| Publikationstyp: | Konferenz: Paper |
| Art der Begutachtung: | Peer review (Publikation) |
| Titel: | Explainable deep learning for medical time series data |
| Autor/-in: | Frick, Thomas Glüge, Stefan Rahimi, Abbas Benini, Luca Brunschwiler, Thomas |
| et. al: | No |
| DOI: | 10.1007/978-3-030-70569-5_15 |
| Tagungsband: | Wireless Mobile Communication and Healthcare |
| Seite(n): | 244 |
| Seiten bis: | 256 |
| Angaben zur Konferenz: | International Conference on Wireless Mobile Communication and Healthcare (MobiHealth), Online, 18 December 2020 |
| Erscheinungsdatum: | 21-Feb-2021 |
| Reihe: | Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering |
| Reihenzählung: | 362 |
| Verlag / Hrsg. Institution: | Springer |
| Verlag / Hrsg. Institution: | Cham |
| ISBN: | 978-3-030-70568-8 978-3-030-70569-5 |
| Sprache: | Englisch |
| Schlagwörter: | Explainable deep learning; Convolutional neural network; Explanation quality metric; Medical time series data |
| Fachgebiet (DDC): | 006: Spezielle Computerverfahren 362: Gesundheits- und Sozialdienste |
| Zusammenfassung: | Neural Networks are powerful classifiers. However, they are black boxes and do not provide explicit explanations for their decisions. For many applications, particularly in health care, explanations are essential for building trust in the model. In the field of computer vision, a multitude of explainability methods have been developed to analyze Neural Networks by explaining what they have learned during training and what factors influence their decisions. This work provides an overview of these explanation methods in form of a taxonomy. We adapt and benchmark the different methods to time series data. Further, we introduce quantitative explanation metrics that enable us to build an objective benchmarking framework with which we extensively rate and compare explainability methods. As a result, we show that the Grad-CAM++ algorithm outperforms all other methods. Finally, we identify the limits of existing explanation methods for specific datasets, with feature values close to zero. |
| URI: | https://digitalcollection.zhaw.ch/handle/11475/21995 |
| Volltext Version: | Publizierte Version |
| Lizenz (gemäss Verlagsvertrag): | Lizenz gemäss Verlagsvertrag |
| Departement: | Life Sciences und Facility Management |
| Organisationseinheit: | Institut für Computational Life Sciences (ICLS) |
| Enthalten in den Sammlungen: | Publikationen Life Sciences und Facility Management |
Dateien zu dieser Ressource:
Es gibt keine Dateien zu dieser Ressource.
Zur Langanzeige
Frick, T., Glüge, S., Rahimi, A., Benini, L., & Brunschwiler, T. (2021). Explainable deep learning for medical time series data [Conference paper]. Wireless Mobile Communication and Healthcare, 244–256. https://doi.org/10.1007/978-3-030-70569-5_15
Frick, T. et al. (2021) ‘Explainable deep learning for medical time series data’, in Wireless Mobile Communication and Healthcare. Cham: Springer, pp. 244–256. Available at: https://doi.org/10.1007/978-3-030-70569-5_15.
T. Frick, S. Glüge, A. Rahimi, L. Benini, and T. Brunschwiler, “Explainable deep learning for medical time series data,” in Wireless Mobile Communication and Healthcare, Feb. 2021, pp. 244–256. doi: 10.1007/978-3-030-70569-5_15.
FRICK, Thomas, Stefan GLÜGE, Abbas RAHIMI, Luca BENINI und Thomas BRUNSCHWILER, 2021. Explainable deep learning for medical time series data. In: Wireless Mobile Communication and Healthcare. Conference paper. Cham: Springer. 21 Februar 2021. S. 244–256. ISBN 978-3-030-70568-8
Frick, Thomas, Stefan Glüge, Abbas Rahimi, Luca Benini, and Thomas Brunschwiler. 2021. “Explainable Deep Learning for Medical Time Series Data.” Conference paper. In Wireless Mobile Communication and Healthcare, 244–56. Cham: Springer. https://doi.org/10.1007/978-3-030-70569-5_15.
Frick, Thomas, et al. “Explainable Deep Learning for Medical Time Series Data.” Wireless Mobile Communication and Healthcare, Springer, 2021, pp. 244–56, https://doi.org/10.1007/978-3-030-70569-5_15.
Alle Ressourcen in diesem Repository sind urheberrechtlich geschützt, soweit nicht anderweitig angezeigt.