WO2014206467A1 - Procédé de quantification de coloration artérielle - Google Patents

Procédé de quantification de coloration artérielle Download PDF

Info

Publication number
WO2014206467A1
WO2014206467A1 PCT/EP2013/063471 EP2013063471W WO2014206467A1 WO 2014206467 A1 WO2014206467 A1 WO 2014206467A1 EP 2013063471 W EP2013063471 W EP 2013063471W WO 2014206467 A1 WO2014206467 A1 WO 2014206467A1
Authority
WO
WIPO (PCT)
Prior art keywords
window
sequence
ray images
calculating
staining
Prior art date
Application number
PCT/EP2013/063471
Other languages
English (en)
Inventor
Carlo Gatta
Xavier CARRILLO SUÁREZ
Simone Balocco
Simeon PETKOV
David ROTGER MUÑOZ
Oriol RODRIGUEZ LEOR
Eduard FERNANDEZ NOFRERIAS
Josepa MAURI FERRE
Antoni BAYES GENIS
Original Assignee
Centre De Visió Per Computador (Cvc)
Universitat De Barcelona
Fundació Institut D'investigació En Ciències De La Salut Germans Trias I Pujol
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Centre De Visió Per Computador (Cvc), Universitat De Barcelona, Fundació Institut D'investigació En Ciències De La Salut Germans Trias I Pujol filed Critical Centre De Visió Per Computador (Cvc)
Priority to PCT/EP2013/063471 priority Critical patent/WO2014206467A1/fr
Publication of WO2014206467A1 publication Critical patent/WO2014206467A1/fr

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0012—Biomedical image inspection
    • G06T7/0014—Biomedical image inspection using an image reference approach
    • G06T7/0016—Biomedical image inspection using an image reference approach involving temporal comparison
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/48—Diagnostic techniques
    • A61B6/481—Diagnostic techniques involving the use of contrast agents
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/48—Diagnostic techniques
    • A61B6/486—Diagnostic techniques involving generating temporal series of image data
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/50—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment specially adapted for specific body parts; specially adapted for specific clinical applications
    • A61B6/504—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment specially adapted for specific body parts; specially adapted for specific clinical applications for diagnosis of blood vessels, e.g. by angiography
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/50—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment specially adapted for specific body parts; specially adapted for specific clinical applications
    • A61B6/507—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment specially adapted for specific body parts; specially adapted for specific clinical applications for determination of haemodynamic parameters, e.g. perfusion CT
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/52—Devices using data or image processing specially adapted for radiation diagnosis
    • A61B6/5211—Devices using data or image processing specially adapted for radiation diagnosis involving processing of medical diagnostic data
    • A61B6/5217—Devices using data or image processing specially adapted for radiation diagnosis involving processing of medical diagnostic data extracting a diagnostic or physiological parameter from medical diagnostic data
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10016—Video; Image sequence
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10116—X-ray image
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30101—Blood vessel; Artery; Vein; Vascular

