, their outlook on how they can serve their country is suddenly changed. We then need to consider the effects this has on the mental health of those dealing with these issues. We realize that bringing a group of healthy individuals and purposely damaging their bones to see if we can make a cyst develop is not a reasonable way to study the problem. Luckily, we’re in the boom of computational modeling, and a problem like this is perfect to study using such a tool. We also wanted to begin our work with the horse because our collaborator, Dr. Santschi is one of the smartest minds leading the research efforts behind this problem. Not only this, but it’s much easier and safer to go from an idea to an actual surgical implementation in a horse rather than a human. Once we get it fixed there we can move confidently into the human sphere. We believe our work is going to make a substantial impact and really change the way this problem is handled worldwide.

Describe how you executed the simulations.

We ran two different “groups” of simulations for this project. One was determining the effect of high impact loading on subchondral bone, and the other was to evaluate bone stimulus with and without a compressive lag screw across the defect. For both groups, the same basic procedures were used. I ran a quasi-static analysis with geometric nonlinearity accounted for and used *analysis=discontinuous. I used C3D4 elements (tetrahedral elements were my only choice given the complex geometry I was working with). I was able to determine via convergence analysis that C3D10 were unnecessary. Encastre boundary conditions were used on the bottom face of the tibia, and uniform pressure was applied to the top face of the femur. I implemented general contact (frictionless) for all articulating surfaces. I used grounded springs to help restrict the menisci from excessive translation (mimicking the joint capsule and MCL restraint). I used nonlinear connector elements for the MCL, LCL, PCL, and ACL. I used a rigid body tie constraint on the top face of the femur to apply rotational boundary conditions, such as no varus/valgus rotation. Lastly, I took advantage of *model change to swap elements in and out during the screw evaluation procedures.

Were there any key technical challenges you faced along the way? How did you solve them?

Oh, boy – plenty of challenges. Perhaps the most difficult challenge was to figure out how to reliably mimic the compression across the screw that comes from the torque applied in the surgical procedure. I was limited by my meshing software in that I couldn’t get the micro structure of the threading in the screw, nor did I have a clean, flat interior surface in the screw to apply a bolt load. I played with thermal contraction to get compression across the screw, but with this being a 3D model and plenty of material interactions, I wasn’t able to accurately determine how much compression I was actually getting. Also, since we wanted to test a wide range of models and procedures, we needed a method that was consistent every time we built a new model. I had an idea of removing the elements in the middle portion of the screw and pulling the two exposed surfaces together with a known, applied force. I am very grateful and fortunate that I have a mentor who’s well-versed with Abaqus. I ran the idea by him, and he quickly responded by introducing me to the *model change feature in Abaqus. I was able to do exactly what I wanted to do. I selected a group of elements in the middle of the screw that passed through the void and did not have any contact interaction with the surrounding bone and made them a set. I removed these elements in the first step, used connector elements with a specified pre-load, and let Abaqus solve this initial configuration. In the next step, I put the elements back into place, again with *model change, and applied the regular joint load to mimic light walking. It worked wonderfully, and I was able to get the exact compression I wanted in each model.

What were the advantages of using simulation in your project?

A problem like this is the perfect place for simulation for two reasons, which go with the two major objectives of this project. The first objective was to gain some better insight into how these holes develop. I wouldn’t recommend bringing in healthy individuals and banging their bones until a defect forms! With simulation, we can create these realistically high impact loads and safely observe whether or not the stresses and strains exceed a damage threshold. With that information we can look at it and say, ok, it’s very plausible that repetitive high impact loading is enough to create these subchondral defects, since we see very high stresses in the location they typically form. All of this information can be gathered without ever touching a bone! Secondly, we want to test surgical procedures. Just as I wouldn’t recommend banging on bones, I wouldn’t recommend bringing in people to do some trial and error surgical procedures! Using simulation, we can gain a lot of confidence in a certain surgery without ever touching a scalpel, and we can also begin to understand the mechanics of why or why not a surgery would work.

