The use of a modified causal index facilitates the response interpretation of an artificial neural network (ANN) trained by data of whole body plethysmography in a knock-out mouse model

R. Linder, D. Theegarten, S. Mayer, O. Anhenn, M. Ebsen, J. Schwarze, J. Neesen, H. Sudhoff, S. J. Pöppl, M. Wagner*

*Corresponding author for this work

Abstract

Motivation: Human decisions often proceed in two steps. Initially those most preferred are chosen followed by a subsequent choice of these preferences. Applying one artificial neural network (ANN), a classification is limited to the preselection process. The final categorization is only possible by a subsequent ANN that distinguishes the pre-chosen classes. Existing strategies using coupled ANNs are discussed and a new approach particularly suited for multiclass classification problems is introduced (‘Subsequent ANN’, SANN). Results: Evaluating a simulated data base comprising 3 classes, classification results of SANN were obviously superior to those achieved by ANN. To evaluate a real-world data base the microarray benchmark GCM (14 classes) was chosen. The ANN results reached 72%, comparable to previous results. Using SANN, up to 81% of the tumors were correctly classified.
Translated title of the contributionDer einsatz eines modifizierten causal-index erleichtert die interpretation des antwortverhaltens eines mit daten einer whole-body-plethysmographie an einem knock-out-mausmodell trainierten artifiziellen neuronalen netzwerks (ANN)
Original languageEnglish
JournalAtemwegs- und Lungenkrankheiten
Volume29
Issue number7
Pages (from-to)340-343
Number of pages4
ISSN0341-3055
Publication statusPublished - 01.07.2003

Fingerprint

Dive into the research topics of 'The use of a modified causal index facilitates the response interpretation of an artificial neural network (ANN) trained by data of whole body plethysmography in a knock-out mouse model'. Together they form a unique fingerprint.

Cite this