Definitions

  • the present invention relates to a method of quantifying arterial staining from a temporal sequence of X-ray images of an area of a human body with at least one artery, each X-ray image being a grey scale image having a two- dimensional structure of pixels common to all the X-ray images, and the sequence of X-ray images having been taken during motion of a contrast liquid injected in the artery.
  • the present invention also relates to a computer program product, to a system, and to a computing system suitable for carrying out such a method.
  • cardiovascular disease is used to refer to all kinds of diseases related to the heart or blood vessels (arteries and veins). This term describes any disease that affects the cardiovascular system, and is commonly used to refer to those related to atherosclerosis (arterial disease). These conditions have causes, mechanisms, and similar treatments.
  • Some medical diagnosis methods are based on the corresponding physician directly observing and interpreting images resulting from medical imaging procedures, such as e.g. Magnetic Resonance Imaging (MRI), Single Photon
  • Emission Computed Tomography SPECT
  • Computed Tomography CT
  • Said images are normally taken during motion of a contrast liquid injected in the artery, said motion of the contrast liquid causing an arterial staining which is visible in the obtained images.
  • Drawbacks of these diagnosis methods may be that quantification of the arterial staining is completely physician-subjective and that the images obtained for said quantification may not be cheap.
  • Computer-assisted image analysis techniques using cheaper images and providing automatic (or semi-automatic) quantification of arterial flow eliminating subjectiveness may thus be desirable. Diverse attempts have been made to tackle such difficult problem from a computer vision perspective.
  • An aspect of this method may thus be that four parameters (k s , a 3 , u s and " ⁇ ) are used to describe the phenomenon under analysis. Details about said sinusoidal function and its parameters may be obtained from [Gil et al., 2008].
  • the present invention aims at further improving these prior art methods and systems.
  • the present invention provides a method of quantifying arterial staining from a temporal sequence of X-ray images of an area of a human body with at least one artery.
  • Each of said X-ray images is a grey scale image having a two-dimensional structure of pixels common to all the X-ray images.
  • the sequence of X-ray images has been taken during motion of a contrast liquid injected in the artery, such that the sequence of X-ray images represents how the motion of the contrast liquid evolves over time.
  • the method comprises dividing the common two-dimensional structure of pixels into a plurality of windows, each window comprising a plurality of contiguous pixels.
  • the method also comprises calculating, for each window and each X- ray image, an average value of the grey levels of the pixels of the X-ray image within the window.
  • the method further comprises calculating, for each window, the parameters ⁇ 4, ⁇ 3 ⁇ 4 and t>A of the following Rayleigh-like function:
  • t is the time in the sequence of X-ray images.
  • This calculation of parameters ⁇ 4, ⁇ u and ⁇ A is performed by applying a non-linear optimization method to approximate the average values of the grey levels calculated for the window with the Rayleigh-like function.
  • the parameters ⁇ 4, " ⁇ and ⁇ A are provided to the non-linear optimization method initialized with suitable values preventing said method to fall in a local minima.
  • An advantage of the proposed method may be that it only requires a sequence of X-ray images, which may be cheaper to obtain than other types of images obtained by other imaging techniques (e.g. MRI, SPECT, CT, etc.).
  • Another advantage of this method may be that only three parameters ( ⁇ , ' > ⁇ and ⁇ A ) are used to describe the phenomenon of arterial staining due to arterial flow.
  • a further advantage of this method may be that the proposed Rayleigh-like function allows modelling the phenomenon under analysis with better accuracy than the prior art methods (such as e.g. [Gil et al., 2008]).
  • a computer program product comprising program instructions for causing a computer to perform said method of quantifying arterial staining.
  • the invention also relates to such a computer program product embodied on a storage medium (for example, a CD-ROM, a DVD, a USB drive, on a computer memory or on a read-only memory) or carried on a carrier signal (for example, on an electrical or optical carrier signal).
  • the present invention provides a system for quantifying arterial staining from a temporal sequence of X-ray images of an area of a human body with at least one artery.
  • Each of said X-ray images is a grey scale image having a two-dimensional structure of pixels common to all the X- ray images.
  • the sequence of X-ray images has been taken during motion of a contrast liquid injected in the artery, such that the sequence of X-ray images represents how the motion of the contrast liquid evolves over time.
  • the system comprises computing means for dividing the common two-dimensional structure of pixels into a plurality of windows, each window comprising a plurality of contiguous pixels.
  • the system also comprises computing means for calculating, for each window and each X-ray image, an average value of the grey levels of the pixels of the X-ray image within the window.
  • the system further comprises computing means for calculating, for each window, the parameters ⁇ ' 3 ⁇ 4 , "A and of the following Rayleigh-like function:
  • t is the time in the sequence of X-ray images.
