neuroinformatika Dotaz Zobrazit nápovědu
Příspěvek obsahuje krátkou informaci o úloze informatiky v neurovědách a existujícím projektu jejího uplatňování v této oblasti.
- Klíčová slova
- projekt Lidský mozek, simulace mozku, neuroinformatika,
- MeSH
- informatika * trendy MeSH
- lidé MeSH
- mapování mozku * trendy MeSH
- mozek * fyziologie MeSH
- neurologie trendy MeSH
- počítače * trendy MeSH
- Check Tag
- lidé MeSH
Využití metod průzkumové analýzy dat (exploratory data analysis, EDA) je při hodnocení klinických dat v medicíně klíčovou fází. Vizualizační principy, modely poukazující na trendy vývoje či např. znázornění potenciálních závislostí, pomáhají k lepší interpretaci měření a v diagnostickém rozhodování. Počet dostupných moderních EDA balíků pro vývojáře v posledních letech roste v souvislosti s rozvojem oboru Data Science. NeuroEDA je interaktivní webová aplikace pro hodnocení biomedicínských dat. Aplikace byla naprogramována ve statistickém jazyce R, v rámci reaktivního paradigmatu frameworku Shiny. Je dále rozvíjena a využívána Katedrou biomedicínské informatiky FBMI ČVUT ve spolupráci s Neurologickou klinikou 1. LF UK a VFN v Praze, především pro hodnocení pacientů s dystoniemi a Parkinsonovou nemocí. Zpracování uživatelských dat v tabulkové formě (.csv, excel) probíhá v serverové části. Kromě základních popisných statistik, průzkumových grafů a shlukové analýzy, které jsou vhodné i pro hodnocení velkých dat, nabízí aplikace metody pro robustní a neparametrickou analýzu. Ty jsou v neurologii obzvlášť vhodné. Typicky z důvodu malých počtů a vlivných pozorování. Dále kvůli častému nesplnění dalších statistických předpokladů. Mezi její výhody patří snadná rozšiřitelnost o nové R balíky a rychlá odezva ve webových prohlížečích. Uživatelské interaktivní prostředí umožňuje práci s funkcemi jazyka R bez znalosti skriptování.
Usage of methods for exploratory data analysis (EDA) plays an important role in assessment of clinical medical data. Visualizations, models and illustration of dependency can help for better understanding of measurements in the diagnostics and making decision. The number of available modern EDA packages for developers is increasing as well as the development of the Data Science field. NeuroEDA is an interactive web application for biomedical data assessment. The application was programmed in the R statistical language, based on reactive paradigm framework called Shiny. It is further developed and used by the Department of Biomedical Informatics FBME CTU in cooperation with the Neurological Clinic of the 1st Medical Faculty, Charles University in Prague. Especially for the patients with dystonia and Parkinson's disease. Computations and processing of user data are executed on the server side. Basic descriptive statistics, interactive graphs, clustering, robust and LOESS regression were implemented and are suitable for analytics of big data. Robust regression is especially suitable in neurology. Typically, due to the small numbers of measured observations and failure to prove other statistical assumptions. Among its advantages we can consider easy expandability of new R packages and quick response in web browsers. User interface allows to work with the R language features without any scripting knowledge.
- Klíčová slova
- neuroinformatika, neuroEDA,
- MeSH
- internet MeSH
- interpretace statistických dat MeSH
- lékařská informatika * MeSH
- navrhování softwaru MeSH
- neurologie * MeSH
- software MeSH
- statistika jako téma MeSH
- Publikační typ
- práce podpořená grantem MeSH
Background: INeuroinformatics is a rapidly developing interdisciplinary field which provides an enormous amount of data to be classified, evaluated and interpreted. Usage of exploratory data analysis (EDA) methods is essential in evaluating clinical data in medicine and this analysis remains a big challenge because each new system has specific requirements. Visualizations, models and illustrations of dependency can help in better understanding of measurements in diagnostics and in decision making. The number of available modern EDA packages for developers is increasing as well as the development in the Data Science field. The development of modern methods of data analysis must also be incorporated in university education. Objective: The aim of the study is to design and develop software, which implements current EDA packages and model making procedures for neurological data analyses which could be easily modified. The second objective is to evaluate the possibility of supporting the education of biomedical engineering students at the undergraduate level in order to provide effective support in biomedical data analysis. Methods: An application has been created under the reactive Shiny framework in the R language. Data in .csv or .tsv format are processed on the server side of the application. Results: We have developed a new easy-to-use software named NeuroEDA for interactive web-based biomedical data assessment. This application covers basic descriptive statistics, exploratory graphs and cluster analysis, which is also suitable for big data examination. Furthermore, this application offers methods for robust and non-parametric analysis. These are particularly useful in neuroinformatics from our long-term experience. The application was practically deployed in the evaluation of clinical neurological data and in teaching the subject Biomedical Statistics. Conclusion: We have introduced the possibility of creating biomedical software for clinical use and demonstration in teaching. Among the advantages of the application, is that it is easily expandability with new R packages and quick processing in web browsers. The interactive user interface allows one to work with R’s functions without needing scripting/programming knowledge. Students can acquire practical experience in processing and transformation of heterogeneous medical data not only in biomedical engineering fields, but also at the medical faculties for Medical Informatics. This application is actively used for neuroinformatics data assessment and in discovering some potentially useable hypotheses.
