Simulation in urology – a role for virtual reality?

Authors


 Miss Jyoti Shah, Department of Urology, St Mary's Hospital, Praed Street, London W2 1NY, UK.
e-mail: jyoti.shah@ic.ac.uk

Introduction

The teaching of surgical skills has traditionally followed the Halstedian master-apprenticeship model, where surgeons learn to operate on the philosophy of ‘see one, do one, teach one’. Although this model relies heavily on opportunity, based on the random flow of patients, it has certainly produced excellent surgeons over the years. However, healthcare provision has changed, as has surgical education.

The introduction of the Calman Report for Specialist Training, and the New Deal for junior doctors' hours, has made training more structured, although this is within a shorter period. This, combined with the government drive to make the NHS a consultant-led service, reduce waiting lists and increase the number of day-case procedures, has also reduced the chance that trainees have to learn to operate within the operating theatre; training inevitably adds time to a list and therefore reduces the patient throughput. Finally, the increasing expectations of patients are such that the issue of trainee surgeons undertaking surgery is coming under increasing scrutiny, and some patients will refuse to allow a trainee to be the primary operator in their surgery.

Inside or outside the operating theatre?

The operating room has been described as ‘the surgeon’s classroom and laboratory extraordinaire' [1], and many have argued that it is the best classroom for surgical education. However, there are several disadvantages to training within the operating theatre; there are always numerous distractions that have priority over surgical education. Indeed, most of the distractions will have nothing to do with teaching [2].

Furthermore, it is well known that the operating room is a stressful environment, as a result of many factors, e.g. time constraints and the need to complete the operating list on time, technical problems with the procedure itself, which may be beyond the trainee's abilities and therefore of no educational value, problems with the equipment, new theatre staff who may not be familiar with the way the operation is carried out or the surgeon's idiosyncrasies, the constant movement of people in and out of the theatre posing distractions and finally, concern for the patient. The trainee may also arrive highly stressed or exhausted. Much work has been carried out on the effect of stress on learning and it has been shown that the greatest learning occurs in a moderately stressful environment. Where fine motor control is required in combination with cognition, low stress levels enhance performance [3].

It is also possible that the trainer may not be a good teacher, which further questions whether the operating theatre is the best place for the trainee to learn to operate [4]. Other issues, such as the cost of training in the operating theatre and the risk to the patient if a trainee operates without adequate training or supervision, also argue against the use of the live situation in the operating theatre as the ideal teaching environment.

On the other hand, teaching outside the operating theatre in a simulated environment has some clear advantages. Trainees can be taught in a controlled and stepwise manner such that individual components of a procedure can be learnt and eventually put together, and the training can be tailored to the individual needs of the trainee.

The use of simulators allows trainees to practice, a vital component of learning any motor skill, according to published work on how motor skills are learnt. With appropriate feedback, trainees can operate on a patient for the first time but have already gained some experience and skills [5]. There are many types of simulation that have become established over time in surgical training, e.g. latex models of skin, artificial organs and other tissues, fresh tissue such as an anaesthetized pig, animal parts such as pigs feet for suturing, and virtual reality.

Virtual reality

The term ‘virtual reality’ (VR) was first coined by Jaron Lanier, who founded the first commercial enterprise, VPL Research, in the late 1980s. It is defined as a human-computer interface that simulates realistic environments as a three-dimensional (3D) digital world, whilst enabling participant interaction [6]. Simulation has been used in the aviation industry for several years and it is commonplace for pilots to spend hours ‘flying’ in a flight simulator before flying a real aircraft. Particular tasks in the simulator require a specified number of hours before certification, and it has been shown that trainees could reduce the number of hours of in-flight training by spending time in the simulator [7].

There are many similarities that can be drawn between pilots and surgeons; both have to learn to manage stressful and potentially life-threatening situations that are unpredictable and subject to change at any moment. The benefits of VR simulation noted in the aviation industry have inspired attempts to bring VR into surgical training.

Elements of a VR simulator

There are many requirements of a VR training tool. From a clinician's perspective, the procedures that would benefit from VR training need to be defined and in the first instance simulators need to provide training for common procedures. They should provide accurate detail, be anatomically precise and be highly interactive. From an industry perspective, factors such as the number of trainees need to be considered, as well as the frequency with which an operation is performed. Obviously, development costs are likely to prohibit the creation of simulators for more specialized procedures that are rarely performed. Stacey summarized the desired features, shown in Table 1[8].

Table 1.  Elements that need to be incorporated into medical simulators [8]
FeatureComment
Visual realitySimulators need sufficiently high resolution to look realistic and like a patient.
Physical realityThe system needs to be interactive, as in real life. The organs need to be elastic and there should be
dynamic realism when organs and tissues are touched or grasped; they need to bend and deform,
as they do in reality.
Simulator devices need to be able to react to forces applied by the trainee.
Physiological realityTissues and organs need to show signs of life, e.g. organ peristalsis. They also need to react with reality
when manipulated, e.g. when tissues are cut they should bleed or leak tissue fluids; muscle needs to
show contractions and tissues to bruise in appropriate situations.
Tactile realityThe trainee needs to be able to feel forces and pressure between the medical device and the tissues.