  • the parameters ⁇ A, ' > ⁇ are provided to the non-linear optimization method initialized with suitable values preventing said method to fall in a local minima.
  • the invention provides a computing system for quantifying arterial staining from a temporal sequence of X-ray images of an area of a human body with at least one artery.
  • Each of said X-ray images is a grey scale image having a two-dimensional structure of pixels common to all the X-ray images.
  • the sequence of X-ray images has been taken during motion of a contrast liquid injected in the artery, such that the sequence of X- ray images represents how the motion of the contrast liquid evolves over time.
  • This computing system comprises a processor and a memory, the memory storing computer executable instructions that, when executed by the processor, cause the computing system to perform the method provided by the present invention.
  • Figure 1 a is a graphic showing three curves representing grey level variations corresponding to a same arterial region experiencing staining due to blood flow in the artery, one of these curves having been obtained by applying a Rayleigh-based fitting used in embodiments of the invention;
  • Figure 1 b is a graphic showing three curves obtained under the same criteria than those used for the curves of Figure 1 a, but, in this particular case, corresponding to another arterial region;
  • Figure 1 c is a graphic showing three curves obtained under the same criteria than those used for the curves of Figures 1 a and 1 b, but, in this particular case, the curves corresponding to a different arterial region;
  • Figure 2 is a graphic showing three curves representing grey level variations corresponding to a same tissue region experiencing staining due to blood perfusion from the artery, one of these curves having been obtained by applying a Gaussian-based fitting used in embodiments of the invention;
  • Figure 3a is a graphic showing a grey-level temporal variation, a Rayleigh- based curve and a Gaussian-based curve corresponding to a same arterial and tissue region experiencing staining;
  • Figure 3b is a graphic obtained under the same criteria than those used to obtain Figure 3a, but, in this particular case, with minimization of the approximation error in L1 norm and considering a pseudo-probability of vessel presence;
  • Figure 4a is a graphic obtained under the same criteria than those used to obtain Figure 3a, but, in this particular case, corresponding to another arterial and tissue region experiencing staining;
  • Figure 4b is a graphic obtained under the same criteria than those used to obtain Figure 4a, but, in this particular case, with minimization of the approximation error in L1 norm and considering a pseudo-probability of vessel presence;
  • Figure 5 is a graphic showing diverse parameters quantifying arterial and tissue staining, which have been obtained according to embodiments of the invention.
  • the calculation of the function and its parameters may be performed, for each defined window (of e.g. N x N pixels), by applying non-linear regression (a well-known non-linear optimization method) to approximate the average values of the grey levels calculated for the window with the function M A ⁇ t).
  • non-linear regression a well-known non-linear optimization method
  • another type of non-linear optimization method could be similarly applied to obtain similar results, such as e.g. gradient descent, conjugate gradient descent, genetic optimization, and so on.
  • the parameters » , 3 ⁇ 4 and ⁇ A may be provided to the non-linear optimization method initialized with the following values:
  • Figure 1 a is a graphic showing three curves 102, 103, 104 obtained from a same sequence of X-ray images and a same window defined on the X-ray images, said window comprising representation of a same arterial region experiencing staining due to blood flow in the artery.
  • the first curve 102 corresponds to an averaged grey level variation
  • the second curve 103 corresponds to a Rayleigh-based fitting used in embodiments of the invention
  • the third curve 104 corresponds to a prior art sinusoidal-based fitting.
  • Figure 1 b is a graphic showing three curves 105, 106, 107 obtained under the same criteria than those used for the curves of Figure 1 a, but, in this particular case, considering another defined window representing another arterial region.
  • Figure 1 c is a graphic showing three curves 108, 109, 1 10 obtained under the same criteria than those used for the curves of Figures 1 a and 1 b, but, in this particular case, considering a different defined window representing a different arterial region.
  • an X-Ray-based arterial flow analysis consists in injecting a contrast agent in the pathological artery while acquiring the X-ray sequence.
  • the contrast liquid may be injected by inserting an intravascular catheter into the arterial system. Then, the contrast liquid flows through the artery and irrigates nearby tissue through capillaries, and is collected by the venous system before returning back to the heart.
  • each physiological media is progressively stained during the transit of the contrast agent and successively returns to the original appearance (grey level).
  • each X-ray image of the sequence is a grey scale image having a two-dimensional structure of pixels common to all the X-ray images.
  • the sequence of X-ray images has been taken during motion of a contrast liquid injected in the artery, such that the sequence of X- ray images represents how the motion of the contrast liquid evolves over time.
  • the common two-dimensional structure of pixels has been divided into a plurality of windows, each window comprising a plurality of contiguous pixels. For each defined window and each X-ray image, an average value of the grey levels of the pixels of the X-ray image within the window has been calculated.
