Universität zu Köln

Windisch, Paul ORCID: 0000-0003-1040-4888, Weber, Pascal, Fuerweger, Christoph, Ehret, Felix, Kufeld, Markus, Zwahlen, Daniel and Muacevic, Alexander (2020). Implementation of model explainability for a basic brain tumor detection using convolutional neural networks on MRI slices. Neuroradiology, 62 (11). S. 1515 - 1519. NEW YORK: SPRINGER. ISSN 1432-1920

Full text not available from this repository.

Abstract

Purpose While neural networks gain popularity in medical research, attempts to make the decisions of a model explainable are often only made towards the end of the development process once a high predictive accuracy has been achieved. Methods In order to assess the advantages of implementing features to increase explainability early in the development process, we trained a neural network to differentiate between MRI slices containing either a vestibular schwannoma, a glioblastoma, or no tumor. Results Making the decisions of a network more explainable helped to identify potential bias and choose appropriate training data. Conclusion Model explainability should be considered in early stages of training a neural network for medical purposes as it may save time in the long run and will ultimately help physicians integrate the network's predictions into a clinical decision.

Item Type:	Journal Article
Creators:	Creators Email ORCID ORCID Put Code Windisch, Paul UNSPECIFIED orcid.org/0000-0003-1040-4888 UNSPECIFIED Weber, Pascal UNSPECIFIED UNSPECIFIED UNSPECIFIED Fuerweger, Christoph UNSPECIFIED UNSPECIFIED UNSPECIFIED Ehret, Felix UNSPECIFIED UNSPECIFIED UNSPECIFIED Kufeld, Markus UNSPECIFIED UNSPECIFIED UNSPECIFIED Zwahlen, Daniel UNSPECIFIED UNSPECIFIED UNSPECIFIED Muacevic, Alexander UNSPECIFIED UNSPECIFIED UNSPECIFIED
URN:	urn:nbn:de:hbz:38-330617
DOI:	10.1007/s00234-020-02465-1
Journal or Publication Title:	Neuroradiology
Volume:	62
Number:	11
Page Range:	S. 1515 - 1519
Date:	2020
Publisher:	SPRINGER
Place of Publication:	NEW YORK
ISSN:	1432-1920
Language:	English
Faculty:	Unspecified
Divisions:	Unspecified
Subjects:	no entry
Uncontrolled Keywords:	Keywords Language Clinical Neurology; Neuroimaging; Radiology, Nuclear Medicine & Medical Imaging Multiple languages
URI:	http://kups.ub.uni-koeln.de/id/eprint/33061