Advantages of VR

VR allows trainees to learn real-life events that have been simulated on computers, without causing any discomfort or risk to the patient [9]. VR avoids the feeling that many patients experience when operated on by trainee surgeons, i.e. that they are being ‘experimented upon’. VR allows the trainee to practice each task as frequently as is necessary before undertaking the entire procedure. The various theories surrounding the learning of motor skills suggest that repeated trials or practice are essential for learning, and that the maximum benefit is obtained when simulations closely approximate the real environment in which the particular skill will eventually be used. When this practice is combined with appropriate feedback from the trainer directed at specific weaknesses, the training situation is ideal. Many of the simulators also allow ‘action replay’ facilities so that both the trainer and trainee can evaluate performance.

Trainees have different learning rates and this is one of its greatest criticisms of recent ‘streamlined’ specialist training; not all trainees can learn the necessary skills within the designated 5 or 6 years. VR allows training out of the operating theatre to be tailored to the individual's needs, and avoids the embarrassment of slow progress around peers. The endpoints of the task that is being practised can also be altered to meet the trainee's needs, and the simulated operation can be abandoned when the trainee feels ‘saturated’.

Training on VR also allows convenient learning, in that the trainee can learn when time allows and does not have to wait for a particular operation when there is scarcity of in-patients upon which to operate. When training in the operating theatre, much operating time is lost and therefore VR training does not slow the progress of the lists or reduce staff time for training.

Many of the VR systems have facilities for objective data collection, therefore allowing objective feedback of variables such as the time taken to complete the procedure, error rates and economy of motion, so that with practice the trainee can aim to improve performance.

These advantages also translate into financial benefits; the presence of surgical residents in the operating theatre in the USA has been estimated to cost $53 million per year for general surgery alone and it is thought that once VR training becomes established the cost of training will be considerably reduced [10].

Disadvantages of VR

At present VR training is in its infancy and is used primarily as a research tool. Initial studies validating existing simulators have shown that simulator task performance is related to actual task performance [11]. However, transfer efficiency ratios need to be developed for VR simulators to give trainers an indication of what the equivalence of time spent on a VR simulator is in terms of time spent operating on real patients [12]. Systems also need to be modified so that a moderate element of stress is added to VR training, as this is known to be the optimum learning environment.

Although excellent graphics are available, there remains a time delay in the screen-refresh rate so that rapid interaction between the computer and user is currently lacking. However, computers are doubling their processing power and halving their size every 18 months (Moore's Law) and thus as computers become more powerful this problem can be addressed.

Current systems also lack haptic feedback, which mediates the sense of touch, and at present there is no sensory input of pressure or texture. The tissues and organs do not deform when grasped or touched, and there is no bleeding, damage to tissues or organ peristalsis. Computer algorithms need to detect when objects are in contact with each other and when there are forces between them [13]. Available systems are also costly and not affordable by all training institutions; this raises issues of the source of funding for VR training

Current applications in urology

Teaching anatomy

The National Library of Medicine in the USA has made available the ‘Visible Human Project’[14], which has created a large database of 15 GB of computer memory consisting of digitized CT, MRI and tissue images that were generated at 1 mm intervals from a 39-year-old convict on death row. More recently the organisation has developed a ‘virtual female’ dataset.

Information was extracted from the Visible Human dataset and 3D pelvic reconstructions generated that are pertinent to male pelvic anatomy [15]. Insight into the relationship of pelvic structures can be gained by rotating these 3D images, and these models serve as useful tools for teaching and investigating male pelvic anatomy. As computer software advances, trainees will be able to perform virtual dissections, but ultimately these reconstructions will be used in VR simulations of urological procedures.

DRE simulator

In men, prostate malignancies are the second leading cause of death and up to 25% of patients die from the disease [16]. Although disease that is confined to the prostate is thought to be potentially curable, up to a third of all patients with prostate cancer have advanced disease at the time of diagnosis [17]. An early diagnosis is therefore crucial and at present screening tools include the DRE, PSA assay and TRUS.

Traditionally, students train to undertake a DRE on a mechanical simulator that consists of a rubber model of the prostate, with beads inserted into it to simulate malignancies. A plastic cover blocks the trainee's view of the phantom during diagnosis and a rotating plate allows the trainee to switch prostate types (Merck's Procar Simulator, Heath Edco, ‘Prostate training kit’, Waco, TX, USA). This is followed by training on patients, although this relies on the availability of patients, their willingness to allow medical students to train on them, and the discomfort that the procedure may cause the patient when carried out by an inexperienced trainee. A major problem is also the lack of feedback and therefore mistakes cannot be corrected. One study has shown that only 31% of medical students routinely undertake DREs, and that this lack of adequate exposure and training eventually leads to unnecessary specialist referrals [18].

