Research Group for Surgical Simulators Linz

PatientSim

PatientSim [1], [2] is our pilot project, in which we are developing a hybrid surgical simulator for cement augmentation techniques (see Figure 1).

Motivation

Vertebral compression fractures (VCFs) are mainly caused by osteoporosis and primarily affect the vertebral bodies of the thoracolumbar region [3]. Their incidence was 1.4 million in Europe in 2002 [4] and 700,000 in the U.S. in 1997 [5]. Three types of compression fractures with different degrees of severity can be distinguished. In nearly all types of VCFs, the cranial and caudal endplates of the vertebral body are disrupted to some extent [6]. A suitable treatment, conservative or surgical, should therefore only be selected after a detailed assessment of the fracture [7], [8].

PatientSim hybrid surgical simulator with patient phantom, instruments and simulated fluoroscopy display
Figure 1: PatientSim – a hybrid surgical simulator for cement augmentation techniques.

If surgical treatment is chosen, cement augmentation techniques such as vertebroplasty or kyphoplasty are very common [9]. In both interventions, artificial bone cement (mostly PMMA, polymethyl methacrylate) is injected into the affected vertebra(e). In kyphoplasty, an inflatable balloon is inserted beforehand to restore the original height of the vertebral body. This creates a cavity into which a more viscous cement can be injected at lower pressure than in vertebroplasty, which reduces the risk of cement leakage.

There are different ways of learning these procedures:

  • A mentor teaches novice surgeons directly on patients, which compromises patient safety.
  • Novice surgeons attend an intervention-specific course in which they operate on human specimens under expert supervision. Although such courses are close to reality, they have several drawbacks: besides high costs, availability and especially ethical concerns limit the use of human specimens. Furthermore, fluoroscopy is used for instrument guidance, which leads to radiation exposure and limits the training time.
  • A surgical simulator, the approach addressed by this project, overcomes these drawbacks [10], [11].

Mixed Reality Simulator

Because of its advantages over VR simulators and box trainers, a mixed reality simulator is being developed. The simulation environment is designed as a three-layered structure (see Research, Figure 1). The key element of the hardware layer is the patient phantom with artificial, physical vertebrae and soft tissue. The surgeon works on these structures with real surgical instruments, supported by a detailed visualisation. Image guidance is provided by a fluoroscopy simulation, which relies solely on 3D models of the anatomy and the instruments and therefore requires no X-ray radiation. The rigid tools required for transpedicular bone access are equipped with sensor coils, whose position and orientation in 3D are measured by an electromagnetic tracking system.

Computer application

At the core of the hardware abstraction layer is a C++ console application, the device server. Connected clients send scripting commands to communicate with specific hardware (e.g. the tracking system) via the User Datagram Protocol (UDP) or the Transmission Control Protocol (TCP). On the application layer, a graphical user interface (GUI) provides a detailed visualisation of the anatomical data. The GUI is written in C++ using open-source libraries such as Qt and Kitware’s Visualization Toolkit (VTK). Each anatomical structure can be loaded as a CAD model and registered using the calibrated insertion instruments. The surgical area is visualised in two-dimensional projections to improve the surgeon’s spatial perception. During each insertion trial, task-specific parameters (duration, instrument path length, etc.) are recorded for post-processing.

Patient phantom

The main components of the patient phantom are artificial vertebrae and artificial soft tissue. Three exchangeable artificial vertebrae (e.g. L2 to L4) are clamped in a torso made of silicone rubber and covered by artificial muscle and skin [12], [13] (see the research area Artificial Bones and Soft Tissue). After a training session, the penetrated vertebrae are replaced by new ones. Both the recorded task-specific parameters and the accessed vertebrae, which can be cut open with a band saw, can be used for a final debriefing with a medical expert.

Artificial trabecular bone

The inner, spongy structure of the vertebrae is imitated by open-cell polyurethane foams. We developed two groups of recipes with different mineral fillers, calcium carbonate and tricalcium phosphate (25–100 % by weight), and varying amounts of water as a blowing agent, and characterised them by compression testing and micro-computed tomography [14], [15], [16]. Apart from the structural anisotropy, their compressive and morphometric properties mirror those of human vertebral trabecular bone in good approximation, and they can even reproduce age-related changes of the bone structure. The calcium carbonate recipe is slightly better suited for mechanically correct imitations, the tricalcium phosphate recipe for morphologically correct ones. Further material compositions also reproduce the haptics of pedicle screw placement [17], [18], [19].

Vertebrae for cement augmentation

The first synthetic vertebrae used a closed-cell foam, which allowed realistic needle insertion but no balloon dilation or cement injection. We therefore developed open-celled synthetic vertebrae that support the whole procedure: transpedicular needle insertion, balloon tamp dilation (kyphoplasty) and bone cement application (vertebroplasty) [20], [21], [22]. They were validated against 38 human vertebrae from six donors and compared with commercially available vertebra surrogates (see Figures 2–4).

