US2017039689A1PendingUtilityA1
Systems and methods for enhancement of retinal images
Est. expiryOct 22, 2033(~7.3 yrs left)· nominal 20-yr term from priority
G06V 40/193G06T 3/40G06T 2207/20032A61B 3/0025G06T 5/20G06T 5/008G06F 17/30268A61B 3/14A61B 3/12G06T 2207/20016G06T 2207/30041G06V 2201/03G06V 40/18G06V 40/14G16Z 99/00G06V 10/758G06V 10/50G06V 10/44G06V 10/267G16H 30/40G16H 50/20G06F 16/51G06T 7/0012G06T 7/0016G06T 2207/20036G06T 2207/10024G06F 16/583G16H 30/20G06T 2207/30096G06F 16/5866G06T 7/0014G06T 2207/30104G06T 2207/30168G06T 5/94G06T 3/14G06T 3/18
54
PatentIndex Score
0
Cited by
0
References
0
Claims
Abstract
Embodiments disclose systems and methods that aid in screening, diagnosis and/or monitoring of medical conditions. The systems and methods may allow, for example, for automated identification and localization of lesions and other anatomical structures from medical data obtained from medical imaging devices, computation of image-based biomarkers including quantification of dynamics of lesions, and/or integration with telemedicine services, programs, or software.
Claims
exact text as granted — not AI-modified1 . A computing system for enhancing a retinal image, the computing system comprising:
one or more hardware computer processors; and one or more storage devices configured to store software instructions configured for execution by the one or more hardware computer processors in order to cause the computing system to:
access a medical retinal image I for enhancement, the medical retinal image related to a subject;
estimate the background I Back, of the image at single or multiple scales;
scale the intensity I(x,y) at a first pixel location in the medical retinal image adaptively based on the intensity at a same position in the background image I Back, (x,y) for generating an enhanced image.
2 . The computation system of claim 1 , wherein the background is estimated by a median filtered image with a median computed over a geometric shape.
3 . The computation system of claim 1 , wherein the background is estimated at more than one scales by progressively changing the size of a geometric shape for a filter used to compute the background.
4 . The computation system of claim 1 , wherein the computing system is further configured to process the retinal image using a noise-removing filter.
5 . The computing system of claim 1 , wherein if the intensity I(x,y) is lower than intensity at the same position in the background I Back, (x,y), then the enhanced image pixel intensity at location (x,y) is set to a value around a middle of a minimum and a maximum intensity value for the medical retinal image C mid scaled by a ratio of intensity at medical retinal image to intensity in the background image as expressed by
C
mid
·
I
(
x
,
y
)
(
x
,
y
)
.
6 . The computing system of claim 1 , wherein if the intensity I(x,y) is not lower than intensity at the same position in the background image I Back, (x,y), then the enhanced image pixel intensity at location (x,y) is set to a sum of around the middle of the minimum and the maximum intensity value for the medical retinal image, C mid , and (C mid −1) scaled by a ratio of a difference of intensity of the background image from intensity of the medical retinal original image to a difference of intensity of the background image from a maximum possible intensity value C max , expressed as
C
mid
+
(
(
C
mid
-
1
)
·
I
(
x
,
y
)
-
(
x
,
y
)
C
max
-
(
x
,
y
)
)
.
7 . The computing system of claim 1 , wherein the computing system is further configured to automatically identify one or more abnormalities or anatomical structures in the retinal image.
8 . The computing system of claim 1 , wherein the computing system is further configured to automatically analyze a medical condition of the subject.
9 . The computing system of claim 1 , wherein a filter is used to compute the background using a geometric shape that is one or more of: a circle, a square, or a regular polygon.
10 . The computing system of claim 1 , wherein the computing system is further configured to automatically perform at least one of image registration, lesion localization, screening, quality assessment, interest region detection, or descriptor computation.
11 . The computing system of claim 1 , wherein the retinal image is a single or multidimensional image that has been captured using an imaging method from one or more of: color retinal imaging, fluorescein angiography, adaptive optics-based imaging, optical coherence tomography, hyperspectral imaging, scanning laser ophthalmoscopy, wide-field imaging, or ultra-wide-field imaging.
12 - 22 . (canceled)
23 . Non-transitory computer storage that stores executable program instructions that, when executed by one or more computing devices, configure the one or more computing devices to perform operations comprising:
accessing a medical retinal image for enhancement, the medical retinal image related to a subject; estimating the background I Back, of the image at single or multiple scales; scaling the intensity I(x,y) at a first pixel location in the medical retinal image adaptively based on the intensity at a same position in the background image I Back, (x,y) for generating an enhanced image.
24 . The non-transitory computer storage of claim 23 , wherein the background is estimated by a median filtered image with a median computed over a geometric shape.
25 . The non-transitory computer storage of claim 23 , wherein the background is estimated at more than one scales by progressively changing the size of a geometric shape for a filter used to compute the background.
26 . The non-transitory computer storage of claim 23 , further comprising processing the retinal image using a noise-removing filter.
27 . The non-transitory computer storage of claim 23 , wherein if the intensity I(x,y) is lower than intensity at the same position in the background I Back, (x,y), then the enhanced image pixel intensity at location (x,y) is set to a value around a middle of a minimum and a maximum intensity value for the medical retinal image C mid scaled by a ratio of intensity at medical retinal image to intensity in the background image as expressed by
C
mid
·
I
(
x
,
y
)
(
x
,
y
)
.
28 . The non-transitory computer storage of claim 23 , wherein if the intensity I(x,y) is not lower than intensity at the same position in the background image I Back, (x,y), then the enhanced image pixel intensity at location (x,y) is set to a sum of around the middle of the minimum and the maximum intensity value for the medical retinal image, C mid , and (C mid −1) scaled by a ratio of a difference of intensity of the background image from intensity of the medical retinal original image to a difference of intensity of the background image from a maximum possible intensity value C max , expressed as
C
mid
+
(
(
C
mid
-
1
)
·
I
(
x
,
y
)
-
(
x
,
y
)
C
max
-
(
x
,
y
)
)
.
29 . The non-transitory computer storage of claim 23 , further comprising automatically identifying one or more abnormalities or anatomical structures in the retinal image.
30 . The non-transitory computer storage of claim 23 , further comprising automatically analyzing a medical condition of the subject.
31 . The non-transitory computer storage of claim 23 , wherein a filter is used to compute the background using a geometric shape that is one or more of: a circle, a square, or a regular polygon.
32 . The non-transitory computer storage of claim 23 , further comprising automatically performing at least one of image registration, lesion localization, screening, quality assessment, interest region detection, or descriptor computation.
33 . The non-transitory computer storage of claim 23 , wherein the retinal image is a single or multidimensional image that has been captured using an imaging method from one or more of: color retinal imaging, fluorescein angiography, adaptive optics-based imaging, optical coherence tomography, hyperspectral imaging, scanning laser ophthalmoscopy, wide-field imaging, or ultra-wide-field imaging.Cited by (0)
No later patents cite this yet.
References (0)
No backward citations on record.