Using CT-Based CAE in Plastic Surgery

Posted on 31 January 2019 by Jessica James

Professor Tomohisa Nagasao of Kagawa University was recently interviewed by our Japanese resellers JSOL. They talked about using CAE in plastic surgery. Professor Nagasao discussed how pre-surgical simulation using CT data and medical image segmentation in Synopsys Simpleware software is offering valuable insights for his work.

Q1: When did you start using simulation for your work as a doctor?

Professor Nagasao: It was around 1997 or 1998 that I had the idea of using structural analysis and the finite element method in my work. I developed my own program, but could not do what I wanted with my research, so I started using commercial software from about the year 2000.

Q2: What was your opportunity for using simulation?

Professor Nagasao: In plastic surgery, we can extend or change the shape of the human body, for example bones and skins, muscles, and so on. The more I researched simulation, the more useful it seemed for solving problems to do with the body in plastic surgery.

Q3: How did you learn engineering?

Professor Nagasao: When I was starting as a clinician, I studied the finite element method and structural analysis. I originally liked mathematics, so once I understood the method, I decided to play with real software. Going from medical science to engineering is often thought to be difficult, but I saw the possibility of combining them and losing these preconceptions. For example, in a cheekbone fracture, you need to return the bone to its original position using a small screw called a miniplate. However, if you strike the screw improperly, it will collapse.

The same thing happens with bones – let’s say you want to adjust the bones and stop the plate at a certain point. The bone is attached with muscles around it and receives pressure from the skin, so it is common for the screws to come off and the small plates with screws to bend. This is like designing the hinge for a door, and for patients I saw that the burden is completely difference if the screw is 2mm in diameter, or 4mm in diameter. If you strike a screw with a large diameter, it can cause problems later with the wound.

So you apply engineering principles to restore and fix the position of the bones in surgery. Changes to the human body should be planned more precisely than houses, so simulating the procedure makes it more efficient for the patient.

There is a disease in my special field called funnel breast, where the ribs are deformed. You can create a 3D model from CT images of the patient, and then 3D print the model so you can see the bone and cartilage separately. The model is useful to see how the parts fit together, similar to what a carpenter is doing with say, a CAD design in architecture.

You can also apply this thinking to diseases like cleft palate, which affects about 1 in 500 people. Surgeries to fix the cleft palate have evolved from older methods, but there is still a risk of sewing too forcibly, and reopening the wound – so engineering methods are again useful for better understanding how to work on the skin.

Q4: How can CAE be used in the future for plastic surgery?

Professor Nagasao: Before surgeries, we use a 3D model reconstructed using Simpleware software from the patient’s CT data to simulate the procedure. Improving surgical accuracy is important as human bodies will bleed a lot even if the bones are shifted by 1cm. If you do not plan surgeries extensively, the operation will take longer than needed, compared to working from a clear plan. For example, if you have a plastic model, it is useful, but can’t take into account factors like the eyes or skin. You can use simulation to evaluate these risks and improve the surgery, like you would if you were building a house.

Q5: Is there anything you’d like to see from image-based CAE in the future?

Professor Nagasao: It’s hard to model cartilage in three dimensions in the same way as bone from CT or MRI. It is important for surgical planning to take into account these issues like cartilage without manual work. Pattern recognition might help with this, in addition to thresholding, to make things easier. Taken further, patient-specific data could be automatically extracted as a finite element model to speed up workflows.

JSOL: These are currently difficult requests, as CT scanning does not capture cartilage, and it is hard to segment cartilage from MRI. Artificial intelligence and the use of automatic recognition and databases might help in the future. It is currently very difficult for engineers to separate a truly small color difference in a scan, such as between cartilage or ligaments, so it would be nice to add in a doctor’s experience to achieve this.

Professor Nagasao: I think some parts of simulation are in the future, but for now I think many areas can already be captured quite accurately. In the world of medicine, I think that crossovers with engineering fields will become more common, so that the benefits of this research will increase in medicine.

Special Thanks To:

Chief Professor Tomohisa Nagasao

Official website: https://www.kagawa-u.ac.jp/english/

Original interview (in Japanese): https://www.jsol-cae.com/product/tool/simpleware/cases/caseI03/#breadcrumb

Contact Us

Want to know about how we can help with your workflow? Contact us at simpleware@synopsys.com, or try the software with a free trial.