- Klíčová slova
- neuroEDA, neuroinformatika,
- MeSH
- biomedicínské inženýrství MeSH
- interpretace statistických dat MeSH
- lékařská informatika * MeSH
- navrhování softwaru MeSH
- neurologie MeSH
- software * MeSH
- statistika jako téma * MeSH
- vyučování MeSH
- vzdělávání odborné MeSH
- Publikační typ
- práce podpořená grantem MeSH
En Diciembre de 2001 la Comisión Europea organizó una conferencia a la que asistieron más de 400 expertos y en la que se analizaron las sinergias existentes entre la Bioinformática (BI), la Informática MÉdica (MI) y la Neuroinformática. En Noviembre de 2002, en el congreso internacional BIOINFORSALUD (Valencia, España) se presentó el White Paper del proyecto BIOINFOMED, cuyo objetivo era el de contribuir a las estrategias de Investigación y Desarrollo de la Comunidad Europea en las mencionadas áreas informáticas en la próxima década. En este documento estratégico encargado por la Comisión se analizaban las relaciones establecidas entre la BI y la MI, resultando en una nueva definición de la disciplina de Informática Biomédica (BMI), cuya finalidad es facilitar el desarrollo de la Medicina Genómica. Este White Paper se realizó gracias a la colaboración de un comité de 30 expertos internacionales coordinados por el Instituto de Salud Carlos III (ISCIII) que diseñaron una agenda con 18 líneas de investigación para corroborar la sinergia existente y en el que además se señalaron las áreas en las que se debería priorizar el esfuerzo. El propósito de este artículo es actualizar este estudio detallando algunos de los avances más recientes surgidos en este campo. Durante los tres años transcurridos entre la publicación del White Paper hasta la fecha actual han surgido numerosas iniciativas que han sido cuidadosamente revisadas seleccionando aquellos proyectos en función de su mayor relevancia. Algunos de los proyectos analizados en este documento son: la infraestructura informática para biobancos desarrollada bajo la supervisión del Consorcio “Public Population Project in Genomics” (P3G), la Red de Excelencia (INFOBIOMED) en Informática Biomédica constituida bajo el Sexto Programa Marco de la Unión Europea para la Investigación y Desarrollo Tecnológico, la iniciativa encabezada por “Health Level Seven” (HL7) para incluir la información genética en la historia clínica electrónica, la propuesta de un Proyecto Fenoma Humano, la Red Temática española de Investigación Cooperativa in Informática Biomédica (INBIOMED), los nuevos Centros Nacionales de Computación Biomédica (NCBCs), financiados por los NIH, bajo la iniciativa BISTI y los proyectos relacionados con la simulación y modelado de la Fisiología Humana.
In December 2001, the European Commission promoted a Conference in which more than 400 experts analyzed the synergies arisen between Bioinformatics (BI), Medical Informatics (MI) and Neuroinformatics. In November 2002, and in order to contribute to the strategy of the European R&D policy for the following decade in such areas, the White Paper of the BIOINFOMED project was presented at the international congress BIOINFORSALUD 2002 (Valencia, Spain). In the strategic document entrusted by the Commission, the relations established between BI and MI were analyzed, resulting in a new definition of the discipline of Biomedical Informatics (BMI) that aims to facilitate the developments of Genomic Medicine. To elaborate the White Paper, a committee of 30 international experts coordinated by the Institute of Health "Carlos III" (ISCIII) designed the agenda with 18 lines of research that corroborated the existing synergy. They pointed out those areas in which the efforts had to be prioritized. In this article, we update this study by highlighting recent outstanding developments in this field. The projects presented respond to a careful selection carried out among the numerous initiatives that have arisen in the three years passed between the publication of the White Paper and this article. Some of the projects analyzed in this paper are: the IT infrastructure for biobanks under the Public Population Project in Genomics (P3G) Consortium, a Network of Excellence (INFOBIOMED) in Biomedical Informatics constituted within the European Union Sixth Framework Program for Research and Technological Development, the initiative headed by HL7 to include genetic information in the electronic health record, the proposal of a Human Phenome Project, a Spanish Cooperative Research Thematic Network (INBIOMED) in Biomedical Informatics, the new National Centers for Biomedical Computing (NCBCs), funded by NIH, under the BISTI initiative and the projects related with the simulation and modeling of Human Physiology.
