![](bg4.png)
and study of the progression of a disease.
The method proposed for the segmentation takes
advantage of some of the techniques mentioned in
the previous section, besides new contributions, to
achieve the correct segmentation and to avoid the dis-
advantages of using them separately. Specifically, it
is based mainly on morphological operators but also
used principal curvature information, allowing an en-
hanced detection of the vessels on to the background
of the image. This algorithm has been compared with
several methods and, apart from improving the accu-
racy ratio, it must be stood out that the optic disc edge
is not detected as vessel unlike the most edge detec-
tion methods. Figure 4 shows our method along with
the proposed in (Martinez-Perez et al., 2007). The
non-detection of this feature is crucial to avoid multi-
ple errors in the tracking process.
(a) (b) (c)
Figure 4: Segmentation comparison: (a) Proposed method,
(b) (Martinez-Perez et al., 2007) method and (c) Manually
segmented image belonging to the DRIVE database.
On the other hand, the included measures are ac-
curate and reliable but also dependent on a correct
image analysis and rectification of some significant
points by the user.
About future work lines, a clinical validation will
be carried out to determine the specificity of the sys-
tem to distinguish between healthy and ill patients.
Afterwards, the method will be applied to analyze
the retinal microvascular architecture of children with
low birthweight and to use it as a prognostic marker
of cardiovascular risk.
ACKNOWLEDGEMENTS
This work has been funded by the project IMIDTA/
2010/47 and partially by projects Consolider-C
(SEJ2006-14301/PSIC), ”CIBER of Physiopathology
of Obesity and Nutrition, an initiative of ISCIII” and
Excellence Research Program PROMETEO (Gener-
alitat Valenciana. Conselleria de Educaci
´
on, 2008-
157). We would like to express our deep gratitude
to Imex Clinic S.L., the Department of Pediatrics of
General Hospital of Valencia and the Fundaci
´
on Of-
talmol
´
ogica del Mediterr
´
aneo for its participation as
well as the DRIVE database.
REFERENCES
Chaudhuri, S., Chatterjee, S., Katz, N., Nelson, M., and
Goldbaum, M. (1989). Detection of blood vessels in
retinal images using two-dimensional matched filters.
Medical Imaging, IEEE Transactions, 8(3):263 –269.
Hoover, A., Kouznetsova, V., and Goldbaum, M. (2000).
Locating blood vessels in retinal images by piecewise
threshold probing of a matched filter response. IEEE
Transactions on Medical Imaging, 19:203–210.
Jiang, X. and Mojon, D. (2003). Adaptive local threshold-
ing by verification-based multithreshold probing with
application to vessel detection in retinal images. IEEE
Trans. Pattern Anal. Mach. Intell., 25:131–137.
MacQueen, J. B. (1967). Some methods for classification
and analysis of multivariate observations. In Proc. of
the fifth Berkeley Symposium on Mathematical Statis-
tics and Probability, volume 1, pages 281–297.
Martinez-Perez, M., Hughes, A., Stanton, A., Thom, S.,
Bharath, A., and Parker, K. (1999). Retinal blood
vessel segmentation by means of scale-space analy-
sis and region growing. In Medical Image Comput-
ing and Computer-Assisted Intervention MICCAI 99,
volume 1679 of Lecture Notes in Computer Science,
pages 90–97.
Martinez-Perez, M. E., Hughes, A. D., Stanton, A. V.,
Thorn, S. A., Chapman, N., Bharath, A. A., and
Parker, K. H. (2002). Retinal vascular tree morphol-
ogy: A semi-automatic quantification. In Biomedical
Engineering, volume 49.
Martinez-Perez, M. E., Hughes, A. D., Thom, S. A.,
Bharath, A. A., and Parker, K. H. (2007). Segmen-
tation of blood vessels from red-free and fluorescein
retinal images. Medical Image Analysis, 11:47–61.
Niemeijer, M., Staal, J., van Ginneken, B., Loog, M., and
Abramoff, M. (2004). Comparative study of retinal
vessel segmentation methods on a new publicly avail-
able database. In SPIE Medical Imaging, volume
5370, pages 648–656.
Sinthanayothin, C., Boyce, J. F., Cook, H. L., and
Williamson, T. H. (1999). Automated localisation of
the optic disc, fovea, and retinal blood vessels from
digital colour fundus images. British Journal of Oph-
thalmology, 83:902–910.
Soille, P. (2003). Morphological Image Analysis: Prin-
ciples and Applications. Springer-Verlag New York,
Inc., 2nd edition.
Staal, J., Abrmoff, M. D., Niemeijer, M., Viergever, M. A.,
and Ginneken, B. V. (2004). Ridge-based vessel seg-
mentation in color images of the retina. IEEE Trans-
actions on Medical Imaging, 23:501–509.
Walter, T. and Klein, J. (2002). A computational approach
to diagnosis of diabetic retinopathy. In 6th Conference
on Systemics, Cybernetics and Informatics (SCI).
Zana, F. and Klein, J. C. (2001). Segmentation of vessel-
like patterns using mathematical morphology and cur-
vature evaluation. IEEE Transactions on Image Pro-
cessing, 10:1010–1019.
BIOSIGNALS 2012 - International Conference on Bio-inspired Systems and Signal Processing
324