One solution to this problem is the use of VR simulation and has been addressed by Burdea et al.[19]. Twelve ‘virtual’ patients were diagnosed using a ‘PHANToM’ haptic interface that provides feedback to the trainee's finger during the DRE. They undertook human factor studies on medical students and urology residents, finding that both groups were able to palpate and correctly diagnose the virtual cases presented to them. Their results are encouraging and show the feasibility of a VR-based DRE simulator, although improvements are needed [19].

TURP simulator

TURP is an essential skill required by every urologist but reduced training opportunities, pressures of waiting lists, alternative treatments for BPH and problems with litigation have diminished the number of TURPs that trainees now perform during the course of their training. Again, VR training has a potential role in this situation.

One group has developed a computer-based VR system to create 3D images of the prostatic lumen and resectoscope loop. User interaction via a magnetic sensor attached to the resectoscope has five degrees of freedom. The software allows movements of the resectoscope and when the loop occupies the same virtual space as the prostate, tissue is eliminated, creating a furrow [20]. A series of red points as the loop moves away from the cut surface simulates bleeding.

This TURP simulator enables trainees to become familiar with the resectoscope and the limits of TURP resection, with no time constraints and with a high degree of realism; it allows trainees to acquire some of the basic resection skills without the costs of training in the operating theatre, and presents no risk to patients. However, the simulator is limited by the lack of haptic feedback and delayed images. Once it is fully developed, the simulator will become an important part of training and assessing the techniques of trainees and experienced resectionists.

To date, interactive simulators have used red dots on the computer screen to simulate bleeding [20,21]. More recently, another group have enhanced the realism presented by the simulator of Ballaro et al.[21] and highlighted the need to learn the art of haemostasis in a procedure that is carried out in a fluid environment [22]. This group have captured video footage of bleeding and have texture-mapped this blood-flow sequence onto a virtual surface. During the running of the simulation, when the prostatic tissue is resected, bleeding is systematically triggered. This model allows the trainee to practice with added reality and with the appropriate responses to the challenges of haemostasis under a variety of different flow conditions.

Flexible ureteroscopy simulator

Flexible ureteroscopy has limited tactile feedback to the urologist and therefore is a relatively straightforward procedure to represent when using VR techniques. The first flexible ureteroscopy simulator was recently developed (Karl Storz Endoscopy America, Culver City, CA) with intrarenal pathologies such as a kidney stone, a small tumour and a normal air bubble positioned at various locations within the intrarenal collecting system [23]. The locations of these lesions can be altered, thereby changing the simulation experience that the individual urology trainee will encounter. The software allows for measurements such as the time taken to navigate up the ureter and through the inside of the kidney, and the accuracy of identifying the locations of the various lesions.

The simulator was introduced at the annual meeting of the AUA (May 1995, Las Vegas, NV) and was evaluated by over 300 urologists. Over 98% of the endoscopic surgeons thought that the simulation was representative of true flexible ureteroscopy. The model is being validated at present but has the foundations to allow trainees to practise a procedure requiring complex hand-eye co-ordination before operating on real patients.

Laparoscopic trainer

The minimally invasive surgical trainer (MIST) is a laparoscopic-skills trainer that simulates a range of skills required, such as grasping and diathermy. This computer-based model is linked to a frame and has two laparoscopic instruments attached to it. During the tasks, the number of errors, time taken and economy of path length are objectively recorded, and at the end of the task provide differential right- and left-handed analyses. Movements of these instruments are shown on the monitor. The MIST has been validated as a training tool that can distinguish grades of surgeons [24]. However, the problem with MIST is that there is no force-feedback from the instruments when the instrument touches objects. This is the basis of the next generation of VR simulators and ReachIn (Sweden) is an example of a company that has developed a 3D system with haptic feedback, and is in the process of modifying this programme for surgical applications.

Other potential uses of VR

Although this review has covered the role of VR in training, one group has developed a use of VR in treating patients. A VR system has been developed for treating erectile dysfunction and enables the patient to follow pathways that accelerate a psychodynamic process, which directly simulates the subconscious [25]. The experience brings to light the obstacles that led to their sexual dysfunction. In that study, the authors reported that after therapy sessions including VR, the total (defined as the return of adequate erection with completion of sexual activity) and partial positive response rate was 73%. The patients who had benefited from the VR therapy were contacted 6 months after the last session and most reported no adverse physical reactions, and expressed a desire to prolong the VR experience. This highlights the potential future roles of VR in treating and educating patients.

Conclusions

Although VR in urology is in its infancy, it is anticipated that as improvements in hardware and software continue, VR applications will become more lifelike and enhance surgical training. There is a growing body of literature now supporting training in a structured environment outside the operating theatre, apart from the ethical and safety issues of training on patients in an age of increasing litigation.

Current applications of VR in urology have confirmed feasibility in teaching, training of urologists, planning of surgical procedures and in treating specific urological conditions. It is possible that urological simulations may be incorporated into the revalidation process [26].

Authors

J. Shah, BSc, MRCS, Clinical Research Fellow.

S. Mackay, Clinical Research Fellow.

J. Vale, MS, FRCS, Consultant Urologist.

A. Darzi, MD, FRCS, FRCSI, FACS, Professor of Surgery.

Ancillary