  • Each window for which Figures 1 a, 1 b and 1 c show corresponding curves is one of said defined windows into which the common two-dimensional structure of pixels has been divided.
  • Each graphic illustrated by Figures 1 a, 1 b and 1 c comprises a vertical axis 100 which corresponds to the grey level variation, and a horizontal axis 101 which corresponds to time.
  • the curves 102, 105 and 108 relate to respective averaged grey level variations 100 over time 101 , each one corresponding to different arterial regions. These curves 102, 105 and 108 reach their respective minima (maximum staining) at increasing instants in the sequence.
  • the arterial curves 102, 105 and 108 decreases rapidly in the first part of the sequence and increases gradually and slowly.
  • the contrast liquid motion depends on the local pressure induced by the injection, so that the darkening is faster in proximal parts than in distal parts of the vessel. Moreover, the washing out is slower than the filling since, when the contrast liquid injection is over, the pressure decreases to physiological values. Taking this into account, it has been theoretically estimated and experimentally verified that the grey level variation induced by the arterial staining could be very accurately modelled by using a Rayleigh-like function defined as follows:
  • &A is the time corresponding to the minimum of the curve
  • A is the curve amplitude
  • ⁇ A is a grey level offset.
  • the curves 103, 106 and 109 correspond to respective representations of said function for three different defined windows.
  • Figures 1 a, 1 b and 1 c show the three arterial curves 102, 105 and 108, together with the fitted (or approximated) arterial model M A (t) 103, 106, 109 and also the sinusoidal-based fitting proposed in [Gil et al., 2008] 104, 107, 1 10.
  • the sinusoidal function 104, 107, 1 10 systematically underestimates the amplitude of the grey-level variation.
  • the sinusoidal function 104, 107, 1 10 is always smoother than the experimental curve 102, 105, 108, so that the staining temporal interval is always overestimated.
  • the M A (t) model 103, 106, 109 better estimates the time of maximal staining, with respect to the sinusoidal function 104, 107, 1 10.
  • Figures 1 a, 1 b and 1 c illustrates that the M A (t) model 103, 106, 109 allows better describing the arterial staining while requiring one parameter less than the prior art model proposed in [Gil et al., 2008] 104, 107, 1 10.
  • a parametric map may be obtained for each obtained parameter » , & A and ⁇ of the M A (t) model.
  • These parametric maps may have a structure equivalent to the common two-dimensional structure of pixels divided into the same plurality of windows (of e.g. N x N pixels), which has been used to process the sequence of X-ray images.
  • Different colours may be defined depending on the value of the corresponding parameter, such that each window of the map will show a particular colour or another depending on the value of the resulting value for the parameter in the window.
  • These coloured maps may be very useful for the pertinent physician to identify alterations in an artery.
  • the parametric map of ⁇ A will represent the "background” objects, since it captures the gray-level offset values that does not change in the temporal sequence for each window.
  • the parametric map of "i will represent the intensity of the arterial staining in each window (e.g. dark blue for low intensities, dark red for high intensities, and “intermediate” colors for intensities lying in between said high and low intensities).
  • the parametric map of 1>A will represent the time of maximal arterial staining in each window (e.g. dark blue for short times, light yellow for long times, and "intermediate” colors for times lying in between said long and short times).
  • the application of the non-linear optimization method may comprise minimizing the approximation error in L1 norm:
  • I( x , y-, t) is the average value of the grey levels calculated for the window ( x i v) corresponding to time t in the sequence of X-ray images.
  • minimizing the approximation error may comprise calculating and applying, for each time t in the sequence of X-ray images, a pseudo-probability of vessel presence in the corresponding X-ray image.
  • a pseudo-probability of vessel presence in the corresponding X-ray image is higher the higher the value of said pseudo-probability and is lower the lower the value of said pseudo-probability.
  • This calculation of the pseudo- probability may be performed by applying a method for enhancement of tubular structures. An example of how this calculation may be performed is disclosed in [F ngi, A.F., Niessen, W.J., Vincken, K.L., Viergever, M.A., 1998. Muliscale vessel enhancement filtering, in: MICCAI '98: Proceedings of the First International Conference on Medical Image Computing and Computer- Assisted Intervention, Springer-Verlag, London, UK. pp. 130-137].
  • E A (k A , a A , b A ) ⁇ V(x. y. t ) ⁇ ⁇ I(x, y, t) - m A (t) ⁇ ( 4 )
  • the signal obtained i.e. grey level values over time
  • the signal obtained may be sawtooth shaped (i.e. not smooth).
  • the application of a pseudo-probability of vessel presence as described before may permit obtaining an improved model m A (t) of arterial staining.
  • Figures 3a and 3b illustrate an example of how application of a pseudo- probability of vessel presence may allow obtaining an improved model m A ⁇ t) of arterial staining.
  • a vertical axis 300 corresponding to grey level variation and a horizontal axis 301 corresponding to time are shown.
