Claims Analysis
Total Claims: 9
Claim 1
Independent Claim
184 words
A digital angiography method comprising acquiring a digital angiographic image representing vessels on a background, and automatically performing the following steps in an image processing system: identifying pixels of the digital angiographic image situated on median lines of vessels, referred to as median pixels; examining intensity values at the identified median pixels; identifying vessel segments corresponding to median pixels exhibiting a local variation of intensity with respect to median pixels situated upstream and downstream on the same median line, as potential stenosis zones; subsequent to said identifying potential stenosis zones, extracting pixels at edges of the vessel segments only in said potential stenosis zones; identifying strictures in said vessel segments in said potential stenosis zones based on the locations of extracted pixels at edges of the vessel segment; determining for each potential stenosis zone, upon characteristics including the intensity of median pixels and distance between edges of the vessel segment in said potential stenosis zone at an identified stricture, whether or not the identified stricture is a true stenosis; and providing the digital angiographic image to a display device including markings identifying true stenoses.
Claim 2
Dependent Claim
184 words
A digital angiography method comprising acquiring a digital angiographic image representing vessels on a background, and automatically performing the following steps in an image processing system: identifying pixels of the digital angiographic image situated on median lines of vessels, referred to as median pixels; examining intensity values at the identified median pixels; identifying vessel segments corresponding to median pixels exhibiting a local variation of intensity with respect to median pixels situated upstream and downstream on the same median line, as potential stenosis zones; subsequent to said identifying potential stenosis zones, extracting pixels at edges of the vessel segments only in said potential stenosis zones; identifying strictures in said vessel segments in said potential stenosis zones based on the locations of extracted pixels at edges of the vessel segment; determining for each potential stenosis zone, upon characteristics including the intensity of median pixels and distance between edges of the vessel segment in said potential stenosis zone at an identified stricture, whether or not the identified stricture is a true stenosis; and providing the digital angiographic image to a display device including markings identifying true stenoses.
Claim 3
Dependent Claim
55 words
A method as claimed in claim 1, wherein the step of determining whether or not an identified stricture within a potential stenosis zone is a true stenosis or not is carried out by a neural network which supplies, for each potential stenosis zone, the probability that it contains a true stenosis or a false alarm.
Claim 4
Dependent Claim
41 words
A method as claimed in claim 2, wherein the pixels of the edges of the vessels are extracted by correlating intensity profiles perpendicularly to the axis of the median line of the vessel segments in the potential stenosis zones with crenellations.
Claim 5
Dependent Claim
56 words
A method as claimed in claim 2, wherein the edges of the vessel segments in a potential stenosis zone which correspond to each point of the median line are determined as points of inflection of the corresponding intensity profile, perpendicularly to the median lines of the vessels, by calculation of the derivative of this intensity profile.
Claim 6
Dependent Claim
39 words
A method as claimed in claim 1, wherein the identifying of median pixels is preceded by a step of extracting the background so as to produce the image of the vessels as contrasting objects on a substantially uniform background.
Claim 7
Dependent Claim
39 words
A method as claimed in claim 2, wherein the identifying of median pixels is preceded by a step of extracting the background so as to produce the image of the vessels as contrasting objects on a substantially uniform background.
Claim 8
Dependent Claim
39 words
A method as claimed in claim 3, wherein the identifying of median pixels is preceded by a step of extracting the background so as to produce the image of the vessels as contrasting objects on a substantially uniform background.
Claim 9
Dependent Claim
39 words
A method as claimed in claim 4, wherein the identifying of median pixels is preceded by a step of extracting the background so as to produce the image of the vessels as contrasting objects on a substantially uniform background.