Why did you choose Abaqus over other simulation products?

I played with other software to see if we could accomplish the things we wanted to, but I always came back to Abaqus. Some other solvers struggled with the high degree of nonlinearity in the model, most notably with all of the contact. Abaqus was able to handle the nonlinearity relatively easily. I was also impressed with the arsenal of features available in the CAE. The documentation that’s available online was also a tremendous help. I don’t think the value of these resources online can be overstated. Plenty of blogs, technical documentation, and help websites are out there, and if I ran into a problem I was always able to find something out there that would help steer me in the right direction. Another feature of Abaqus that proved to be very helpful was its ability to handle initial overclosures. I had to come up with some creative meshing solutions to get high quality meshes, but when I began to add all the pieces into Abaqus, there were small overclosures and gaps as a consequence of me prioritizing mesh quality. These were handled nicely, and didn’t cause any issues.

What did you like best about using Abaqus?

Abaqus is the only software I’ve used that has been able to handle my model. The features like general contact and model change has made a world of difference in my simulation. The CAE is elegant and intuitive, and the ability to create and visualize custom outputs using python has been a game changer.

Do you feel that learning simulation skills in university will provide you with an advantage in your career? 

Absolutely! I think simulation is incredibly important now in industry and will only grow in its importance. The future of simulation is bright, and I’m very grateful that I have had an opportunity to learn it while studying here at the University of Kansas. I’ve just recently started searching for jobs, and I’m very encouraged by the high amount of jobs available for those with a simulation skillset.

Is there anything else we should know about this project? 

This project is highly collaborative, and I sincerely believe this work wouldn’t be successful without the different minds involved. After all, the transcondylar screw was not my idea, it was Dr. Elizabeth Santschi’s! She’s been performing this surgery all across the country and teaching the technique to surgeons worldwide. My advisor is Dr. Kenneth Fischer, a professor affiliated with both mechanical and bioengineering, here at the University of Kansas. He has been so much more than just an advisor to me, and I can only hope to mentor someone half as well as he has mentored me. And I alluded to her earlier – Dr. Santschi is my co-advisor and has been absolutely incredible to work with. She is truly brilliant. I tell every incoming engineer at KU, “go find yourself a clinician like Dr. Santschi before you pick up any medical related problem!” I can’t overstate how much admiration I have for both of them. They think about problems differently and between the three of us, we’re always able to come up with creative solutions. Also, having the different perspectives sort of creates our own internal checks and balances. There’s no use for a brilliant solution to a mechanical problem that has no feasibility in the clinical world! I could learn from these two as long they’ll let me. I also would like to acknowledge another mentor of mine, Chip Degrace. He’s a mechanical engineer working in the automotive industry, and is an Abaqus mastermind. If I ever have an idea of what to do, he usually knows the how. He has also been very supportive in this project and has had some valuable insight into things I hadn’t yet thought about.

What advice do you have for students just beginning to learn simulation tools?

Find a mentor who is willing to invest in you. Ask questions, and never get discouraged that you don’t know or understand how to do something. Simulation is so valuable and always keep the bigger picture in mind: you’re learning an incredibly valuable tool. And above all else, it’s genuinely exciting! I will never forget the first time I got my knee model to solve. The results were almost certainly wrong – but it solved! Stick with it and enjoy the process of learning.

What are your hobbies and interests?

I’m a huge hockey fan, so anytime my beloved Boston Bruins are playing I make sure to watch! I have 2 dogs that my wife and I love being active with. My friends and I are also pretty into the local trivia scene here. But as it turns out…a team of engineers makes a terrible trivia team!

Your favorite quote?

“A ship is always safe at shore, but that is not what it’s built for.” – Einstein


To read the full submission of Lance’s project, click here.

Stay up to date

Receive monthly updates on content you won’t want to miss

Subscribe

Register here to receive a monthly update on our newest content.