  • Figure 3a shows an example of sawtooth shaped signal 302 and a m A ⁇ t) curve 304 obtained without application of a pseudo-probability of vessel presence.
  • Figure 3b shows the same sawtooth shaped signal 302 and an improved m A ⁇ t) curve 306 obtained with application of a pseudo-probability of vessel presence calculated in accordance with [Frangi et al., 1998].
  • Figures 4a and 4b illustrate another example of how application of a pseudo- probability of vessel presence may allow obtaining an improved model m A (t) of arterial staining.
  • a vertical axis 400 corresponding to grey level variation and a horizontal axis 401 corresponding to time are shown.
  • Figure 4a shows an example of sawtooth shaped signal 402 and a m A (t) curve 404 obtained without application of a pseudo-probability of vessel presence.
  • Figure 4b shows the same sawtooth shaped signal 402 and an improved m A ⁇ t) curve 406 obtained with application of a pseudo-probability of vessel presence calculated in accordance with [Frangi et al., 1998].
  • the common two-dimensional structure of pixels may comprise a first region of interest (ROI) comprising one or more contiguous windows.
  • This first ROI may correspond to a region of the artery under analysis and may be defined by the corresponding physician.
  • the method may further comprise calculating, for each X-ray image, two further parameters: " ⁇ and &A.
  • the parameter n A is an average of the values ⁇ > ⁇ of the windows comprised in said first ROI.
  • the parameter &A is an average of the values &A of the windows comprised in said first ROI.
  • the method may also comprise calculating an average Rayleigh-like function according to the following formula:
  • the method may further comprise calculating an Arterial Staining Intensity ASI according to the following formula:
  • the method may further comprise calculating an Arterial Staining Duration ASD according to the following formula:
  • ASD i end - 3 ⁇ 4 nit [sec] ( 7 )
  • the method may further comprise calculating an Arterial Time at Maximal Staining ATMS according to the following formula:
  • ATMS arg niin(m J (t)) [sec] ( 1 0 )
  • Figure 5 shows an example of a m A ⁇ t) curve 502 (in black solid line), together with different parameters as defined before.
  • the graphic of Figure 5 shows a vertical axis 500 which corresponds to grey level variation and a horizontal axis 501 which corresponds to time.
  • the reference number 504 corresponds to the Arterial Staining Intensity ASI parameter for the 771 ⁇ ( ⁇ ) curve 502
  • the reference number 505 corresponds to the Arterial Staining Duration ASD for the m A (t) curve 502
  • the reference number 506 corresponds to the Arterial Time at Maximal Staining ATMS for the m A ⁇ t) curve 502.
  • nnit may be defined as the instant in which the m A ⁇ t) curve 502 crosses - ASI/2 while decreasing, whereas d is the moment in which the M A(t) curve 502 crosses - ASI/2 while increasing.
  • This way of measuring the duration of the arterial staining is designed to be as much independent as possible of different ASIs.
  • This parameter can be used to estimate the duration of the wash-out phase.
  • the method may further comprise quantifying tissue staining induced by perfusion of the contrast liquid from the artery, said quantification comprising calculating, for each window, the parameters * , A M, O ⁇ M and of the following Gaussian-like function:
  • t is the time in the sequence of X-ray images.
  • This calculation of this Gaussian-like function is performed by applying a nonlinear optimization method to approximate the average values of the grey levels calculated for the window with the Gaussian-like function.
  • the parameters 1 u , "M, 3 ⁇ 4 and k M are provided to the non-linear optimization method initialized with suitable values preventing said method to fall in a local minima.
  • An aspect of these last embodiments is that a good differentation of the arterial staining and the tissue staining phenomena may be achieved.
  • the arterial staining may be well modeled by applying a Rayleigh-like function, whereas tissue staining may be well modeled by applying a Gaussian-like function.
  • Figure 2 shows a graphic illustrating that said Gaussian-like function m M ⁇ t) allows obtaining better results than the prior art method disclosed in [Gil et al., 2008].
  • This graphic comprises a vertical axis 100 corresponding to grey level variation and a horizontal axis 101 corresponding to time.
  • three curves 200, 201 , 202 are shown, which have been obtained from a same sequence of X-ray images and a same window defined on the X-ray images, said window comprising representation of a same tissue region experiencing staining due to perfusion.
  • the curve 200 corresponds to an averaged grey level variation
  • the curve 201 corresponds to a Gaussian-based fitting used in embodiments of the invention ( «3 ⁇ 4 (*))
  • the curve 202 corresponds to a prior art sinusoidal-based fitting.
  • Figure 2 just shows a particular result of the diverse experiments performed to confirm that the proposed Gaussian-like function m M ⁇ t) 201 allows obtain a better model of tissue staining than the prior art sinusoidal function 202 disclosed in [Gil et al., 2008].
  • the calculation of the function m M (t) and its parameters ⁇ ' w , «M, m and k M ) may be performed, for each defined window (of e.g. N x N pixels), by applying non-linear regression (a well-known non-linear optimization method) to approximate the average values of the grey levels calculated for the window with the function m M (t) .
  • non-linear regression a well-known non-linear optimization method
  • another type of non-linear optimization method could be similarly applied to obtain similar results, such as e.g. gradient descent, conjugate gradient descent, genetic optimization, and so on.
