EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Masked Modeling-Based Ultrasound Image Classification via Self-Supervised Learning

Kele Xu iD, Kang You iD, Boqing Zhu iD, Ming Feng iD, Dawei Feng iD, Cheng Yang iD

IEEE Open Journal of Engineering in Medicine and Biology · 2024 · 5 · 226-237

DOI: 10.1109/ojemb.2024.3374966 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Open Journal of Engineering in Medicine and Biology, 2024. Verfasst von Kele Xu und 5 weiteren Autor:innen. Beteiligte Einrichtung: National University of Defense Technology Changsha 410073 China. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 82 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Recently, deep learning-based methods have emerged as the preferred approach for ultrasound data analysis. However, these methods often require large-scale annotated datasets for training deep models, which are not readily available in practical scenarios. Additionally, the presence of speckle noise and other imaging artifacts can introduce numerous hard examples for ultrasound data classification. In this paper, drawing inspiration from self-supervised learning techniques, we present a pre-training method based on mask modeling specifically designed for ultrasound data. Our study investigates three different mask modeling strategies: random masking, vertical masking, and horizontal masking. By employing these strategies, our pre-training approach aims to predict the masked portion of the ultrasound images. Notably, our method does not rely on externally labeled data, allowing us to extract representative features without the need for human annotation. Consequently, we can leverage unlabeled datasets for pre-training. Furthermore, to address the challenges posed by hard samples in ultrasound data, we propose a novel hard sample mining strategy. To evaluate the effectiveness of our proposed method, we conduct experiments on two datasets. The experimental results demonstrate that our approach outperforms other state-of-the-art methods in ultrasound image classification. This indicates the superiority of our pre-training method and its ability to extract discriminative features from ultrasound data, even in the presence of hard examples.

Schlagwörter

Stichwörter der Autor:innen

Pre-trainingmasked modelingself-supervisedultrasound image

Autor:innen und Institutionen

  • National University of Defense Technology Changsha 410073 China.
  • Shanghai Jiao Tong University Shanghai 200240 China.
  • TongJi University Shanghai 200070 China.

ORCID-Profile: Kele Xu, Kang You, Boqing Zhu, Ming Feng, Dawei Feng, Cheng Yang

Zitieren

APA

Xu, K., You, K., Zhu, B., Feng, M., Feng, D., & Yang, C. (2024). Masked Modeling-Based Ultrasound Image Classification via Self-Supervised Learning. IEEE Open Journal of Engineering in Medicine and Biology, 5, 226-237. https://doi.org/10.1109/ojemb.2024.3374966

Vancouver

Xu K, You K, Zhu B, Feng M, Feng D, Yang C. Masked Modeling-Based Ultrasound Image Classification via Self-Supervised Learning. IEEE Open Journal of Engineering in Medicine and Biology. 2024;5:226-237. doi:10.1109/ojemb.2024.3374966

BibTeX
@article{Xu2024,
  author = {Xu, Kele and You, Kang and Zhu, Boqing and Feng, Ming and Feng, Dawei and Yang, Cheng},
  title = {Masked Modeling-Based Ultrasound Image Classification via Self-Supervised Learning},
  journal = {IEEE Open Journal of Engineering in Medicine and Biology},
  year = {2024},
  volume = {5},
  pages = {226--237},
  doi = {10.1109/ojemb.2024.3374966}
}
Kostenlos starten →