DBIS EPub

Studying the Potential of Multi-Target Classification to Characterize Combinations of Classes with Skewed Distribution

Schneck, Arne and Kalle, Sven and Pryss, Rüdiger and Schlee, Winfried and Probst, Thomas and Langguth, Berthold and Landgrebe, Michael and Reichert, Manfred and Spiliopoulou, Myra (2017) Studying the Potential of Multi-Target Classification to Characterize Combinations of Classes with Skewed Distribution. In: 30th IEEE International Symposium on Computer-Based Medical Systems (CBMS 2017), June 22 - 24, 2017, Thessaloniki, Greece. (Accepted for Publication)

[img] PDF - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
203Kb

Abstract

The identification of subpopulations with particu-lar characteristics with respect to a disease is important for personalized diagnostics and therapy design. For some diseases, the outcome is described by more than one target variable. An example is tinnitus: the perceived loudness of the phantom signal and the level of distress caused by it are both relevant targets for diagnosis and therapy. In this work, we study the potential of multi-target classification for the identification of those screening variables, which separate best among the different subpopula-tions of patients, paying particular attention to subpopulations with discordant value combinations of loudness and distress. We analyse the screening data of 1344 tinnitus patients from the University Hospital Regensburg, including questions from 7 questionnaires, and report on the performance of our workflow in target separation and in ranking the questionnaires’ variables on their discriminative power.

Item Type:Conference or Workshop Item (Paper)
Uncontrolled Keywords:multi-target classification on skewed data; tinnitus handicap; tinnitus loudness; medical mining
Subjects:DBIS Research > Publications
ID Code:1525
Deposited By: Prof. Dr. Manfred Reichert
BibTex Export:BibTeX
Deposited On:12 Jul 2017 17:12
Last Modified:12 Jul 2017 17:12

Repository Staff Only: item control page