  • the parameters *M , «M, M and k M may be provided to the non-linear optimization method initialized with the following values: where I( x i V, t) is the average value of the grey levels calculated for the window ( x > y) corresponding to time t in the sequence of X-ray images.
  • I( x i V, t) is the average value of the grey levels calculated for the window ( x > y) corresponding to time t in the sequence of X-ray images.
  • k M have been theoretically estimated and experimentally verified as suitable to prevent the non-linear optimization method to fall in a local minima.
  • other initial values could be used to obtain similar results, such as e.g. extensive search of optimal initial values by means of bootstrapping.
  • a parametric map may be obtained for each obtained parameter *M , « ⁇ , JM and & of the m M ⁇ t) model. Equal or similar principles to those applied to obtain the parameters ⁇ , ⁇ A and ⁇ A of the m A (t) model may also be applied to obtain the parametric maps of the parameters * , « , VM and k M of the M M if) model.
  • the parametric map of k&i will represent the "background" objects.
  • the parametric map of « will represent the intensity of the tissue staining in each defined window.
  • the parametric map of *M will represent the time of maximal tissue staining in each defined window.
  • the map of &M will represent a value proportional to the temporal extent of the tissue perfusion.
  • the application of the non-linear optimization method may comprise minimizing the approximation error in L1 norm: wherein I( x , is the average value of the grey levels calculated for the window ( x i y) corresponding to time t in the sequence of X-ray images.
  • minimizing the approximation error may comprise calculating and applying, for each time t in the sequence of X-ray images, a pseudo-probability of vessel presence in the corresponding X-ray image, such that the weight of each value ⁇ I ?/ ⁇ ) - M M (*)
  • the calculation of said pseudo-probability is performed by applying a method for enhancement of tubular structures.
  • this calculation may be performed by applying the method disclosed in [Frangi et al., 1998], in a similar way than for the function m A ⁇ t) .
  • Figures 3a and 3b also show an example of corresponding curves m M (t) 303, 305 with and without application of pseudo-probability of vessel presence respectively.
  • Figures 4a and 4b show another example of corresponding curves m M (t) 403, 405 with and without application of pseudo-probability of vessel presence respectively. These figures illustrate that applying said pseudo-probability of vessel presence may generally improve the modelling of tissue staining m M (t) .
  • the common two-dimensional structure of pixels may comprise a second ROI comprising one or more contiguous windows.
  • This second ROI may correspond to a region of tissue under analysis and may be defined by the corresponding physician.
  • the method may further comprise calculating, for each X-ray image, three further parameters: %, t M and 1 ⁇ 2.
  • the parameter " is an average of the values a M of the windows comprised in said second ROI.
  • the parameter t M is an average of the values * of the windows comprised in said second ROI.
  • the parameter M is an average of the values ⁇ ⁇ of the windows comprised in said second ROI.
  • the method may further comprise calculating an average Gaussian-like function according to the following formula:
  • the method may further comprise calculating a Tissue Staining Intensity TSI according to the following formula:
  • TSD *end - *init M ⁇ 1 5 >
  • the method may further comprise calculating a Tissue Time at Maximal Staining TTMS according to the following formula:
  • TTMS arg min(m M (t)) [sec ( 1 8 )
  • Figure 5 shows an example of a m M (t) curve 503, together with different parameters as defined before.
  • the reference number 507 corresponds to the Tissue Staining Intensity TSI parameter for the m M (t) curve 503
  • the reference number 508 corresponds to the Tissue Staining Duration TSD for the m M (i) curve 503
  • the reference number 509 corresponds to the Tissue Time at Maximal Staining TTMS for theTMM(£) curve 503.
  • Some embodiments of the method may further comprise calculating a Staining Temporal Delay (STD) between the arterial and the tissue staining according to the following formula:
  • FIG. 5 also illustrates graphically this last parameter STD 510 for the curves m A ⁇ t) and m M ⁇ t) .
  • This parameter STD 510 allows quantifying this characteristic.
  • the embodiments of the invention described with reference to the drawings comprise computer apparatus and processes performed in computer apparatus, the invention also extends to computer programs, particularly computer programs on or in a carrier, adapted for putting the invention into practice.
  • the program may be in the form of source code, object code, a code intermediate source and object code such as in partially compiled form, or in any other form suitable for use in the implementation of the processes according to the invention.
  • the carrier may be any entity or device capable of carrying the program.
  • the carrier may comprise a storage medium, such as a ROM, for example a CD ROM or a semiconductor ROM, or a magnetic recording medium, for example a floppy disc or hard disk.
  • the carrier may be a transmissible carrier such as an electrical or optical signal, which may be conveyed via electrical or optical cable or by radio or other means.
  • the carrier may be constituted by such cable or other device or means.
  • the carrier may be an integrated circuit in which the program is embedded, the integrated circuit being adapted for performing, or for use in the performance of, the relevant processes.

