Case-Centred Multidimensional Scaling for Classification Visualisation in Medical Diagnosis

详细信息    查看全文

• 作者：Frank Klawonn (20) (21)
  Werner Lechner (22)
  Lorenz Grigull (22) (23)
  
• 刊名：Lecture Notes in Computer Science
• 出版年：2013
• 出版时间：2013
• 年：2013
• 卷：7798
• 期：1
• 页码：149-159
• 全文大小：378KB
• 参考文献：1. Grigull, L., Lechner, W.: Supporting diagnostic decisions using hybrid and complementary data mining applications: a pilot study in the pediatric emergency department. Pediatric Research聽71, 725鈥?31 (2012) CrossRef
  2. Kruskal, J., Wish, M.: Multidimensional Scaling. SAGE Publications, Beverly Hills (1978)
  3. Berthold, M., Borgelt, C., H枚ppner, F., Klawonn, F.: Guide to Intelligent Data Analysis: How to Intelligently Make Sense of Real Data. Springer, London (2010) CrossRef
  4. Shawe-Taylor, J., Cristianini, N.: Kernel Methods for Pattern Analysis. Cambridge University Press, Cambridge (2004) CrossRef
  5. Kuncheva, L.: Combining Pattern Classifiers: Methods and Algorithms. Wiley, Chichester (2004) CrossRef
  6. Pekalska, E., de Ridder, D., Duin, R., Kraaijveld, M.: A new method of generalizing Sammon mapping with application to algorithm speed-up. In: Boasson, M., Kaandorp, J., Tonino, J., Vosselman, M. (eds.) ASCI 1999: Proc. 5th Annual Conference of the Advanced School for Computing and Imaging, Delft, ASCI, pp. 221鈥?28 (1999)
  7. Rehm, F., Klawonn, F., Kruse, R.: MDS _{/ polar} : A new approach for dimension reduction to visualize high dimensional data. In: Famili, A.F., Kok, J.N., Pe帽a, J.M., Siebes, A., Feelders, A. (eds.) IDA 2005. LNCS, vol.聽3646, pp. 316鈥?27. Springer, Heidelberg (2005) CrossRef
  8. Rehm, F., Klawonn, F.: Improving angle based mappings. In: Tang, C., Ling, C.X., Zhou, X., Cercone, N.J., Li, X. (eds.) ADMA 2008. LNCS (LNAI), vol.聽5139, pp. 3鈥?4. Springer, Heidelberg (2008) CrossRef
  9. Fran莽ois, D., Wertz, V., Verleysen, M.: The concentration of fractional distances. IEEE Trans. Knowl. Data Eng.聽19(7), 873鈥?86 (2007) CrossRef
  10. Jayaram, B., Klawonn, F.: Can unbounded distance measures mitigate the curse of dimensionality? Int. Journ. Data Mining, Modelling and Management聽4, 361鈥?83 (2012) CrossRef
  
• 作者单位：Frank Klawonn (20) (21)
  Werner Lechner (22)
  Lorenz Grigull (22) (23)
  
  20. Department of Computer Science, Ostfalia University of Applied Sciences, Salzdahlumer Str. 46/48, D-38302, Wolfenbuettel, Germany
  21. Bioinformatics & Statistics, Helmholtz Centre for Infection Research, Inhoffenstr. 7, D-38124, Braunschweig, Germany
  22. Improved Medical Diagnostics Pte Ltd, 190 Middle Road, #19-05 Fortune Centre, Singapore, 188979
  23. Department of Pediatric Haematology and Oncology, Medical University, Carl-Neubergstr. 1, Hannover, Germany
  
• ISSN：1611-3349

文摘

Computer-based decision support can assist a medical doctor to find the right diagnosis. The knowledge and experience of the medical doctor is enhanced by a much larger data set of patients than the doctor will ever see in her or his life. The decision support system can derive possible diagnoses for a new patient based on a suitable classifier built on the patients in the patient database. However, since such a system cannot replace a medical doctor and should only support her or him, it should also provide information about the certainty of its recommendation. In this paper, we propose to visualise how close or similar the new patient is to others in the database by a modified multidimensional scaling technique that focuses on the correct positioning of the new patient in the visualisation. In this way, the medical doctor can easily see whether the diagnosis recommended by the system is reliable when all patients close to the new patient have the same diagnosis or whether it is quite uncertain when the new patient is surrounded by patients with different diagnoses.