KIT | KIT-Bibliothek | Impressum | Datenschutz

Methods for the frugal labeler: Multi-class semantic segmentation on heterogeneous labels

Schutera, M. 1; Rettenberger, L. ORCID iD icon 1; Pylatiuk, C. 1; Reischl, M. ORCID iD icon 1
1 Institut für Automation und angewandte Informatik (IAI), Karlsruher Institut für Technologie (KIT)


Deep learning increasingly accelerates biomedical research, deploying neural networks for multiple tasks, such as image classification, object detection, and semantic segmentation. However, neural networks are commonly trained supervised on large-scale, labeled datasets. These prerequisites raise issues in biomedical image recognition, as datasets are generally small-scale, challenging to obtain, expensive to label, and frequently heterogeneously labeled. Furthermore, heterogeneous labels are a challenge for supervised methods. If not all classes are labeled for an individual sample, supervised deep learning approaches can only learn on a subset of the dataset with common labels for each individual sample; consequently, biomedical image recognition engineers need to be frugal concerning their label and ground truth requirements. This paper discusses the effects of frugal labeling and proposes to train neural networks for multi-class semantic segmentation on heterogeneously labeled data based on a novel objective function. The objective function combines a class asymmetric loss with the Dice loss. The approach is demonstrated for training on the sparse ground truth of a heterogeneous labeled dataset, training within a transfer learning setting, and the use-case of merging multiple heterogeneously labeled datasets. ... mehr

Verlagsausgabe §
DOI: 10.5445/IR/1000143280
Veröffentlicht am 03.03.2022
DOI: 10.1371/journal.pone.0263656
Zitationen: 6
Web of Science
Zitationen: 4
Zitationen: 6
Cover der Publikation
Zugehörige Institution(en) am KIT Institut für Automation und angewandte Informatik (IAI)
Publikationstyp Zeitschriftenaufsatz
Publikationsjahr 2022
Sprache Englisch
Identifikator ISSN: 1932-6203
KITopen-ID: 1000143280
HGF-Programm 47.14.02 (POF IV, LK 01) Information Storage and Processing in the Cell Nucleus
Erschienen in PLoS ONE
Verlag Public Library of Science (PLoS)
Band 17
Heft 2
Seiten Art.-Nr.: e0263656
Vorab online veröffentlicht am 08.02.2022
Nachgewiesen in Dimensions
Web of Science
KIT – Die Forschungsuniversität in der Helmholtz-Gemeinschaft
KITopen Landing Page