  • Needle insertion: the characteristic force pattern while passing the cortical shell and the cancellous bone was reflected by one of the synthetic vertebra types.
  • Balloon dilation: dilation pressure (13.0 ± 2.4 bar) and volume (2.3 ± 1.5 ml) were in good accordance with human vertebrae (10.7 ± 3.4 bar, 3.1 ± 1.1 ml).
  • Cement application: the application forces matched well, while the spreading pattern of the cement inside the vertebra differed from the human reference.
Fluoroscopy images of a needle inserted into a synthetic and a human vertebra
Figure 2: Fluoroscopy during transpedicular needle insertion: (a) synthetic vertebra, (b) human vertebra.
Fluoroscopy images of balloon dilation in a synthetic and a human vertebra
Figure 3: Balloon dilation (kyphoplasty): (a) synthetic vertebra, (b) human vertebra.
Fluoroscopy images with coloured cement distribution in a human and a synthetic vertebra
Figure 4: Cement application (vertebroplasty), the cement spread is shown in steps of increasing volume: (a) human vertebra, (b) synthetic vertebra.

Validation

The simulator was validated at the Klinikum Wels-Grieskirchen with 43 participants: 22 novices, 12 intermediates and 9 experts in image-guided surgery and pedicle tool insertion [23], [24]. Each performed a transpedicular bone access through both pedicles into an artificial L3 vertebra, locating the pedicle with a tracked probe and inserting an 11-gauge bone access needle under simulated fluoroscopy.

  • Face validity: 95 % of the novices rated the attractiveness and 91 % the general appearance of the simulator as good or very realistic.
  • Content validity: 92 % of the intermediates and 89 % of the experts would recommend the simulator; all experts found it useful for practising image-guided surgery skills.
  • Construct validity: a simulation-based performance score, computed from the tracked instrument paths, the number of fluoroscopy images and the time needed, distinguished experts clearly from novices (p = 0.001) and intermediates (p = 0.01).
  • Concurrent validity: the performance score correlated strongly with the rating of a specialist using a standardised assessment (OSATS), R = 0.86.

The simulator thus provides a realistic tool for training image-guided spine surgery, and its automatic assessment reliably reflects the trainee’s level of expertise.

Funding

This work was supported by the Austrian Research Promotion Agency (FFG) within the programme line Cooperation & Innovation (COIN), project number 845436.

Figures 2–4 from [22], licensed under CC BY 4.0; white margins removed.

