Simulation design for procedural technique training: We need you!

Simulation design for procedural technique training: We need you!

The Veterinary Journal 196 (2013) 143–144 Contents lists available at SciVerse ScienceDirect The Veterinary Journal journal homepage: www.elsevier.c...

151KB Sizes 4 Downloads 46 Views

The Veterinary Journal 196 (2013) 143–144

Contents lists available at SciVerse ScienceDirect

The Veterinary Journal journal homepage:

Guest Editorial

Simulation design for procedural technique training: We need you!

When designing simulations for training individuals in the delivery of either medical or veterinary procedural techniques, a wide range of parameters must be considered in order to circumscribe a system that, ultimately, provides efficacy. What sensory modalities need to be emulated? Which are critical to learning and required to properly execute the procedure? The parameters to be considered can be as wide as when to use physical models versus synthetic representations, or more detailed, such as whether to employ surface rendering or volume graphics, or both, for visual representation. Are auditory cues essential for the emulation? Is force-feedback based on tissue correlates, or synthesized? Often the choice is not clear, and clarity only arises through multiple iterations of trial and error. Presented within this issue of The Veterinary Journal is an excellent example of a group looking to create a simulation environment for the specific procedure of canine venepuncture (Lee et al., 2013). In the article, the authors provide a quintessential overview, from their choice of a target audience to data acquisition, representation and integration, and finally evaluation of their approach to simulating a basic everyday veterinary procedural technique. The article provides an excellent ‘inside look’ at the determinants and choices that Lee and his colleagues pursued in their unique design to create a hybrid world, one of augmented reality, to provide a useful and usable venepuncture simulation environment. It should be noted that accurate emulation of what appears to be a straightforward and simple technique, is actually quite difficult. It is essential to acknowledge the inherent complexity of the execution of the coordinated task. Following the flow of data acquisition, representation, and evaluation, I would like to suggest some basic points to consider. The first consideration in designing a simulation is to decide on the target audience. Is it comprised of novices, intermediates, experts, or all of the above? This parameter will determine the level and sophistication of the simulation. It also defines how schematic or real the simulation should be. Excessive realism to convey simple relationships of related structures and procedural intervention is not necessary for training novices and its associated cost is often not warranted. However, if the target audience is higher-level intermediates, such as advanced residents or experts considering professional training, a high level of visual realism and interventional sophistication is necessary to engage the user. Without these elements, users might see the simulation as a trivial game, a time-wasting scenario where learning is unlikely.

1090-0233/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.

A second consideration is whether to employ physical models or to synthesize the structural representation, or a combination of both. Both have their advantages. Physical correlates can be constructed from different materials, some cost-effective and some costly. Depending on the source, they can provide striking verisimilitude. However, they also suffer from key limitations associated with physical models. These limitations include expense, poor correlation to actual tissue mechanics and response, breakdown with repeated use, and in many cases, creating a mess. A different approach is to synthesize the representation, whether visual, aural, haptic or all of the above. Increasingly, there is a trend to represent the requisite structures for simulation by using data acquired from clinical imaging i.e. computer tomography (CT) and magnetic resonance imaging (MRI). The advantage is that reconstructions from these sources can provide a patient/ pathology specific model, which in the case of rare conditions, might not be regularly encountered in a training program or residency. They can also be easily and inexpensively copied and disseminated. Additionally, 3D reconstructions from clinical acquisitions can be used in 3D printing, thus providing an accurate and relatively cost-effective method to ‘print’ a phantom model to use as a physical correlate. Techniques are available which use a single source for data and subsequently represent it at various levels of complexity i.e. schematic to simple to realistic. This provides amortization of the costs of acquisitions. Once acquired, the data are integrated into the simulation environment. Depending on the selected procedural technique, interface considerations must be defined. Again, is the visual schematic or realistic, or is the type of representation to be dynamically determined? Is the visual in color? Can coloration be obtained from clinical imaging such as endoscopy? Is it necessary to provide stereo visuals and/or auditory displays? Is the tissue rigid, like bone, or dynamic, like arterial structures, viscera, or muscle? Ultimately, the value of the system lies in its efficacy. That is, is the system useful and usable and is it constructive in providing transfer of proficiency from the synthetic environment to the real-world setting? Usually, this efficacy is compared with a reference standard. However, quite often, that reference standard does not exist. Many training institutions employ a variety of approaches. Furthermore, if a reference standard does exist, it often has been developed de facto, and has not been subjected to the rigor of trial and validation. Especially in an academic environment, the introduction of any new methodology without subjecting it


Guest Editorial / The Veterinary Journal 196 (2013) 143–144

to the rigors of validation naturally leads to reluctance to its adoption. In closing, I would reiterate that human endeavors, even in what seems simple and straightforward, are often complex and difficult to emulate precisely. I would challenge all, students, residents, staff and faculty, to engage in the effort to simulate procedural techniques for training. Whether you are a system designer, or participate in the evaluation of a system as a human subject, your expertise is essential to the many trials, errors and multiple iterations that will be required to create useful and usable systems, ones that promote learning and the achievement of proficiency and lead to improved efficiency and outcomes in practice. As residency programs limit the number of duty hours, as expert faculty are under increasing demands for clinical production, as learning materials become increasingly rare and expensive, there is a need to adapt new emerging technologies, to improve them and to use them to provide engaging environments to purposefully practice and hone one’s skills. So please, get involved!

Don Stredney Ohio Supercomputer Center, The Department of Otolaryngology, The Ohio State University, Columbus, Ohio 43212, USA Ohio Supercomputer Center, The Department of Biomedical Informatics, The Ohio State University, Columbus, Ohio 43212, USA E-mail address: [email protected]

Reference Lee, S., Lee, J., Lee, A., Park, N., Lee, S., Song, S., Seo, A., Lee, H., Kim, J.-I., Eom, K., 2013. Augmented reality IV injection simulator based 3D medical imaging for veterinary medicine. The Veterinary Journal 196, 197–202.