EUVIMEDEUVIMED entdecken
Crossref

Transfer Learning with Limited Data for Mice Lung Segmentation in Synchrotron X-Ray Tomography

Andjela Blagojević, Ognjen Obradović, Tijana Geroski, Ognjen Pavić, Min Woo Kim, Nenad Filipović

Annals of Biomedical Engineering · 2026-08 · 54 (8) · 2763-2775

DOI: 10.1007/s10439-026-04036-6 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Annals of Biomedical Engineering, online erschienen am 03.03.2026. Verfasst von Andjela Blagojević und 5 weiteren Autor:innen. Beteiligte Einrichtung: Faculty of Engineering, University of Kragujevac, Kragujevac, Serbia. andjela.blagojevic@kg.ac.rs. Hauptthemen laut MEDLINE: Lung, Synchrotrons, Tomography, X-Ray Computed, Image Processing, Computer-Assisted. Die Arbeit nennt 19 Literaturangaben. Gefördert von: Ministarstvo Prosvete, Nauke i Tehnološkog Razvoja, Ministry of Education, Science and Technology.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

PURPOSE: Due to the small scale and complexity of the acinus, the demarcation between tissue and air is often unclear, making segmentation challenging. Accurate segmentation is essential for analyzing lung development, structural changes, and functional roles, offering insights into lung growth and disease. Traditional methods require large, annotated datasets, which are difficult to obtain, especially in medical imaging, where annotations are time-consuming and costly. Annotating mice lungs imaged by synchrotron radiation-based X-ray tomographic microscopy (SRXTM) is particularly difficult due to their small size, complexity, and subtle boundaries. METHODS: This paper presents a novel approach using transfer learning with limited data for segmenting mice lungs in SRXTM images. A U-net model, evaluated on a dataset of 59 image/mask pairs of lung sections, was pretrained on lung section images and then used to segment entire lungs in images. Segmentation accuracy improved through fine-tuning with transfer learning, using only 5 labeled entire lung image/mask pairs, with 2 pairs for testing. This highlights the potential of transfer learning in addressing the challenge of limited annotated data. RESULTS: The results show that U-net applied on the lung sections images achieves accurate prediction of the lung section masks, with the following metrics DSC = 0.9069, mIOU = 0.895, and BCE = 0.141. After transfer learning to entire lung images, segmentation results were DSC = 0.8915, mIOU = 0.8895, and BCE = 0.141. CONCLUSION: Experimental results demonstrate the effectiveness of the proposed method, achieving competitive segmentation performance compared to traditional approaches, while requiring significantly fewer annotated images.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

LungSynchrotronsTomography, X-Ray ComputedImage Processing, Computer-AssistedAnimalsMice

Stichwörter der Autor:innen

Lung segmentationSynchrotron radiation-based X-ray tomographic microscopy imagingTransfer learning with limited dataU-net

Autor:innen und Institutionen

  • Faculty of Engineering, University of Kragujevac, Kragujevac, Serbia. andjela.blagojevic@kg.ac.rs.
  • Bioengineering Research and Development Center (BioIRC), Kragujevac, Serbia. andjela.blagojevic@kg.ac.rs.
  • Faculty of Engineering, University of Kragujevac, Kragujevac, Serbia.
  • Bioengineering Research and Development Center (BioIRC), Kragujevac, Serbia.
  • Institute of Information Technologies Kragujevac, University of Kragujevac, Kragujevac, Serbia.
  • Pohang Accelerator Laboratory (PAL), POSTECH, Pohang, 37673, South Korea.

Förderung

  • Ministarstvo Prosvete, Nauke i Tehnološkog Razvoja
  • Ministry of Education, Science and Technology

Zitieren

APA

Blagojević, A., Obradović, O., Geroski, T., Pavić, O., Kim, M. W., & Filipović, N. (2026). Transfer Learning with Limited Data for Mice Lung Segmentation in Synchrotron X-Ray Tomography. Annals of Biomedical Engineering, 54(8), 2763-2775. https://doi.org/10.1007/s10439-026-04036-6

Vancouver

Blagojević A, Obradović O, Geroski T, Pavić O, Kim MW, Filipović N. Transfer Learning with Limited Data for Mice Lung Segmentation in Synchrotron X-Ray Tomography. Annals of Biomedical Engineering. 2026;54(8):2763-2775. doi:10.1007/s10439-026-04036-6

BibTeX
@article{Blagojevi2026,
  author = {Blagojević, Andjela and Obradović, Ognjen and Geroski, Tijana and Pavić, Ognjen and Kim, Min Woo and Filipović, Nenad},
  title = {Transfer Learning with Limited Data for Mice Lung Segmentation in Synchrotron X-Ray Tomography},
  journal = {Annals of Biomedical Engineering},
  year = {2026},
  volume = {54},
  number = {8},
  pages = {2763--2775},
  doi = {10.1007/s10439-026-04036-6}
}
Kostenlos starten →