Table 7 The impact of implemented ontologies for the management of data quality

Management ( 9 papers )

(Li and Ko 2007)

To develop automated ontology approach to manage nutrients in a diabetes diet care knowledge management

-Used expert opinions to decide which are the important nutrients to include in the diabetes diet and therefore the ontology

Diabetes diet care in Taiwan

-This is face validity and consistency of the data

-Authors suggested that there is a further step using ontology approach for more efficient diet knowledge management

(Esposito 2008a)

To detect abnormalities and malformations due to heart diseases

-Use as an ontology approach and rules to perform the instance and consistency checking and verifies that patient information violates the normal cardiovascular model loaded based on the SNOMED vocabulary

Congenital Heart Disease (CHD) dataset in Italy

-Theoretical discussion on data consistency

-Researchers show applicability of ontology to define either the anatomy of the cardiovascular system in normal patients or the anatomy characterized by malformations or abnormalities in CHD patients to support cardiologist in the identification of diseases

(Nimmagadda et al. 2008)

To provide a solution to problems around handling increasing amounts of clinical information and solves some issues related to managing large

-Simulate human body disorders into metadata through ontology based data warehouse modelling

Human body anatomy and pathology dataset in Australia

-Theoretical discussion on managing accuracy and correctness of data

-Authors states ontology can facilitate logic processes and semantics for data quality management and decision support for health care providers and clinicians

(Min et al. 2009)

To collect/retrieve information intelligently and address the semantic heterogeneity problem from the integration of data from multiple information resources

-Apply ontology mapped with medical thesaurus to integrate and retrieve the data from two independent database systems

3000 records registered for the prostate cancer patients and Tumour Registry in US

-Theoretical discussion about data consistency

-Authors state that ontology can solve the semantic heterogeneity problem from the integration of two databases by recognition of inconsistence data

(Brüggemann and Grüning 2009)

To improve the outcome of data quality management (DQM)

-Use an algorithm and data model for consistency checking, an algorithm for detecting duplicates and give three examples of DQM-specific metadata tasks (data provenance, data quality annotations at schema and instance level and an ontology for the DQM domain)

Cancer registries in Germany

-Authors mentioned the usefulness of their ontology approach to define a shared vocabulary for improved interoperability, and performing DQM include consistency checking, data duplicate detecting and metadata management

(Topalis et al. 2011)

To retrieve data and information extraction

-Use ontology based model to integrate and capture the right terms (variables) and the relationships between such concepts in a disease map

Neurological disease, malaria, vector-borne diseases in Greece

-Theoretical discussion about data accuracy in multiple information sources

-Authors demonstrate the importance of capturing the right terms in ontologies to use both in the development of specific databases and, in the construction of decision support systems to control diseases for biologists, and epidemiologists

(Perez-Rey et al. 2006)

To develop a method and tool for database integration from remote sources

-Test the implemented ontology on eight different private databases with biomedical data stored in different database management packages such as MySQL, PointBase, Access, and others and provide integrated access to their data

Public genomic and clinical databases in Spain

-Use case study to retrieve information in three sources using queries and theoretical discussion on data consistency

-Authors believe that ontologies are the most suitable representation formalism for schemas in database integration system

(Lee et al. 2009)

To classify a person as a diabetic patient

-Represent new ontology methods for fuzzy medical relationship using taxonomical knowledge

Diabetes domain in Taiwan

-Manage accuracy of data

-Authors state that fuzzy ontology can effectively develop semantic decision making and reduce uncertainty (inaccurate data) to classify patients for medical staffs

(O’Donoghue et al. 2009)

To demonstrate the data quality benefits of integrating remote patient monitoring solutions

-Use a Body Area Network (BAN) datasets within patient EHR solutions

Three patient types are identified 1) Non-Athletic Adult, 2) Athletic Adult and 3) Child from Ireland

-Use Jade Content Ontology classes for their the Medical Knowledge Base agent

-Use 2 experiments (with/without knowledge base) for effect on risk prediction accuracy

-Focus on data accuracy and correctness

-Authors states that ontology can improve patient management through the reduction of false alarm generations and facilitate the categorisation of the data to indicate risk categories for decision support