While surgeon and surgical educator Howard Champion was conducting research and training for improving U.S. military combat care, he noticed the need for the development of a simulation-based platform for training surgeons in open surgery. Hoping to address this issue, Champion founded SimQuest (Silver Spring, MD) in 2001. The company set out to develop a platform for open surgical simulation that would be useful for educating trauma — an unconventional idea at the time, since most research was focused on minimally invasive techniques rather than open surgery. The technology has also naturally evolved into a potential training tool for surgical residents.
Since its inception, SimQuest has obtained most of its funding through grants from the National Institute of Standards and Technology (NIST), the National Institutes of Health (NIH), National Science Foundation, and the Department of Defense.
How it Works
SimQuest's surgical skills trainer uses physics-based simulation in an immersive desktop-type environment. It is designed not only for surgical residents seeking to practice their skills, but also for trauma surgeons, who could benefit from practicing with simulated, unpredictable scenarios similar to those they will face in the field.
For the initial learning content in the basic open surgical skill training simulator, instruction in basic instrument handling and surgical techniques is designed to be followed by technical exercises embedded in case scenarios of varying difficulty and complexity, with an emphasis on decision making, especially in adverse environments, e.g., when less than optimum information is available or under pressure.
Users are positioned to look down at their hands as they normally would at the operating table. The simulator supports expandable teaching content using rich media (for browser-based presentation) and learning management systems that allow remote monitoring of user progress. Low-level metrics extracted from the simulator's operation can be aggregated to produce proficiency assessment metrics.
SimQuest’s first surgical instruction scenario for the simulator teaches basic wound closure by suturing in support of the American College of Surgeons (ACS)/American Program Directors of Surgery (APDS) basic surgical resident curriculum. By subclassifying these activities into fundamental manipulation steps and deconstructing these steps into descriptions of the physical objects and their interactions, the company specified the required physics-based simulations needed for each interaction and used this as a guide for the design/construction of the simulator software and interface devices.
The guiding principles for the system’s physical configuration are that it must: (1) be compact enough to be easily moved; (2) be height-adjustable to support a broad range of individuals standing or sitting at the simulated operating table; and (3) enable users to hold real surgical tools.
In order to objectively assess and refine the simulation, SimQuest constructed a quantitative evaluation rig that allowed surgical tools to manipulate inanimate as well as appropriately prepared in vitro animal tissue, measured the loads between the tool and tissue, and measured tissue surface deformation.
Using this assessment rig, a series of data collection runs were performed with both inanimate and in vitro animal tissues, where the forceps were used to grasp the tissue surface and a number of movements perpendicular to the tissue surface as well as parallel to the plane of the sample, with the tissue surface shape measured at the end of the displacement simultaneously with recording the applied loads.
Two generations of systems have been built, with the second being an immersive OR table interface using a consumer-class 120 Hz stereo display with active shutter glasses combined with a two-handed haptic interface device configuration developed by modifying the consumer mass-market Novint Falcon to provide force and torque feedback as well as tracking of surgical tool jaws that provide closure feedback and measurement of strength of grasp (7 DOF).
For the initial wound closure scenario, SimQuest modeled a forearm consisting of multi-layer skin, fascia, muscle, nerves, blood vessels, and bone. Needle driver, forceps, and sutures were provided in order to allow learners to manipulate the wound for closure.
The company carried out extensive numerical simulation prototyping using multiple algorithms, both commercial (e.g. NVIDIA PhysX) and academic (e.g. PhysBAM), and found that it was necessary to develop multiple additional finite element codes (including a corotational approach) running on the CPU and GPU specifically adapted for tool-tissue interactions.
Multiple codes were developed to achieve an accurate and flexible simulation of suture thread interaction with tools and tissue resulting in a custom approach that incorporates known and novel mechanics/numerics techniques, combined with a continuous collision detection and response system, so as to provide complex real-time behaviors such as thread self-collision, needle and thread passage through tissue, thread and/or needle being grasped by more than one tool, and thread wrapping around tools, all supporting haptic feedback.
SimQuest implemented and assessed numerous collision detection and re - sponse approaches from the literature as well as commercial (PhysX) and open source (Bullet), and ultimately developed its own approach that supports continuous collision detection combined with force and linear complementarity geometric constraints which have been refined to resolve hundreds of simultaneous contacts of many types, including constraints that conflict with one another, e.g. pressing a suture into the tissue surface with a tool.
The system provides force/torque feedback from the tools interacting with the simulated surgical scene through a multi-rate approach that provides effective feel independent of the varying computational load on the multiple processor cores. This information is relayed to haptic devices that provide simultaneous two-handed 7 DOF interaction.
The simulator can extract metrics for objective assessment of proficiency, from geometric measures of tool-to-tissue activity as well as kinetic, force-based monitoring of interaction, by exposing the internal variables that reflect the ongoing physics state during the simulation as well as specific geometric measures of the tool-tissue interaction in the surgical scenario so that they are accessible via the Lua scripting language.
The system creates scenario-based surgical training content using a collection of software tools that were developed to form the equivalent of the classic computer gaming tool chain. By using these tools, new scenarios can be developed without having to write code with the resulting set of XML-based data files used to define the simulation when the program starts.
The didactic learning interface is implemented in HTML5 and constructed such that its display automatically adapts to whatever display device is used.
Where it Stands
The Open Incision Surgery Simulator system has been developed to the point of assessing the transfer of training for an initial scenario; plans for expansion of this content and further development of the system as a commercial platform are underway.
SimQuest is currently approaching its first hands-on assessment of the simulator at Old Dominion University in Norfolk, VA. The company plans to use the results of this assessment to update the system before moving into the production phase. In the future, the company hopes to gain validation from the American College of Surgeons as a recommended addition to the Surgical Skills Curriculum.
For more information about the Open Incision Surgery Simulator, visit http://info.hotims.com/34460-160.