EUVIMEDEUVIMED entdecken
Crossref

Self-supervised-RCNN for medical image segmentation with limited data annotation

Banafshe Felfeliyan, Nils D Forkert, Abhilash Hareendranathan, David Cornel, Yuyue Zhou, Gregor Kuntze, Jacob L Jaremko, Janet L Ronsky

Computerized Medical Imaging and Graphics · 2023

DOI: 10.1016/j.compmedimag.2023.102297 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Computerized Medical Imaging and Graphics, 2023. Verfasst von Banafshe Felfeliyan und 7 weiteren Autor:innen. Beteiligte Einrichtung: Department of Biomedical Engineering, University of Calgary, Calgary, AB, Canada; McCaig Institute for Bone & Joint Health, University of Calgary, Ca…. Hauptthemen laut MEDLINE: Data Curation, Osteoarthritis. Die Arbeit wurde bisher 32-mal zitiert (Crossref) und nennt 48 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Many successful methods developed for medical image analysis based on machine learning use supervised learning approaches, which often require large datasets annotated by experts to achieve high accuracy. However, medical data annotation is time-consuming and expensive, especially for segmentation tasks. To overcome the problem of learning with limited labeled medical image data, an alternative deep learning training strategy based on self-supervised pretraining on unlabeled imaging data is proposed in this work. For the pretraining, different distortions are arbitrarily applied to random areas of unlabeled images. Next, a Mask-RCNN architecture is trained to localize the distortion location and recover the original image pixels. This pretrained model is assumed to gain knowledge of the relevant texture in the images from the self-supervised pretraining on unlabeled imaging data. This provides a good basis for fine-tuning the model to segment the structure of interest using a limited amount of labeled training data. The effectiveness of the proposed method in different pretraining and fine-tuning scenarios was evaluated based on the Osteoarthritis Initiative dataset with the aim of segmenting effusions in MRI datasets of the knee. Applying the proposed self-supervised pretraining method improved the Dice score by up to 18% compared to training the models using only the limited annotated data. The proposed self-supervised learning approach can be applied to many other medical image analysis tasks including anomaly detection, segmentation, and classification.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Data CurationOsteoarthritisHumansKnee JointMachine LearningImage Processing, Computer-AssistedSupervised Machine Learning

Stichwörter der Autor:innen

Deep learningImage segmentationLimited annotationsMagnetic resonance imaging (MRI)Self-supervised learning

Autor:innen und Institutionen

  • Department of Biomedical Engineering, University of Calgary, Calgary, AB, Canada; McCaig Institute for Bone & Joint Health, University of Calgary, Calgary, AB, Canada. Electronic address: Banafshe.Felfeliyan@uCalgary.ca.
  • Department of Biomedical Engineering, University of Calgary, Calgary, AB, Canada.
  • Department of Radiology & Diagnostic Imaging, University of Alberta, Edmonton, AB, Canada.
  • McCaig Institute for Bone & Joint Health, University of Calgary, Calgary, AB, Canada.
  • Department of Biomedical Engineering, University of Calgary, Calgary, AB, Canada; McCaig Institute for Bone & Joint Health, University of Calgary, Calgary, AB, Canada; Mechanical & Manufacturing Engineering, University…

Zitieren

APA

Felfeliyan, B., Forkert, N. D., Hareendranathan, A., Cornel, D., Zhou, Y., Kuntze, G., Jaremko, J. L., & Ronsky, J. L. (2023). Self-supervised-RCNN for medical image segmentation with limited data annotation. Computerized Medical Imaging and Graphics. https://doi.org/10.1016/j.compmedimag.2023.102297

Vancouver

Felfeliyan B, Forkert ND, Hareendranathan A, Cornel D, Zhou Y, Kuntze G, et al. Self-supervised-RCNN for medical image segmentation with limited data annotation. Computerized Medical Imaging and Graphics. 2023. doi:10.1016/j.compmedimag.2023.102297

BibTeX
@article{Felfeliyan2023,
  author = {Felfeliyan, Banafshe and Forkert, Nils D and Hareendranathan, Abhilash and Cornel, David and Zhou, Yuyue and Kuntze, Gregor and Jaremko, Jacob L and Ronsky, Janet L},
  title = {Self-supervised-RCNN for medical image segmentation with limited data annotation},
  journal = {Computerized Medical Imaging and Graphics},
  year = {2023},
  doi = {10.1016/j.compmedimag.2023.102297}
}
Kostenlos starten →