ARK-Virus: An ARK Platform Extension for Mindful Risk Governance of Personal Protective Equipment Use in Healthcare

In this demonstration we present the Access Risk Knowledge (ARK) Platform - a socio-technical risk governance system. Through the ARK Virus Project, the ARK Platform has been extended for risk management of personal protective equipment (PPE) in healthcare settings during the COVID-19 pandemic. ARK demonstrates the benefits of a Semantic Web approach for supporting both the integration and classification of qualitative and quantitative PPE risk data, across multiple healthcare organisations, in order to generate a unique unified evidence base of risk. This evidence base could be used to inform decision making processes regarding PPE use.


INTRODUCTION
This paper describes the Access Risk Knowledge (ARK) platform for socio-technical risk governance and structured evidence collection [3]. ARK's risk governance system is based on the Cube -an established methodology used for analysing socio-technical systems and for managing associated risks [2,6]. The Cube provides a framework This  for initiating and analysing organisational change through interlinked, dimension-specific questionnaires. These questionnaires are used to identify, assess, and classify risks, as well as to plan, execute, and evaluate risk mitigation actions.
The ARK-Virus Project, see Section 2, extends the ARK Platform, see Section 3, for use in the healthcare domain -specifically for risk management of personal protective equipment (PPE) use for the purpose of COVID-19 infection prevention and control (IPC).

ARK-VIRUS PROJECT
PPE is a critical component of IPC [5,8] as such, up to date, situationaware risk management is essential in ensuring that key mechanisms of PPE practice are understood, appropriate actions are carried out and outcomes are monitored [9].
The risk management infrastructure of the Health Service Executive 1 (HSE) in Ireland is currently limited to traditional risk registers, focused on manual reporting, that are siloed in individual organisations. This means that while local innovation around PPE use is possible, this data is rarely shared with other organisations.
The ARK-Virus Project aims to provide a collaborative space for PPE risk management. The ARK Platform uses Semantic Web technologies to model, integrate, and classify PPE risk data, from both qualitative and quantitative sources, into a unified knowledge graph of PPE for COVID-19 IPC. This graph will facilitate the interlinking of risk data with other available datasets on PPE practices, risks and evidence. The ARK Platform will also support the CUBE mindful governance methodology for the analysis, identification and management of PPE risks by healthcare safety experts.
The ARK-Virus project is a collaboration between an academic team and a community of practice in COVID-19 PPE deployment including St James's Hospital 2 , Dublin Fire Brigade 3 , Tulsa Child and Family Agency 4 , Beacon Renal 5 dialysis services and Stewarts Care 6 services for people with intellectual disabilities. By putting ARK in place over many PPE deployment projects, it will be possible to conduct semi-automated multi-project analysis and distillation of best practice data into a shareable, privacy-aware, Linked Data knowledge base. Development of this integrated evidence base is critical in optimising the effectiveness of PPE and in understanding the factors influencing compliance with evidence-based IPC measures.

ARK PLATFORM
The ARK Platform leverages Semantic Web technologies and W3C standards to integrate qualitative clinical risk management data with quantitative operational data and analytics. At the technical centre of the ARK Platform is a unified model of risk, expressed using an OWL knowledge graph and SKOS taxonomies, that bridges qualitative and quantitative data. This makes large-scale evidence collection and risk analysis more tractable by transforming humanoriented quantitative risk information into structured data using the ARK Cube Ontology 7 and the ARK Platform Vocabulary 8 , and by enabling text fields to be annotated with concepts from the ARK Risk 9 and the ARK Health 10 taxonomies. Domain experts can interact with the platform via a web application.

CUBE Ontology & Platform Vocabulary
The ARK Cube ontology and the ARK Platform vocabulary enable data capture and analysis through the Cube methodology. The ontologies were developed by mixed teams of organisational psychologists and knowledge engineers. The ontologies were implemented using the Web Ontology Language 11 (OWL) specification in Protégé 12 . The Cube ontology is used in the overall architecture of the ARK Platform by serving as a repository of the Cube questionnaire and by modelling questionnaire responses. The Platform vocabulary allows for the modelling of platform users, access controls and user permissions. It also provides a data classification system, based on the HSE Information Classification and Handling Policy [1], that allows data to be categorised according to the sensitivity of its contents. The ontologies contain machine and human-readable metadata, are free of inconsistencies, and are extensible and customisable for different CUBE questionnaires. They have also been validated using the Ontology Pitfall Scanner 13 .

ARK Application
The ARK Application 16 (see Figure 1) is built on open source Jena 17 knowledge base technology and uses a Node.js 18 web front end to enable users to interact with the system. The platform consists of four main components -a Risk Register, Project Analysis, CUBE Analysis and a CUBE Summary. The Risk Register (see Figure 2) allows users to create a risk using the HSE Risk Assessment format. The Project Analysis (see Figure 3) component allows users to create and analyse projects, and sub-projects, based on identified risks. These risks can be newly created or imported from a HSE Risk Assessment. The CUBE Analysis (see Figure 4) component guides users through completing the CUBE questionnaire. Both the second and third components allow users to interlink answers with supporting evidence and to classify answers with risk related concepts using the ARK Terminologies. Finally, the CUBE Summary (see Figure 5) component provides a summary and visualisation of the Project Analysis and the CUBE Analysis.

ARK PLATFORM DEMONSTRATION
A highly interactive demonstration of the ARK-Platform will be provided. The audience will be able to view the CUBE ontology, Project vocabulary, safety management taxonomy and health risk taxonomy via WIDOCO 19 HTML documentations. The audience will also be able to use the ARK application and explore the platform's functionalities using sample PPE risk management data. 19 https://github.com/dgarijo/Widoco