Skip navigation
  •  Inicio
  • UDC 
    • Cómo depositar
    • Políticas do RUC
    • FAQ
    • Dereitos de Autor
    • Máis información en INFOguías UDC
  • Percorrer 
    • Comunidades
    • Buscar por:
    • Data de publicación
    • Autor
    • Título
    • Materia
  • Axuda
    • español
    • Gallegan
    • English
  • Acceder
  •  Galego 
    • Español
    • Galego
    • English
  
Ver ítem 
  •   RUC
  • Facultade de Informática
  • Investigación (FIC)
  • Ver ítem
  •   RUC
  • Facultade de Informática
  • Investigación (FIC)
  • Ver ítem
JavaScript is disabled for your browser. Some features of this site may not work without it.

Enhancing Retinal Blood Vessel Segmentation through Self-Supervised Pre-Training

Thumbnail
Ver/abrir
José_Morano_2020_Enhancing_Retinal_Blood_Vessel_Segmentation.pdf (812.1Kb)
Use este enlace para citar
http://hdl.handle.net/2183/26575
Atribución 4.0 International (CC BY 4.0)
A non ser que se indique outra cousa, a licenza do ítem descríbese como Atribución 4.0 International (CC BY 4.0)
Coleccións
  • Investigación (FIC) [1685]
Metadatos
Mostrar o rexistro completo do ítem
Título
Enhancing Retinal Blood Vessel Segmentation through Self-Supervised Pre-Training
Autor(es)
Morano, José
Hervella, Álvaro S.
Barreira, Noelia
Novo Buján, Jorge
Rouco, J.
Data
2020-08-25
Cita bibliográfica
Morano, J.; Hervella, Á.S.; Barreira, N.; Novo, J.; Rouco, J. Enhancing Retinal Blood Vessel Segmentation through Self-Supervised Pre-Training. Proceedings 2020, 54, 44. https://doi.org/10.3390/proceedings2020054044
Resumo
[Abstract] The segmentation of the retinal vasculature is fundamental in the study of many diseases. However, its manual completion is problematic, which motivates the research on automatic methods. Nowadays, these methods usually employ Fully Convolutional Networks (FCNs), whose success is highly conditioned by the network architecture and the availability of many annotated data, something infrequent in medicine. In this work, we present a novel application of self-supervised multimodal pre-training to enhance the retinal vasculature segmentation. The experiments with diverse FCN architectures demonstrate that, independently of the architecture, this pre-training allows one to overcome annotated data scarcity and leads to significantly better results with less training on the target task.
Palabras chave
Self-supervised learning
Transfer learning
Multimodal
Retinal vasculature segmentation
 
Versión do editor
https://doi.org/10.3390/proceedings2020054044
Dereitos
Atribución 4.0 International (CC BY 4.0)
ISSN
2504-3900

Listar

Todo RUCComunidades e colecciónsPor data de publicaciónAutoresTítulosMateriasGrupo de InvestigaciónTitulaciónEsta colecciónPor data de publicaciónAutoresTítulosMateriasGrupo de InvestigaciónTitulación

A miña conta

AccederRexistro

Estatísticas

Ver Estatísticas de uso
Sherpa
OpenArchives
OAIster
Scholar Google
UNIVERSIDADE DA CORUÑA. Servizo de Biblioteca.    DSpace Software Copyright © 2002-2013 Duraspace - Suxestións