Landscapes

  • Health & Medical Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Medical Informatics (AREA)
  • Public Health (AREA)
  • General Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Physics & Mathematics (AREA)
  • Pathology (AREA)
  • Radiology & Medical Imaging (AREA)
  • Optics & Photonics (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • High Energy & Nuclear Physics (AREA)
  • Biophysics (AREA)
  • Veterinary Medicine (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Dentistry (AREA)
  • Vascular Medicine (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Epidemiology (AREA)
  • Primary Health Care (AREA)
  • Quality & Reliability (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Physiology (AREA)
  • Apparatus For Radiation Diagnosis (AREA)

Abstract

La présente invention concerne un procédé permettant de quantifier la coloration artérielle à partir d'une séquence temporelle d'images à rayons X d'une partie du corps humain comprenant au moins une artère, chaque image à rayons X étant une image à échelle de gris avec une structure de pixels en deux dimensions commune à toutes les images à rayons X, et la séquence d'images à rayons X ayant été prise lors du déplacement d'un liquide de contraste injecté dans l'artère. Le procédé consiste en les étapes suivantes : la division de la structure commune en deux dimensions en une pluralité de fenêtres, chaque fenêtre comprenant une pluralité de pixels contigus ; le calcul, pour chaque fenêtre et chaque image à rayons X, d'une valeur moyenne des niveaux de gris des pixels de l'image à rayons X au sein de la fenêtre ; et le calcul, pour chaque fenêtre, des paramètres d'une fonction de type Raleigh (103), ledit calcul étant effectué en appliquant un procédé d'optimisation non linéaire afin d'évaluer approximativement les valeurs moyennes des niveaux de gris calculés pour la fenêtre avec la fonction de type Raleigh (103).
PCT/EP2013/063471 2013-06-27 2013-06-27 Procédé de quantification de coloration artérielle WO2014206467A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/EP2013/063471 WO2014206467A1 (fr) 2013-06-27 2013-06-27 Procédé de quantification de coloration artérielle

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/EP2013/063471 WO2014206467A1 (fr) 2013-06-27 2013-06-27 Procédé de quantification de coloration artérielle

Publications (1)

Publication Number Publication Date
WO2014206467A1 true WO2014206467A1 (fr) 2014-12-31

Family

ID=48703497

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/EP2013/063471 WO2014206467A1 (fr) 2013-06-27 2013-06-27 Procédé de quantification de coloration artérielle

