Data-Mining in klinischen Datensätzen: Bericht der Arbeitsgruppe Bioinformatik der DGKL
Data mining programs help with extracting knowledge from large amounts of data. Despite decades of experience with profile testing, laboratory medicine is just now realizing the practical application of these programs to highly parallel analytical techniques (genomics, proteomics, etc.). Professiona...
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Dokumenttyp: | Article (Journal) |
| Sprache: | Deutsch |
| Veröffentlicht: |
22. Juli 2010
|
| In: |
Laboratoriumsmedizin
Year: 2010, Jahrgang: 34, Heft: 4, Pages: 227-233 |
| ISSN: | 1439-0477 |
| DOI: | 10.1515/jlm.2010.041 |
| Online-Zugang: | Verlag, lizenzpflichtig, Volltext: https://doi.org/10.1515/jlm.2010.041 Verlag, lizenzpflichtig, Volltext: https://www.degruyterbrill.com/document/doi/10.1515/jlm.2010.041/html 
|
| Verfasserangaben: | Georg Hoffmann, Marc Zapatka, Peter Findeisen, Stefan Wörner, Peter Martus, Michael Neumaier |
| Zusammenfassung: | Data mining programs help with extracting knowledge from large amounts of data. Despite decades of experience with profile testing, laboratory medicine is just now realizing the practical application of these programs to highly parallel analytical techniques (genomics, proteomics, etc.). Professional data preparation, and most specifically data normalization is crucial for the success of any data mining project. Using routine hospital admission data, we demonstrate how explorative cluster analysis can identify meaningful result patterns. Based upon this feasibility study, the German Association for Clinical Chemistry and Laboratory Medicine is now supporting a research and software development project. |
|---|---|
| Beschreibung: | Gesehen am 06.03.2023 |
| Beschreibung: | Online Resource |
| ISSN: | 1439-0477 |
| DOI: | 10.1515/jlm.2010.041 |