Ontology Development Methodology for Virtual Reality-Based Training (VRT) in Ophthalmology Domain: Phaco Cataract Surgery Case Study

Introduction

In the ophthalmology field, cataract is considered as a well-known disease. Despite the easiness, safety and accessibility of good quality services, this disease continues to be the major reason for blindness and visual impairment globally (Chew et al., 2018). Reasons for that, include high costs of equipment maintenance, the availability and training of surgeons and so on (Singh & Strauss, 2015).

Virtual Reality Training (VRT) was obviously recommended as a way that has the possibility to provide an important levels of skills transfer to novice ophthalmic surgeons (Lucas et al., 2019). Moreover, it is appropriate, safe, has a low-pressure environment, and can provide repetitious training. However, although Virtual Reality (VR) is not a novel technology in the healthcare domain, the status in the progress of VR application development is still in its beginning stage. In the literature, the successful average of its adoption, is not well declared (Chang et al., 2019). Some reasons for that may include the development cost (Cook et al., 2019), lack of capturing training scenarios in an unambiguous manner (Radianti et al., 2020). Besides that, VR deals with a quite knowledge-intense domain. For instance, bridging the knowledge between domain experts (e.g., ophthalmologists and VR experts) is crucial and has to be captured and represented in an expressive manner. To simplify a VR development process, there is a requirement to have a modelling tool. In this context, the ontology was called to construct a systematic conceptualization model at a high level of expressiveness in order to provide a shared and common understanding of the real word.

However, thinking about clearance, and semantic expressiveness naturally leads the designers to think about the used methodology and modelling language. According to recent researches findings (Benferdia et al., 2020, 2022; Mohamad et al., 2021) show that majority of ontological designers do not present the methodologies that are utilized to design their ontologies, and they either misuse the existing methodology or use immature ones. Besides that, the existing methodologies have not been widely recommended to be a standard upon which to build an ontology (Corcho et al., 2003; Ahmad et al., 2018; Benferdia et al., 2020), due to the limitation of research that validates the best methodology for designing an ontology. Additionally, all utilize inadequate languages. Unfortunately, these types of languages are unsuccessful in helping to reach the target for representing a domain of interest in an expressive way. Moreover, there is a general lack of capturing perdurant knowledge “Know HOW”, e.g., all steps on how to perform cataract surgery, in all ontologies developed for VRT (Benferdia et al., 2020, 2022; Mohamad et al., 2021). This raises another matter linked to representing perdurant entities, which is one of the key features needed for ontology development. That being said, using the existing methodologies with current conditions will certainly generate a poor model with a lack of expressiveness, truthfulness, and details (Benferdia et al., 2021b). Nevertheless, there is a strong need to look for another solution. Building an ontology for VRT requires a highly established ontology development methodology under a well-researched methodology such as Design Science Research (DSR), as ontology is a type of design artifact.

In this study, two major contributions are made as follows. Firstly, establishing a unified methodology that contains the best practices from existing methodologies and is guaranteed to complete the necessary phases of an ontology lifecycle, would be good. Thus, the adopted and refined ontology development methodology was discussed in depth, analyzed, and presented, and is believed to provide systematic and comprehensive guidelines for building ontologies. Secondly, with a specific emphasis on VRT for cataract surgery, this research used the proposed methodology in order to create a common understanding on how to develop an accurate domain model for supporting VRT in the ophthalmology domain that integrates the OntoPhaco ontology, such that it is able to offer domain experts fundamental information and facilitate in designing VR training scenarios.