A hierarchical, ontology-driven Bayesian concept for ubiquitous medical environments- A case study for pulmonary diseases

Abstract

The present paper extends work on an existing computer-based Decision Support System (DSS) that aims to provide assistance to physicians as regards to pulmonary diseases. The extension deals with allowing for a hierarchical decomposition of the task, at different levels of domain granularity, using a novel approach, i.e. Hierarchical Bayesian Networks. The proposed framework uses data from various networking appliances such as mobile phones and wireless medical sensors to establish a ubiquitous environment for medical treatment of pulmonary diseases. Domain knowledge is encoded at the upper levels of the hierarchy, thus making the process of generalization easier to accomplish. The experimental results were carried out under the Pulmonary Department, University Regional Hospital Patras, Patras, Greece. They have supported our initial beliefs about the ability of Bayesian networks to provide an effective, yet semantically-oriented, means of prognosis and reasoning under conditions of uncertainty.

Development of a Methodological Approach for Data Quality Ontology in Diabetes Management

“…The challenges to the development and validation of ontology-based models to assess and manage DQ include methodological immaturity, immature knowledge base, lack of tools to support ontology-based design of information systems, evaluation of ontological approaches, and engagement of users in design and implementations (Rahimi, Liaw, Ray, & Taggart, 2012;Rahimi, et al, 2014). There have also been several attempts to define ontology evaluation metrics and provide practical techniques to evaluate ontological approaches (in terms of flexibility, scalability and reusability) against non-ontology based models (Cur & #233, 2012;Maragoudakis, Lymberopoulos, Fakotakis, Spiropoulos, & Ieee, 2008). There is a lack of valid and reliable DQ assurance (D. Arts, De Keizer, & Scheffer, 2002;D.…”

Ontological specification of quality of chronic disease data in EHRs to support decision analytics: a realist review

“…Defined purpose Assessed of fitness for purpose using DQ Context Assessment (7 papers) (Jacquelinet et al 2003) To develop semantic data interoperability -Apply an ontological tool to develop semantic data interoperability through domain terminologies using quantitative analysis of the existing coding information system and a qualitative analysis checking completeness, consistency, ambiguity and implicitness of terms Failure, dialysis and transplant datasets from National information system in France -Represent DQ factors such as completeness of data, appropriated terms, structured thesaurus, and terminology standard -Authors state usefulness of ontology based approach to support the processing of texts, and extending a terminological basis for medical experts (Maragoudakis et al 2008 Gene expression data which involve the use of microarrays in UK -Guide the development and use of metrics to measure the complexity and cohesion of ontologies -Authors state that ontology helps to allow a practical division of the work between providers and consumers, in order to minimize the costs to all concerned 2012b; Spasic and Ananiadou 2005;Stvilia et al 2009;Valencia-Garcia et al 2008;Wang et al 2007). Table 9 illustrates various definitions to identify the most common criteria to assess validity of ontologies and data models.…”

“…However, the paper was not explicit about whether was a formal outcome-based comparison of ontological and non-ontological approaches was conducted. Maragoudakis et al (2008) developed an ontology with 5 domains for a clinical Decision Support System (CDSS) for management of Chronic Obstruction Pulmonary Disease (COPD). The ontology, based on hierarchical Bayesian networks, encoded a domain (COPD) and compared the predictive accuracy of this ontology-based hierarchical Bayesian network method with linear programming and artificial neural network methods (Maragoudakis et al 2008).…”

“…Maragoudakis et al (2008) developed an ontology with 5 domains for a clinical Decision Support System (CDSS) for management of Chronic Obstruction Pulmonary Disease (COPD). The ontology, based on hierarchical Bayesian networks, encoded a domain (COPD) and compared the predictive accuracy of this ontology-based hierarchical Bayesian network method with linear programming and artificial neural network methods (Maragoudakis et al 2008). By using 10-fold cross validation and precision and recall metrics, they concluded that the Hierarchical Bayesian method is comparable to Artificial Neural Network (ANN) and far more accurate than linear programming approaches.…”