It is already known that some features of fundus images are significantly altered in subjects suffering from certain pathologies; indeed, by proper observing the alterations it is possible to obtain important biomarkers that help the diagnosis of those pathologies. To this aim, we introduce a new specific tool for detecting and monitoring vessels alterations in fundus images over time. In this work, we show how to extend an existing protocol for fundus image analysis by introducing a skeletal representations (S-Rep) model that represents the blood vessel to analyze. Furthermore, we describe how to perform statistical analysis on those model using the composite principal nested spheres (CPNS) framework. S-Rep formally describes quasi-Tube geometry objects, which are suitable to represent blood vessels structures in fundus images; CPNS allows to statistically discover the presence of longitudinal changes in these structures by analyzing the principal modes. Thanks to these formalisms it is possible to detect specific regions featuring significant longitudinal variations. Our method is experimentally validated over simulated data generated over real images, and results prove the viability of the proposed method.

S-rep model for fundus image analysis

Calimeri F.;Surace L.
2017-01-01

Abstract

It is already known that some features of fundus images are significantly altered in subjects suffering from certain pathologies; indeed, by proper observing the alterations it is possible to obtain important biomarkers that help the diagnosis of those pathologies. To this aim, we introduce a new specific tool for detecting and monitoring vessels alterations in fundus images over time. In this work, we show how to extend an existing protocol for fundus image analysis by introducing a skeletal representations (S-Rep) model that represents the blood vessel to analyze. Furthermore, we describe how to perform statistical analysis on those model using the composite principal nested spheres (CPNS) framework. S-Rep formally describes quasi-Tube geometry objects, which are suitable to represent blood vessels structures in fundus images; CPNS allows to statistically discover the presence of longitudinal changes in these structures by analyzing the principal modes. Thanks to these formalisms it is possible to detect specific regions featuring significant longitudinal variations. Our method is experimentally validated over simulated data generated over real images, and results prove the viability of the proposed method.
2017
Bioinformatics
Biological Image And Signal Processing
Fundus Image
Health Informatics
Shape Analysis
Skeletal Representation
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.11770/417884
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact