HAL Id: hal-01539980
https://hal.archives-ouvertes.fr/hal-01539980
Submitted on 15 Jun 2017
HAL is a multi-disciplinary open access
archive for the deposit and dissemination of
sci-entific research documents, whether they are
pub-lished or not. The documents may come from
teaching and research institutions in France or
abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est
destinée au dépôt et à la diffusion de documents
scientifiques de niveau recherche, publiés ou non,
émanant des établissements d’enseignement et de
recherche français ou étrangers, des laboratoires
publics ou privés.
Copyright
Enhancing Eye Fundus Images for Diabetic Retinopathy
Screening
Guillaume Noyel, Michel Jourlin, Michel Smans, Rebecca Thomas, Simon Iles,
Gavin Bhakta, Andrew Crowder, David Owens, Peter Boyle
To cite this version:
“Prevention is better than healing
because it saves the labour of being sick.” Thomas Adams, 1680
Visit us at www.i-pri.org
“P “P “P “P “P “P “P “P “P “P “P“Prererererererererererererererereveveveveveveveveveveveveventntiontntntntntntntntioioioioioioioioioioioioioioioioioion isn n n n n n n n n n n n isisisisisisisisisis b b b b b b b b b b b b bet b b b b b b b bet b b b b b betetetetetetteetteetetetetetetetetetetetetteter teteteteteteteteteteteteteteteteter r r r r r r r r r r thr r r r r r r r r r ththththththththththththththththththththththththththththanananananananananananananananananananananan h h h h h h h h h hea h h h h h h h h h h h h h h h h h h h heaeaeaeaeaeaeaeaeaeaeaeaeaeaeaeaeaeaeaeaeaeaealililililililililililililililililililililililililingngngngngngngngngngngngngngngngngngngngngngngngngng “P “P “P “P “P “P “P “P “P “P “P ntntntntntntntntnt eaeaeaeaeaeaeaea be be be be be be be be be be be be becacacacacacacacacacaususususususususususe e e e e e e e e e e itititititititititititit s s s s s s s savavesavavavavavavavavaveseseseseseseseseseses t t t t t t t t t t t thehehehehehehehehehehehe l l l l l l l l l labababababababababababououououououououououour r ofr r r r r r r r r ofofofofofofofofof b b b b b b b b b b b b beieieieieieieieieieieieingngngngngngngngngngngngng s sic s s s s s s s s s s sicicicicicicicicicick.k.k.k.k.k.k.k.k.k.k.k.” ” ” ” ” ” ” ” ” be be be be be be be be usususususususususususususe e e e e e e e e e e e e e itititititititititititit avavavavavavavav hehehehehehehehehehe ouououououououour ouour r r r r r ofofofofofofof eieingeieieieieieieingngngngngngngngngngngngngngngngng s s s ” ” ” ” ” ” ” be be be be be be
becacacacacacacacacausususususususususususususususususususususe e e e e e e e e e e e itititititititit s s s s s s s s savavavavavavavavavavav ngngngngngngngng ThoThoThoThoThoThoThoThoThoThoThoThoThoThomasmasmasmasmasmas Admasmasmasmasmasmasmasmasmasmasmas Ad Ad Ad Ad Ad Ad Ad Ad Ad Adamsamsamsamsamsamsams, 1amsamsamsamsamsamsamsamsamsamsamsamsams, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1680680680680680680680680680680680680680680680680680680680
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
s
s
s
s
s
s
s
si
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
i
it
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
t
u
u
u
u
u
u
u
u
u
u
u
us
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
a
a
a
a
a
a
a
a
a
a
t
t
t
t
t
t
t
t
t
t
t
t
w
w
w
ww
w
w
w
w
w
w
w
w
ww
w
w
w
w
w
w
w
w
w
w.
w
w
w
w
w
w
w
w
w
.
.i
.
.
.
.
.
.
.
.
i-
i
i
i
i
i
i
i
i
i
-
-p
-
-
-
-
-
-
p
pr
p
p
p
p
p
p
p
p
p
p
p
p
p
r
ri
r
r
r
r
r
r
r
r
r
i
i
i
i
i
i
i
i
i.
i
.
.o
.
.
.
.
.
.
.
.
or
o
o
o
o
o
o
o
o
o
rg
r
r
r
r
r
r
r
r
r
r
r
r
g
g
g
g
g
g
g
g
g
g
g
g
g
g
Guillaume Noyel1, Michel Jourlin2, Michel Smans1, Rebecca Thomas3, Simon Iles4, Gavin Bhakta4, Andrew Crowder4, David Owens3, Peter Boyle1,51. International Prevention Research Institute, Lyon, France | 2. Laboratoire Hubert Curien, UMR CNRS 5516, Saint-Etienne, France | 3. Diabetes Research Group, Institute of Life Sciences, College of Medicine, Swansea University, Wales, UK 4. Diabetic Eye Screening Wales, Wales, UK | 5. Strathclyde Institute of Global Public Health, University of Strathclyde, Glasgow, UK | Correspondence: guillaume.noyel@i-pri.org
A
ABSTR
BSTR
BSTR
BSTR
BSTRA
ACT
CT
CT
CT
Many eye fundus images present strong variations of contrast which can be a limitation to the diagnosis of the retinopathy. Either some lesions are not taken into account or only a limited part of the domain of the image can be read. Graders have to manually adjust the contrast, which is tedious and not easily reproducible.
We have developed an automatic system, which standardises the colour contrast across the whole domain of the image. The method is consistent with the physical principles or image formation and ensures that the colour aspect of lesions such as micro-aneurysms or anatomical structures such as veins are similar. It is more powerful than the existing grey level methods.
We have tested our approach on several thousand images acquired in good or in harsher conditions. Some were bright while others were dark. Expert graders have checked the enhanced images. Diagnosis becomes more obvious and the grading more comfortable.
Another limitation for the diagnosis is that images of the same patient acquired for different examinations cannot be directly superimposed. Indeed, the eye of the patient is never in the exact same position, the image is a projection of a 3D scene into the plane of the sensor, the optics of the camera creates a radial deformation and the colour of the image may have changed.
We have developed an automatic method to superimpose eye fundus images acquired in the same position (nasal or macular). It is based on contrast standardisation, matching of salient points and a deformation model taking into account two radial distortions.
We have performed tests for 69 patients with pairs of retinal examinations acquired in good conditions at an interval of one year with and without the same camera. A similar test has been performed on 5 patients with 20 pairs acquired in harsher conditions. A minimum of 96% of pairs were correctly superimposed. This is an important step towards the longitudinal analysis of large public health databases.
C
CONTR
ONTR
ONTR
ONTR
ONTR
ONTRA
AST
ST
ST ENH
ENHA
ENH
ENH
ANCEMENT
NCEMENT
NCEMENT
NCEMENT
NCEMENT
NCEMENT
NCEMENT
NCEMENT
NCEMENT
NCEMENT
Figure 1 illustrates the improvement brought by our method of standardizing the colour contrast across the entire Figure 1 illustrates the improvement brought by our method of standardizing the colour contrast across the entire Figure 1 illustrates the improvement brought by our method of standardizing the colour contrast across the entire Figure 1 illustrates the improvement brought by our method of standardizing the colour contrast across the entire Figure 1 illustrates the improvement brought by our method of standardizing the colour contrast across the entire Figure 1 illustrates the improvement brought by our method of standardizing the colour contrast across the entire Figure 1 illustrates the improvement brought by our method of standardizing the colour contrast across the entire image.
image. image.
The enhanced colour images allow a more comfortable grading and a more obvious diagnostic. The enhanced colour images allow a more comfortable grading and a more obvious diagnostic. The enhanced colour images allow a more comfortable grading and a more obvious diagnostic. The enhanced colour images allow a more comfortable grading and a more obvious diagnostic. The enhanced colour images allow a more comfortable grading and a more obvious diagnostic. The enhanced colour images allow a more comfortable grading and a more obvious diagnostic. The enhanced colour images allow a more comfortable grading and a more obvious diagnostic. The enhanced colour images allow a more comfortable grading and a more obvious diagnostic. The enhanced colour images allow a more comfortable grading and a more obvious diagnostic. The enhanced colour images allow a more comfortable grading and a more obvious diagnostic.