References

  1. David Fuerst and Andreas Schrempf (2012): PatientSIM - Development of an augmented reality simulator for surgical training of vertebroplasty and kyphoplasty.. In: BioMed 2012, Innsbruck Austria, 2012.
  2. David Fuerst and Marianne Hollensteiner and Andreas Schrempf (2014): A Novel Augmented Reality Simulator for Minimally Invasive Spine Surgery. In: Proceedings of the 2014 Summer Simulation Multiconference, pp. 28:1–28:5, 2014.
  3. F. Grados and C. Marcelli and P. Dargent-Molina and C. Roux and J. F. Vergnol and P. J. Meunier and P. Fardellone (2004): Prevalence of vertebral fractures in French women older than 75 years from the EPIDOS study.. In: Bone, vol. 34, no. 2, pp. 362–367, 2004.
  4. (2002): Incidence of vertebral fracture in europe: results from the European Prospective Osteoporosis Study (EPOS).. In: J Bone Miner Res, vol. 17, no. 4, pp. 716–724, 2002.
  5. L. J. Melton and M. Thamer and N. F. Ray and J. K. Chan and C. H. Chesnut and T. A. Einhorn and C. C. Johnston and L. G. Raisz and S. L. Silverman and E. S. Siris (1997): Fractures attributable to osteoporosis: report from the National Osteoporosis Foundation.. In: J Bone Miner Res, vol. 12, no. 1, pp. 16–23, 1997.
  6. F-D. Zhao and P. Pollintine and B. D. Hole and M. A. Adams and P. Dolan (2009): Vertebral fractures usually affect the cranial endplate because it is thinner and supported by less-dense trabecular bone.. In: Bone, vol. 44, no. 2, pp. 372–379, 2009.
  7. Wilfred C G. Peh and Peter L. Munk and Faisal Rashid and Louis A. Gilula (2008): Percutaneous vertebral augmentation: vertebroplasty, kyphoplasty and skyphoplasty.. In: Radiol Clin North Am, vol. 46, no. 3, pp. 611–35, vii, 2008.
  8. Pierre Hardouin and Paul Fayada and Hervé Leclet and Daniel Chopin (2002): Kyphoplasty.. In: Joint Bone Spine, vol. 69, no. 3, pp. 256–261, 2002.
  9. N. B. Watts and S. T. Harris and H. K. Genant (2001): Treatment of painful osteoporotic vertebral fractures with percutaneous vertebroplasty or kyphoplasty.. In: Osteoporos Int, vol. 12, no. 6, pp. 429–437, 2001.
  10. Roger Kneebone (2003): Simulation in surgical training: educational issues and practical implications.. In: Med Educ, vol. 37, no. 3, pp. 267–277, 2003.
  11. Shawn Tsuda and Daniel Scott and Jennifer Doyle and Daniel B. Jones (2009): Surgical skills training and simulation.. In: Curr Probl Surg, vol. 46, no. 4, pp. 271–370, 2009.
  12. Marianne Hollensteiner and David Fuerst and Andreas Schrempf (2013): Artificial Vertebrae for a Novel Simulator in Minimally Invasive Spine Surgery.. In: Biomed Tech (Berl), 2013.
  13. Marianne Hollensteiner and David Fuerst and Andreas Schrempf (2014): Artificial muscles for a novel simulator in minimally invasive spine surgery.. In: Conf Proc IEEE Eng Med Biol Soc, pp. 506-509, 2014.
  14. David Fuerst and Daniel Stephan and Peter Augat and Andreas Schrempf (2012): Foam phantom development for artificial vertebrae used for surgical training.. In: Conf Proc IEEE Eng Med Biol Soc, vol. 2012, pp. 5773–5776, 2012.
  15. David Fuerst and Marianne Hollensteiner and Benjamin Esterer and Andreas Schrempf (2016): Druckeigenschaften synthetischer Knochenmodelle zur Nachbildung des trabekulären Knochens von humanen Wirbelkörpern. In: Proceedings of the DWG Hannover 2016, 2016.
  16. David Fuerst and Sascha Senck and Marianne Hollensteiner and Benjamin Esterer and Peter Augat and Felix Eckstein and Andreas Schrempf (2017): Characterization of synthetic foam structures used to manufacture artificial vertebral trabecular bone. In: Materials Science and Engineering C, 2017.
  17. Marianne Hollensteiner and Peter Augat and David Fuerst and Benjamin Esterer and Stefan Gabauer and Klaus Pueschel and Falk Schroedl and Andreas Schrempf (2017): Novel synthetic vertebrae provide realistic haptics for pedicle screw placement. In: Proceedings of the 39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2017.
  18. (2018): Procedure-specific validation of artificial vertebrae. In: 2018.
  19. Marianne Hollensteiner and Benjamin Esterer and David Fürst and Andreas Schrempf and Peter Augat (2019): Development of open-cell polyurethane-based bone surrogates for biomechanical testing of pedicle screws. In: Journal of the Mechanical Behavior of Biomedical Materials, vol. 97, pp. 247–253, 2019, ISSN: 1751-6161, (Funding Information: The Research Group for Surgical Simulators Linz (ReSSL) acknowledges the financial support by the Austrian Research Promotion Agency (FFG) within the program line Cooperation & Innovation (COIN) and project number 845436. Publisher Copyright: © 2019 Elsevier Ltd Copyright: Copyright 2019 Elsevier B.V., All rights reserved.).
  20. Marianne Hollensteiner and Markus Samrykit and Michael Hess and David Fuerst and Andreas Schrempf (2015): Development of Trabecular Bone Surrogates for Kyphoplasty-Balloon Dilatation Training. In: Conf Proc IEEE Eng Med Biol Soc, 2015.
  21. Marianne Hollensteiner and Markus Samrykit and Michael Hess and David Fuerst and Benjamin Esterer and Andreas Schrempf (2016): Inexpensive bone cement substitute for vertebral cement augmentation training. In: Proceedings of the 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2016.
  22. Marianne Hollensteiner and Melanie Botzenmayer and David Fürst and Martin Winkler and Peter Augat and Sabrina Sandriesser and Falk Schrödl and Benjamin Esterer and Stefan Gabauer and Klaus Püschel and Andreas Schrempf (2018): Characterization of polyurethane-based synthetic vertebrae for spinal cement augmentation training. In: Journal of Materials Science: Materials in Medicine, vol. 29, no. 10, pp. 153, 2018, ISSN: 1573-4838.
  23. David Fuerst and Marianne Hollensteiner and Andreas Schrempf (2015): Assessment Parameters for a Novel Simulator in Minimally Invasive Spine Surgery. In: Conf Proc IEEE Eng Med Biol Soc, 2015.
  24. David Fürst and Marianne Hollensteiner and Stefan Gabauer and Benjamin Esterer and Klemens Trieb and Felix Eckstein and Andreas Schrempf (2018): Transpedicular Approach on a Novel Spine Simulator: A Validation Study. In: Journal of Surgical Education, vol. 75, no. 4, pp. 1127–1134, 2018, ISSN: 1931-7204, (Publisher Copyright: © 2018 Association of Program Directors in Surgery).