EUVIMEDEUVIMED entdecken
Crossref

Goblet cells segmentation from confocal laser endomicroscopy with an improved U-Net

Dejian Su iD, Xiangwei Zheng iD, Shaotong Wang, Qingqing Qi, Zhen Li

Biomedical Physics & Engineering Express · 2023

DOI: 10.1088/2057-1976/ace89e ↗

Kurz gefasst

Fachartikel in der Zeitschrift Biomedical Physics & Engineering Express, 2023. Verfasst von Dejian Su und 4 weiteren Autor:innen. Beteiligte Einrichtung: School of Information Science and Engineering, Shandong Normal University, Jinan, People's Republic of China. Hauptthemen laut MEDLINE: Stomach Neoplasms, Precancerous Conditions. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 29 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Shandong Provincial Key Research and Development Program, Natural Science Foundation of Shandong Province.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Gastric intestinal metaplasia (GIM) is regarded as a remarkable precursor for the development of intestinal-type stomach cancer. Goblet cell (GC) segmentation is the crucial step for assessing the degree of GIM by confocal laser endomicroscopy (CLE). However, GC segmentation by hand is difficult, unreliable, and time-consuming. Meanwhile, due to the high resolution and noise interference of CLE images, existing segmentation approaches perform poorly on this task. To tackle those issues, we collected 343 confocal laser endomicroscopy images of 62 patients from a Grade-A tertiary hospital. Each CLE image is manually annotated and then verified three times by skilled medical specialists. Then, U-Net is improved by incorporating the pixel gradient attention mechanism, which focuses on color gradient information around GC and captures color gradient features to direct feature maps in the skip connection layer. At last, the model output vector is used to calculate the possibility map and generate the final segmentation area. Compared with mainstream models, our proposed GC segmentation method from CLE with an improved U-Net (GCSCLE) performs the better segmentation result when tested on our CLE dataset and achieved an IOU of 87.95% and a DICE of 86.64%. Our result shows, the performance of the GCSCLE can be compared with the manual CLE image processing in clinical settings, and it can improve segmentation accuracy and save time and costs.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Stomach NeoplasmsPrecancerous ConditionsHumansGoblet CellsMicroscopy, ConfocalMetaplasiaLasers

Stichwörter der Autor:innen

confocal laser endomicroscopydeep learninggastric intestinal metaplasiagoblet cell

Autor:innen und Institutionen

  • School of Information Science and Engineering, Shandong Normal University, Jinan, People's Republic of China.
  • State Key Laboratory of High-end Server & Storage Technology, Jinan, People's Republic of China.
  • Department of Gastroenterology, Qilu Hospital of Shandong University, Jinan, People's Republic of China.

ORCID-Profile: Dejian Su, Xiangwei Zheng

Förderung

  • National Natural Science Foundation of China
  • Shandong Provincial Key Research and Development Program
  • Natural Science Foundation of Shandong Province

Zitieren

APA

Su, D., Zheng, X., Wang, S., Qi, Q., & Li, Z. (2023). Goblet cells segmentation from confocal laser endomicroscopy with an improved U-Net. Biomedical Physics & Engineering Express. https://doi.org/10.1088/2057-1976/ace89e

Vancouver

Su D, Zheng X, Wang S, Qi Q, Li Z. Goblet cells segmentation from confocal laser endomicroscopy with an improved U-Net. Biomedical Physics & Engineering Express. 2023. doi:10.1088/2057-1976/ace89e

BibTeX
@article{Su2023,
  author = {Su, Dejian and Zheng, Xiangwei and Wang, Shaotong and Qi, Qingqing and Li, Zhen},
  title = {Goblet cells segmentation from confocal laser endomicroscopy with an improved U-Net},
  journal = {Biomedical Physics & Engineering Express},
  year = {2023},
  doi = {10.1088/2057-1976/ace89e}
}
Kostenlos starten →