Publication: Análisis del uso de la inteligencia colaborativa como herramienta para la construcción de bases de conocimiento consensuadas en procesos de diagnóstico médico
Loading...
Identifiers
Publication date
2013
Defense date
2013-09-10
Authors
Tutors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
En la presente tesis se presenta un estudio que gira entorno principalmente a la
Inteligencia Colaborativa como principal característica de las aplicaciones de tipo Social
Media y Web 2.0. Estos conceptos han sido combinados con áreas de estudio como la
medicina, las tecnologías de Web semántica y los sistemas de soporte a la decisión
médica (CDSS), con el objetivo de conocer la forma en cómo la Inteligencia
Colaborativa logra afectar de manera positiva a la obtención de diagnósticos.
Las redes sociales en esta investigación, han sido identificadas como estructuras
sociales conformadas por un grupo de personas cuyo objetivo principal es la
participación en actividades comunes, la mayoría de las veces en busca de la solución a
problemas. Este fenómeno de participación, compartición de información y
colaboración ha sido tomado como base para la creación de redes sociales y demás
plataformas colaborativas en Internet, en donde lo que destaca nuevamente es la
arquitectura de participación de la que hacen uso. Un caso especial y que ha sido objeto
de estudio de esta investigación son aquellas plataformas colaborativas con contenido
médico.
La Web semántica ha jugado un papel fundamental en este estudio ya que permite la
comunicación entre diferentes sistemas para compartir información, es decir, la
interoperabilidad entre sistemas. También facilita la representación del conocimiento en
diferentes áreas y finalmente también permite realizar procesos de inferencia cuando se
aplica a sistemas expertos.
Con base en los conceptos anteriores y respaldada en el concepto de Wisdom of the
crowd (la sabiduría de las multitudes), esta investigación plantea la definición de tres
métodos de consenso que han sido aplicados a bases de conocimiento con contenido
médico. Para la evaluación de los resultados se han utilizado las métricas comunes a los
CDSS siguiendo los criterios propuestos por Kaplan en las diferentes bases de
conocimiento consensuadas, las cuales se han comparado con los valores en las mismas
métricas generadas por un CDSS tradicional que ha sido tomado como estándar de oro.
Finalmente, esta tesis presenta las mejoras que la Inteligencia Colaborativa aporta a la
medicina en términos de exactitud de los diagnósticos y las ventajas que esta representa
cuando se aplica a estos sistemas. -----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
In this thesis a study that primarily revolves around the main characteristic Collaborative Intelligence of Social Media type applications and Web 2.0 is presented. These concepts have been combined with areas of study such as medicine, Semantic Web technologies and Clinical Decision Support Systems (CDSS) with the aim to know the way how the Collaborative Intelligence does positively affect the development of diagnostics. In this research, social networks have been identified as social structures formed by a group of people whose main objective is the participation in common activities, most of the time looking for the solution of problems. This participation, collaboration an information sharing phenomenon has been taken as the basis for social networking and other online collaborative platforms, where it is again highlighting the participation architecture that those systems use. A special case has been studied in this research are those with medical content collaborative platforms. The Semantic Web has played a key role in this study because it allows communication between different systems to share information, ie interoperability between systems. It also facilitates knowledge representation in different areas and finally also allows inference processes when it is applied to expert systems. Based on the above concepts and supported by the concept of Wisdom of the crowd, this research presents the definition of three consensus methods that has been applied to knowledge bases with medical content. For evaluating the results common metrics to CDSS were used following the criteria proposed by Kaplan in the different consensus knowledge bases. The results have been compared with the same metrics values generated by a traditional CDDS which is taken as the gold standard. Finally, this thesis presents the improvements that the Collaborative Intelligence brings to medicine in terms of diagnostic accuracy and the advantages that this represents when applied to this kind of systems.
In this thesis a study that primarily revolves around the main characteristic Collaborative Intelligence of Social Media type applications and Web 2.0 is presented. These concepts have been combined with areas of study such as medicine, Semantic Web technologies and Clinical Decision Support Systems (CDSS) with the aim to know the way how the Collaborative Intelligence does positively affect the development of diagnostics. In this research, social networks have been identified as social structures formed by a group of people whose main objective is the participation in common activities, most of the time looking for the solution of problems. This participation, collaboration an information sharing phenomenon has been taken as the basis for social networking and other online collaborative platforms, where it is again highlighting the participation architecture that those systems use. A special case has been studied in this research are those with medical content collaborative platforms. The Semantic Web has played a key role in this study because it allows communication between different systems to share information, ie interoperability between systems. It also facilitates knowledge representation in different areas and finally also allows inference processes when it is applied to expert systems. Based on the above concepts and supported by the concept of Wisdom of the crowd, this research presents the definition of three consensus methods that has been applied to knowledge bases with medical content. For evaluating the results common metrics to CDSS were used following the criteria proposed by Kaplan in the different consensus knowledge bases. The results have been compared with the same metrics values generated by a traditional CDDS which is taken as the gold standard. Finally, this thesis presents the improvements that the Collaborative Intelligence brings to medicine in terms of diagnostic accuracy and the advantages that this represents when applied to this kind of systems.
Description
Keywords
Inteligencia colaborativa, Redes de colaboración, Redes sociales, Web semántica, Medicina, Diagnóstico médico