Simulators in catheter-based interventional radiology: training or computer games?

Simulation: Moving from Technology Challenge to Human Factors Success

“…Failure to predict performance in the real world can still occur, indicating that the goal of learning has not been attained [51]. Some skills learned in these circumstances might only be relevant to the task performed in the simulator, rather than to the real-world task it was intended to train [22,52]. As mentioned previously, it is even conceivable that the simulator may result in negative training.…”

“…Yet today's commercial computer simulator models still suffer from an unacceptable lack of precision in replicating soft tissue deformation characteristics, physiological behavior, and interventional tool properties. There is also a limited, often arbitrary, range of cases and instruments, commonly from a single medical device manufacturer, presenting a large number of performance metrics that may nonetheless fail to correlate with the critical steps of a task or procedure [20][21][22]. The reasons for these failings are many: computational complexity of real-time soft tissue and physiology simulation, business models based around distribution with limited functionality to promote sales of a particular medical device, lack of initial task analysis and involvement by SMEs, and insufficient quality control of content during development.…”

“…With appropriate development, it should be possible to differentiate between novice and expert performance by using these metrics. Unlike surrogate markers of expertise, such as fluoroscopy time and contrast medium volume [22], a properly developed, evidence-based simulator will use metrics derived from CPS identified from a TA. A novel computer-based liver biopsy simulator model developed in this way (Fig.…”

Using simulation for interventional radiology training

“…Some claims of construct validity for simulatorgenerated assessments have used metrics based, for example, on idiosyncratic opinion (e.g. ''catheter dragging'' in the aortic arch) or high-level events (procedure/ fluoroscopy time, contrast volume, end product) [25][26][27], although other workers dispute the relevance of such metrics [30,31]. Observer-based assessments could overcome technical shortcomings in using computational metrics within simulations, although adopting even validated surgical assessment tools might lose relevance when applied to simulated IR tasks.…”

“…Focusing on known strengths, such as training procedural sequencing, cognitive and communication skills, gross motor behaviours and teamwork within context, represent opportunities to use simulation as a component of a recognised IR curriculum [30,[34][35][36].…”