Implementando o prontuário eletrônico OpenEHR em CMSs: uma aproximação

• Ano de defesa: 2014
• Autor(a) principal: Christiano Pereira Pessanha
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais
• Palavras-chave em Português: Informática na medicina; Ontologias (Recuperação da informação) eses; Ciência da informação; Representação do conhecimento (Sistemas especialistas)
• Link de acesso: https://hdl.handle.net/1843/BUOS-9V4PBM

This research has as motivation the challenges arising from efforts that seek to achieve semantic interoperability in Electronic Health Records (EHR). More specifically, it focuses on the use of OpenEHR standard, which was especially developed to facilitate this type of interoperability. It is argued that archetypes (i.e., artifacts of clinical knowledge of the OpenEHR standard) can be implemented in a Content Management System (CMS) framework, with the advantage of promoting the separation between the domain knowledge (in health) and the EHR code, thus facilitating the semantic interoperability. Other advantages are discussed in the text. The survey follows the Design Science methodology, which indicates the performance of practical experiments as well as the generation of theoretical knowledge from such experiments. Therefore, by way of practical exploration, the research attempts to express OpenEHR archetypes in the technological framework of a specific CMS: the Plone CMS. At first attempt, the expression of OpenEHR archetypes is sought by means of a Plone/Archetypes tool (tool for the automatic generation of content). It has been ascertained that the cost-benefit is unsatisfactory. There are also doubts regarding the sufficiency of the expressive potential of the base language of the CMS used (Python language), due to the complexity of the clinical concepts expressed in the OpenEHR archetypes. Replying to such questions is fundamental to the attempts of expressing CMS Plone archetypes, thus the research explores, in a second attempt, the deployment of OpenEHR information models in Python language (basis of CMS Plone), since these models provide all the required semantics for the construction of OpenEHR archetypes. As a contribution, the thesis performs the previously mentioned deployment and opens the prospect of expression of the clinical knowledge artifacts (OpenEHR Archetypes) in the CMS Plone framework. Besides facilitating the interoperability of the EHR implemented in CMS, benefits arising from its use are also expected. In fact, the use of a CMS carries a positive cost-benefit when compared to the deployment of systems made "from scratch". Some advantages are: automatic generation of content; the management of content in an automated way (via CRUD2); the facilitated creation of interfaces via templates, in addition to others. The investigation into the possibility of expressing OpenEHR semantics in Python suggests the viability of deployment of this standard in the various frameworks of the language. Furthermore, the thesis evidences the alignment between the archetype concept, key to the OpenEHR standard, and the content concept, key to the CMSs. The result of this theoretical effort suggests that, despite the initial difficulty with the Plone/Archetypes tool, there is equivalence between these two concepts. This opens possibilities of expression of archetypes in CMSs in general (not just in Plone).