Figure 1. Contrast enhancement by our colour method
Original images Enhanced images
Retinal exam of the same patient with a year of interval acquired in harsh conditions (strong variation of contrast in the original images)
Original images Enhanced images
I
I
I
IM
MA
A
A
A
A
A
A
A
A
AGE
GE
GE
GE
GE
GE
GE
GE
SUPER
SUPERIIMPOS
SUPER
SUPER
SUPER
SUPER
SUPER
SUPER
SUPER
SUPER
SUPER
SUPER
SUPER
SUPER
SUPER
MPOS
MPOSIIT
MPOS
MPOS
MPOS
MPOS
MPOS
MPOS
MPOS
MPOS
MPOS
MPOS
MPOS
MPOS
MPOS
TIION
T
T
T
T
ON
ON
ON
ON
ON
ON
ON
ON
Figure 2 details (with false colours) the two corrections needed for a proper superimposition (position and distortion). Figure 2 details (with false colours) the two corrections needed for a proper superimposition (position and distortion). Figure 2 details (with false colours) the two corrections needed for a proper superimposition (position and distortion). Figure 2 details (with false colours) the two corrections needed for a proper superimposition (position and distortion). Figure 2 details (with false colours) the two corrections needed for a proper superimposition (position and distortion). Figure 2 details (with false colours) the two corrections needed for a proper superimposition (position and distortion). Figure 2 details (with false colours) the two corrections needed for a proper superimposition (position and distortion). Figure 2 details (with false colours) the two corrections needed for a proper superimposition (position and distortion). Figure 2 details (with false colours) the two corrections needed for a proper superimposition (position and distortion). Figure 2 details (with false colours) the two corrections needed for a proper superimposition (position and distortion). Figure 2 details (with false colours) the two corrections needed for a proper superimposition (position and distortion). Figure 2 details (with false colours) the two corrections needed for a proper superimposition (position and distortion). Figure 2 details (with false colours) the two corrections needed for a proper superimposition (position and distortion). Figure 2 details (with false colours) the two corrections needed for a proper superimposition (position and distortion). Figure 2 details (with false colours) the two corrections needed for a proper superimposition (position and distortion). Figure 2 details (with false colours) the two corrections needed for a proper superimposition (position and distortion). Figure 2 details (with false colours) the two corrections needed for a proper superimposition (position and distortion).
Figure 2. Naive and correct superimposition of pairs of eye fundus images Figure 2. Naive and correct superimposition of pairs of eye fundus images Figure 2. Naive and correct superimposition of pairs of eye fundus images Figure 2. Naive and correct superimposition of pairs of eye fundus images Figure 2. Naive and correct superimposition of pairs of eye fundus images Figure 2. Naive and correct superimposition of pairs of eye fundus images Figure 2. Naive and correct superimposition of pairs of eye fundus images Figure 2. Naive and correct superimposition of pairs of eye fundus images Figure 2. Naive and correct superimposition of pairs of eye fundus images Figure 2. Naive and correct superimposition of pairs of eye fundus images Figure 2. Naive and correct superimposition of pairs of eye fundus images Figure 2. Naive and correct superimposition of pairs of eye fundus images Figure 2. Naive and correct superimposition of pairs of eye fundus images
Naive superimposition showing the difference in size and position of the images
(False colours are used)
Images after correct superimposition Images after correct superimposition Images after correct superimposition Images after correct superimposition Images after correct superimposition Images after correct superimposition Images after correct superimposition Images after correct superimposition
(False colours are used) (False colours are used) (False colours are used) (False colours are used)
Figure 3 illustrates the result of the method.
Figure 3. Correct superimpositions of pairs of eye fundus images
RESULTS AND CONCLUSIONS
Our automatic method will allow an easier follow-up of patients with diabetic retinopathy.