Country Status (1)

Country Link
WO (1) WO2014206467A1 (fr)

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
CARLO GATTA ET AL: "Toward Robust Myocardial Blush Grade Estimation in Contrast Angiography", 10 June 2009, PATTERN RECOGNITION AND IMAGE ANALYSIS, SPRINGER BERLIN HEIDELBERG, BERLIN, HEIDELBERG, PAGE(S) 249 - 256, ISBN: 978-3-642-02171-8, XP019119075 *
FRANGI, A.F.; NIESSEN, WJ.; VINCKEN, K.L.; VIERGEVER, M.A.: "MICCAI '98: Proceedings of the First International Conference on Medical Image Computing and Computer- Assisted Intervention", 1998, SPRINGER-VERLAG, article "Muliscale vessel enhancement filtering", pages: 130 - 137
GIL, D.; RODRIGUEZ-LEOR, O.; RADEVA, P.; MAURI, J.: "Myocardial perfusion characterization from contrast angiography spectral distribution", IEEE TRANS. MED. IMAGING, vol. 27, 2008, pages 641 - 646
GIL, D.; RODRIGUEZ-LEOR, O.; RADEVA, P.; MAURI, J.: "Myocardial perfusion characterization from contrast angiography spectral distribution", IEEE TRANS. MED. IMAGING, vol. 27, 2008, pages 641 - 646, XP011202411 *
SELVATHI D ET AL: "Statistical modeling for the characterization of atheromatous plaque in Intravascular Ultrasound images", DEVICES, CIRCUITS AND SYSTEMS (ICDCS), 2012 INTERNATIONAL CONFERENCE ON, IEEE, 15 March 2012 (2012-03-15), pages 320 - 324, XP032192138, ISBN: 978-1-4577-1545-7, DOI: 10.1109/ICDCSYST.2012.6188729 *

Similar Documents

Publication Publication Date Title
JP7090546B2 (ja) 灌流デジタルサブトラクション血管造影
US8577441B2 (en) System and method for image based physiological monitoring of cardiovascular function
US10275878B2 (en) Method for the quantification of the presence of fats in a region of the heart
US20110150309A1 (en) Method and system for managing imaging data, and associated devices and compounds
CN107280696B (zh) 确定描述侧支内的血液流动的侧支信息的方法和拍摄装置
CN107004262B (zh) 基于设备的运动补偿的数字减影血管造影系统和方法
CN108348206B (zh) 用于无创血流储备分数(ffr)的侧支流建模
JP5658686B2 (ja) 貫壁性のかん流の勾配の画像分析
KR102402628B1 (ko) 인간의 심장 및 심방을 모델링하는 방법 및 시스템
JP2012512729A (ja) 急性脳卒中の組織状態をマッピングする方法及びシステム
US20020168095A1 (en) Analysis of successive data sets
JP2013059620A (ja) 血管狭窄の視覚化及びナビゲーションのシステム及び方法
CN109377481B (zh) 图像质量评价方法、装置、计算机设备和存储介质
US20220319004A1 (en) Automatic vessel analysis from 2d images
KR101611488B1 (ko) 의료 영상에서 아티팩트와 병변을 구분하는 방법
WO2014206467A1 (fr) Procédé de quantification de coloration artérielle
JP2632501B2 (ja) 内臓器官の複数画像フレームを使用した生体内における内臓器官輪郭抽出方法
US11361430B2 (en) System and method for medical imaging
RU2815435C2 (ru) Способ оконтуривания аномальных зон на результатах перкуссионного томографического сканирования
EP4197442A1 (fr) Analyse de perfusion périphérique
WO2009077978A1 (fr) Procédé d'acquisition d'images tridimensionnelles du myocarde, en particulier d'un rougissement du myocarde
Saporito et al. Automatic indicator dilution curve extraction in dynamic-contrast enhanced imaging using spectral clustering
Silva et al. Exploring different parameters to assess left ventricle global and regional functional analysis from coronary ct angiography
JP2023547373A (ja) 血管流体流速を決定する方法
Williams et al. Tissue perfusion diagnostic classification using a spatio-temporal analysis of contrast ultrasound image sequences

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 13732472

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 13732472

Country of ref document: EP

Kind code of ref document: A1