Publication Details

Evaluating deep learning uncertainty measures in cephalometric landmark localization

DREVICKÝ Dušan and KODYM Oldřich. Evaluating deep learning uncertainty measures in cephalometric landmark localization. In: Proceedings of the 13th International Joint Conference on Biomedical Engineering Systems and Technologies - Volume 2: BIOIMAGING. Valetta: Institute for Systems and Technologies of Information, Control and Communication, 2020, pp. 213-220. ISBN 978-989-758-398-8. Available from: http://www.scitepress.org/DigitalLibrary/Link.aspx?doi=10.5220/0009375302130220
Czech title
Vyhodnocení odhadu nejistoty hlubokého učení při lokalizaci cefalometrických bodů zájmu
Type
conference paper
Language
english
Authors
Drevický Dušan, Ing. (FIT BUT)
Kodym Oldřich, Ing., Ph.D. (DCGM FIT BUT)
URL
Keywords

Landmark Localization, Cephalometric Landmarks, Deep Learning, Uncertainty Estimation.

Abstract

Cephalometric analysis is a key step in the process of dental treatment diagnosis, planning and surgery. Localization of a set of landmark points is an important but time-consuming and subjective part of this task. Deep learning is able to automate this process but the model predictions are usually given without any uncertainty information which is necessary in medical applications. This work evaluates three uncertainty measures applicable to deep learning models on the task of cephalometric landmark localization. We compare uncertainty estimation based on final network activation with an ensemble-based and a Bayesian-based approach. We conduct two experiments with elastically distorted cephalogram images and images containing undesirable horizontal skull rotation which the models should be able to detect as unfamiliar and unsuitable for automatic evaluation. We show that all three uncertainty measures have this detection capability and are a viable option when landmark localization with uncertainty estimation is required. 

Funding details

This work was supported by the company TESCAN 3DIM. We would also like to thank the same company for providing us with the CT data used in the experiments.

Published
2020
Pages
213-220
Proceedings
Proceedings of the 13th International Joint Conference on Biomedical Engineering Systems and Technologies - Volume 2: BIOIMAGING
Conference
7th International Conference on Bioimaging, BIOIMAGING 2020 - Part of 13th International Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2020, Valetta - Malta, MT
ISBN
978-989-758-398-8
Publisher
Institute for Systems and Technologies of Information, Control and Communication
Place
Valetta
DOI
UT WoS
000571473800027
EID Scopus
BibTeX
@INPROCEEDINGS{FITPUB12276,
   author = "Du\v{s}an Drevick\'{y} and Old\v{r}ich Kodym",
   title = "Evaluating deep learning uncertainty measures in cephalometric landmark localization",
   pages = "213--220",
   booktitle = "Proceedings of the 13th International Joint Conference on Biomedical Engineering Systems and Technologies - Volume 2: BIOIMAGING",
   year = 2020,
   location = "Valetta, ",
   publisher = "Institute for Systems and Technologies of Information, Control and Communication",
   ISBN = "978-989-758-398-8",
   doi = "10.5220/0009375302130220",
   language = "english",
   url = "https://www.fit.vut.cz/research/publication/12276"
}
Back to top