EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Adaptive neighborhood triplet loss: enhanced segmentation of dermoscopy datasets by mining pixel information

Mohan Xu, Lena Wiese

International Journal of Computer Assisted Radiology and Surgery · 2024

DOI: 10.1007/s11548-024-03241-9 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Computer Assisted Radiology and Surgery, 2024. Verfasst von Mohan Xu und 1 weiteren Autor:innen. Beteiligte Einrichtung: Fraunhofer Institute for Toxicology and Experimental Medicine, Hannover, Germany. mohan.xu@item.fraunhofer.de. Hauptthemen laut MEDLINE: Dermoscopy, Deep Learning, Skin Neoplasms, Data Mining. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 19 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Fraunhofer Internal Programs.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Purpose

The integration of deep learning in image segmentation technology markedly improves the automation capabilities of medical diagnostic systems, reducing the dependence on the clinical expertise of medical professionals. However, the accuracy of image segmentation is still impacted by various interference factors encountered during image acquisition.

Methods

To address this challenge, this paper proposes a loss function designed to mine specific pixel information which dynamically changes during training process. Based on the triplet concept, this dynamic change is leveraged to drive the predicted boundaries of images closer to the real boundaries.

Results

Extensive experiments on the PH2 and ISIC2017 dermoscopy datasets validate that our proposed loss function overcomes the limitations of traditional triplet loss methods in image segmentation applications. This loss function not only enhances Jaccard indices of neural networks by 2.42 $$\%$$ % and 2.21 $$\%$$ % for PH2 and ISIC2017, respectively, but also neural networks utilizing this loss function generally surpass those that do not in terms of segmentation performance.

Conclusion

This work proposed a loss function that mined the information of specific pixels deeply without incurring additional training costs, significantly improving the automation of neural networks in image segmentation tasks. This loss function adapts to dermoscopic images of varying qualities and demonstrates higher effectiveness and robustness compared to other boundary loss functions, making it suitable for image segmentation tasks across various neural networks.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

DermoscopyDeep LearningSkin NeoplasmsData MiningImage Processing, Computer-AssistedImage Interpretation, Computer-AssistedHumansNeural Networks, ComputerMelanomaDatasets as Topic

Stichwörter der Autor:innen

Computer-aided diagnosisImage segmentationNeural networksTriplet loss

Autor:innen und Institutionen

  • Fraunhofer Institute for Toxicology and Experimental Medicine, Hannover, Germany. mohan.xu@item.fraunhofer.de.
  • Fraunhofer Institute for Toxicology and Experimental Medicine, Hannover, Germany.
  • Institute of Computer Science, Goethe University Frankfurt, Frankfurt a. M., Germany.

Förderung

  • Fraunhofer Internal Programs

Zitieren

APA

Xu, M., & Wiese, L. (2024). Adaptive neighborhood triplet loss: enhanced segmentation of dermoscopy datasets by mining pixel information. International Journal of Computer Assisted Radiology and Surgery. https://doi.org/10.1007/s11548-024-03241-9

Vancouver

Xu M, Wiese L. Adaptive neighborhood triplet loss: enhanced segmentation of dermoscopy datasets by mining pixel information. International Journal of Computer Assisted Radiology and Surgery. 2024. doi:10.1007/s11548-024-03241-9

BibTeX
@article{Xu2024,
  author = {Xu, Mohan and Wiese, Lena},
  title = {Adaptive neighborhood triplet loss: enhanced segmentation of dermoscopy datasets by mining pixel information},
  journal = {International Journal of Computer Assisted Radiology and Surgery},
  year = {2024},
  doi = {10.1007/s11548-024-03241-9}
}
Kostenlos starten →