Next Article in Journal
Clinical Applications of Nano-Hydroxyapatite in Dentistry
Next Article in Special Issue
A Self-Supervised Detail-Sensitive ViT-Based Model for COVID-19 X-ray Image Diagnosis: SDViT
Previous Article in Journal
Predicting Hospital Admissions to Reduce Crowding in the Emergency Departments
Previous Article in Special Issue
A Hybrid Linear Iterative Clustering and Bayes Classification-Based GrabCut Segmentation Scheme for Dynamic Detection of Cervical Cancer
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Applied Sciences
Volume 12
Issue 21
10.3390/app122110763
applsci-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Weakly Supervised Learning with Positive and Unlabeled Data for Automatic Brain Tumor Segmentation

by
Daniel Wolf
1,2,
Sebastian Regnery
3,
Rafal Tarnawski
4,
Barbara Bobek-Billewicz
5,
Joanna Polańska
6 and
Michael Götz
1,*
1
Section Experimental Radiology, Department for Diagnostic and Internventional Radiology, University Hospital Ulm, 89081 Ulm, Germany
2
Visual Computing Group, Institute of Media Informatics, Ulm University, 89081 Ulm, Germany
3
Department of Radiation Oncology, Heidelberg University Hospital, 69120 Heidelberg, Germany
4
IIIrd Radiotherapy and Chemotherapy Department, Maria Sklodowska-Curie National Research Institute of Oncology, 44-101 Gliwice, Poland
5
Radiology and Diagnostic Imaging Department, Maria Sklodowska-Curie National Research Institute of Oncology, 44-102 Gliwice, Poland
6
Department of Data Science and Engineering, The Silesian University of Technology, 44-100 Gliwice, Poland
*
Author to whom correspondence should be addressed.
Appl. Sci. 2022, 12(21), 10763; https://doi.org/10.3390/app122110763
Submission received: 5 October 2022 / Revised: 14 October 2022 / Accepted: 19 October 2022 / Published: 24 October 2022
(This article belongs to the Special Issue Applications of Artificial Intelligence in Medical Imaging)
Browse Figures
Versions Notes

Featured Application

The proposed solution provides a quick approach for the annotation of the necessary training data to create an application-specific machine learning model that can be used to annotate medical image studies.

Abstract

A major obstacle to the learning-based segmentation of healthy and tumorous brain tissue is the requirement of having to create a fully labeled training dataset. Obtaining these data requires tedious and error-prone manual labeling with respect to both tumor and non-tumor areas. To mitigate this problem, we propose a new method to obtain high-quality classifiers from a dataset with only small parts of labeled tumor areas. This is achieved by using positive and unlabeled learning in conjunction with a domain adaptation technique. The proposed approach leverages the tumor volume, and we show that it can be either derived with simple measures or completely automatic with a proposed estimation method. While learning from sparse samples allows reducing the necessary annotation time from 4 h to 5 min, we show that the proposed approach further reduces the necessary annotation by roughly 50% while maintaining comparative accuracies compared to traditionally trained classifiers with this approach.
Keywords: image segmentation; tumor segmentation; machine learning; random forests; MRI; PU-learning; semi-supervised learning; weak annotations; sparse annotation; weak supervision image segmentation; tumor segmentation; machine learning; random forests; MRI; PU-learning; semi-supervised learning; weak annotations; sparse annotation; weak supervision
Graphical Abstract

Share and Cite

MDPI and ACS Style

Wolf, D.; Regnery, S.; Tarnawski, R.; Bobek-Billewicz, B.; Polańska, J.; Götz, M. Weakly Supervised Learning with Positive and Unlabeled Data for Automatic Brain Tumor Segmentation. Appl. Sci. 2022, 12, 10763. https://doi.org/10.3390/app122110763

AMA Style

Wolf D, Regnery S, Tarnawski R, Bobek-Billewicz B, Polańska J, Götz M. Weakly Supervised Learning with Positive and Unlabeled Data for Automatic Brain Tumor Segmentation. Applied Sciences. 2022; 12(21):10763. https://doi.org/10.3390/app122110763

Chicago/Turabian Style

Wolf, Daniel, Sebastian Regnery, Rafal Tarnawski, Barbara Bobek-Billewicz, Joanna Polańska, and Michael Götz. 2022. "Weakly Supervised Learning with Positive and Unlabeled Data for Automatic Brain Tumor Segmentation" Applied Sciences 12, no. 21: 10763. https://doi.org/10.3390/app122110763

APA Style

Wolf, D., Regnery, S., Tarnawski, R., Bobek-Billewicz, B., Polańska, J., & Götz, M. (2022). Weakly Supervised Learning with Positive and Unlabeled Data for Automatic Brain Tumor Segmentation. Applied Sciences, 12(21), 10763. https://doi.org/10.3390/app122110763

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop