IADR Abstract Archives
Studying dental dam outcomes using graph database principles
Objectives: Dental dam is a tool, part of a dental armamentarium used to achieve the goals of: safety, moisture control, and efficient isolation of the surgical site. The objective of this study is to map clinical patterns and outcomes for use of the dental dam in implementing new technology, graph databases, and mapping clinical practice guidelines for decision-making.
Methods: The search bibliome included citations found using PubMed, Google, EBSCOhost, and Science Direct. A total of 50 papers including 2 case studies were obtained, reviewed, and analyzed. Patterns and outcomes were mapped using Neo4j graph database principles. A traversal graph of patterns and outcomes achieved Subject Matter Expert level knowledge of node to node relationships with edge-probabilities to each reported outcome terminus. Data points were averaged due to the limited amount of data available using Shapiro-Wilks at p=.05 significance level.
Results: The map began with Patient or conceptual patient node, due to compilation of data points rather than single data points as would occur in an actual patient traversing the graph database. The primary node to node relationships included Decision-maker, Diagnosis, Infection, Caries, Treatment with Attribute node, Basic Armamentarium. Graph then broke into other major and their children nodes to pattern outcome termini. Each was mapped with edge probabilities when data from the literature was evidenced. Mapped outcomes included each of alternative armamentariums to the dental dam currently used in practice. In all, 61 nodes and sub-nodes were mapped and edge probabilities inputted when evidenced.
Conclusions: Mapping of the clinical patterns and outcomes produced a graph model and Subject Matter Expert knowledge database of the dental dam. With this model and further research using systematic review and existing database resources, we will be able to increase our predictive knowledge of the outcomes of the use of the dental dam for surgical dentistry and validate the model.
Methods: The search bibliome included citations found using PubMed, Google, EBSCOhost, and Science Direct. A total of 50 papers including 2 case studies were obtained, reviewed, and analyzed. Patterns and outcomes were mapped using Neo4j graph database principles. A traversal graph of patterns and outcomes achieved Subject Matter Expert level knowledge of node to node relationships with edge-probabilities to each reported outcome terminus. Data points were averaged due to the limited amount of data available using Shapiro-Wilks at p=.05 significance level.
Results: The map began with Patient or conceptual patient node, due to compilation of data points rather than single data points as would occur in an actual patient traversing the graph database. The primary node to node relationships included Decision-maker, Diagnosis, Infection, Caries, Treatment with Attribute node, Basic Armamentarium. Graph then broke into other major and their children nodes to pattern outcome termini. Each was mapped with edge probabilities when data from the literature was evidenced. Mapped outcomes included each of alternative armamentariums to the dental dam currently used in practice. In all, 61 nodes and sub-nodes were mapped and edge probabilities inputted when evidenced.
Conclusions: Mapping of the clinical patterns and outcomes produced a graph model and Subject Matter Expert knowledge database of the dental dam. With this model and further research using systematic review and existing database resources, we will be able to increase our predictive knowledge of the outcomes of the use of the dental dam for surgical dentistry and validate the model.