IERAPSI surgical simulator design submitted to severe task analysis principles

Sale 31 May 2000The IERAPSI consortium has received funding from the European Commission within the Fifth Framework Programme to build an integrated environment for the rehearsal and planning of surgical interventions. In this respect, the IERAPSI team has to deliver a task analysis report to define the human-centred approach of surgical procedures, relating to mastoidectomy, cochlear implantation, and acoustic neuroma resection in particular. With a detailed report at hand, the project partners are able to accurately determine which features the virtual IERAPSI environment has to incorporate in order to meet the required user needs. Based on the guidelines laid out in the ISO 13407 standard for human-centred design in interactive systems, an interim report has been prepared by Professor Robert J. Stone from Virtual Presence Limited, which is one of the project participants.


The ISO 13407 standard which recently came into force, specifies four general principles of human-centred design. These imply an active user involvement with a clear user and task requirements understanding. Second, functions need to be allocated between users and technology. Task analysis forms an iterative process which necessitates the revision of design solutions at multiple stages, whereas in fourth place, the design has to be the result of a multi-disciplinary input. The standard equally highlights four principles for human-centred design activities. Project participants ought to understand as well as specify the context of use. Also a definition of user and organisational requirements is indispensable for a successful outcome. In third place, the multi-disciplinary team has to produce design solutions with an active user involvement. The final design then has to be evaluated against the different requirements.

Colleagues from the Manchester Royal Infirmary's ENT unit, Department of Otolaryngology, Head and Neck Surgery, and the Department of Surgery and North of England Wolfson Centre for Minimally Invasive Therapy assisted Dr. Stone in performing the IERAPSI task analysis. As such, next to the off-line investigation of Internet documents and the method of informal interviewing, Dr. Stone acquired a lot of personal experience in this specific surgical area, using the existing training techniques, such as the cadaveric temporal bone exercises, extensive training sessions with the Temporal Bone Dissector CD-Rom, and dissections by means of a set of temporal bone models.

In addition, Dr. Stone was offered the unique chance to witness and record four live surgical interventions, including an infantile cochlear implant, a case of middle fossa acoustic neuroma, and two translabyrinthine acoustic neuromas. A wealth of useful findings allowed the author of the report to propose two initial human interface characteristics to be integrated into the surgical planning environment of the IERAPSI simulator. The first is an off-line pre-operative system, equipped with a multi-user real time 3D display and appropriate interactive six degrees-of-freedom controls. The second one ideally features an on-line intra-operative, scaled-down version of the multi-user system, ergonomically located to offer maximum benefit to the surgeon with possible interaction via speech recognition.

To his great surprise, Dr. Stone noticed that both registrar and consultant surgeons made little or no use of the radiological patient images during the operations. The records had been consulted on beforehand to part-plan the intervention mentally. Apparently, surgeons seem to rely on their experience to locate key anatomical structures in a timely and safe fashion. At present, IERAPSI focuses heavily on the development of software modules to support the radiological data processing. To this purpose, consultation with users on this particular aspect has to be refined as the concepts emerge.

As for the IERAPSI training environment, the trainee should be provided with adequate tools to practise special procedures and operative skills, relating to the initial bone exposure, drilling and burring actions, and the use of virtual instruments and materials. Because surface marking and incision tasks are basic skills for a surgeon, there is no need to include expensive high-fidelity tissue simulation within IERAPSI, according to Dr. Stone. The author makes a plea for a head or skull section representation in 3D, using a conventional, non-stereoscopic, high-resolution display, completed with a 6/3 degrees-of-freedom data input/haptic feedback device. On the other hand, the system should be capable of indicating gross errors between the trainee's marks and those of an optimised area.

The typical nature of the various drilling and burring actions depends on the type and depth of bone. Sensable Technologies' PHANToM is the only viable and commercially available system to adequately replicate these qualities in a training situation, as Dr. Stone points out. Experienced users can provide useful advice with regard to actually exerted forces for the construction of a haptic sensation library, applied to key temporal bone tasks. Not only tactile sense, but also sound and vision constitute major features in a virtual reality training scenario for surgical drilling and burring. Dr. Stone emphasises the importance of simulating drill site obscuration by bone dust paste to attract the trainee's attention to the need for regular irrigation and suction. Bleeding effects have to be very realistic to indicate damage events, as well as collision detection markers. It is not necessary though, to represent the drilling act as such, nor the actual spinning of the cutter or diamond burr.

Auditory feedback should be kept low profile and restricted to distinctions in frequency, observed in different contact time with the material. In addition, the system also has to provide audible cues, associated with the use of non-drilling instruments, such as facial stimulator and diathermy devices. Next to this, haptic and visual representations are requested for various surgical tools like probes, hooking and peeling tools, clotting mesh, etc. Whether the IERAPSI training system should be designed to evaluate human performance still remains an open question. In order to assess surgical competence, the system has to be ecologically valid which requires detailed consultation with future users.

Finally, although not a requirement of the IERAPSI project, Dr. Stone equally suggests the possibility of decomposing the surgeon's behaviours to a basic level. Here, the design of a simple part-task skills trainer based on abstract task elements, could be useful to train "deep" drill positioning, sensitive structure visual avoidance, thin bone "hooking", bone structure contour following, simple linear tracking, and deep drilling under conditions of partial visual obscuration. A complete and detailed description of the IERAPSI project is available in the VMW March 2000 issue under the title Interactive patient-customised simulator offers surgeons real life "plan, train and rehearse" experience.

Leslie Versweyveld

[Medical IT News][Calendar][Virtual Medical Worlds Community][News on Advanced IT]