EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

BézierSeg: Parametric Shape Representation for Fast Object Segmentation in Medical Images

Haichou Chen, Yishu Deng iD, Bin Li, Zeqin Li, Haohua Chen, Bingzhong Jing, Chaofeng Li

Life · 09.03.2023 · 13 (3) · 743

DOI: 10.3390/life13030743 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Life, online erschienen am 09.03.2023. Verfasst von Haichou Chen und 6 weiteren Autor:innen. Beteiligte Einrichtung: Collaborative Innovation Center for Cancer Medicine, State Key Laboratory of Oncology in South China, Guangdong Key Laboratory of Nasopharyngeal Carc…. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 31 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Fundamental Research Funds for the Central Universities.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Delineating the lesion area is an important task in image-based diagnosis. Pixel-wise classification is a popular approach to segmenting the region of interest. However, at fuzzy boundaries, such methods usually result in glitches, discontinuity or disconnection, inconsistent with the fact that lesions are solid and smooth.

Methods

To overcome these problems and to provide an efficient, accurate, robust and concise solution that simplifies the whole segmentation pipeline in AI-assisted applications, we propose the BézierSeg model which outputs Bézier curves encompassing the region of interest.

Results

Directly modeling the contour with analytic equations ensures that the segmentation is connected and continuous, and that the boundary is smooth. In addition, it offers sub-pixel accuracy. Without loss of precision, the Bézier contour can be resampled and overlaid with images of any resolution. Moreover, clinicians can conveniently adjust the curve’s control points to refine the result.

Conclusions

Our experiments show that the proposed method runs in real time and achieves accuracy competitive with pixel-wise segmentation models.

Schlagwörter

Stichwörter der Autor:innen

Bézier curvesdeep learningobject segmentation

Autor:innen und Institutionen

  • Collaborative Innovation Center for Cancer Medicine, State Key Laboratory of Oncology in South China, Guangdong Key Laboratory of Nasopharyngeal Carcinoma Diagnosis and Therapy, Sun Yat-sen University Cancer Center, Gua…
  • Department of Information, Sun Yat-sen University Cancer Center, Guangzhou 510060, China.
  • School of Electronics and Information Technology (School of Microelectronics), Sun Yat-sen University, Guangzhou 510275, China.

ORCID-Profile: Yishu Deng

Förderung

  • National Natural Science Foundation of China
  • Fundamental Research Funds for the Central Universities

Zitieren

APA

Chen, H., Deng, Y., Li, B., Li, Z., Chen, H., Jing, B., & Li, C. (2023). BézierSeg: Parametric Shape Representation for Fast Object Segmentation in Medical Images. Life, 13(3), 743. https://doi.org/10.3390/life13030743

Vancouver

Chen H, Deng Y, Li B, Li Z, Chen H, Jing B, et al. BézierSeg: Parametric Shape Representation for Fast Object Segmentation in Medical Images. Life. 2023;13(3):743. doi:10.3390/life13030743

BibTeX
@article{Chen2023,
  author = {Chen, Haichou and Deng, Yishu and Li, Bin and Li, Zeqin and Chen, Haohua and Jing, Bingzhong and Li, Chaofeng},
  title = {BézierSeg: Parametric Shape Representation for Fast Object Segmentation in Medical Images},
  journal = {Life},
  year = {2023},
  volume = {13},
  number = {3},
  pages = {743},
  doi = {10.3390/life13030743}
}
Kostenlos starten →