Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
CHAPTER 5
EFFICIENT APPROACHES FOR QUANTIFICATION OF AORTIC
REGURGITATION USING PROXIMAL ISOVELOCITY SURFACE AREA
PROCESS
5.1. Introduction
Aortic Regurgitation is also known as Aortic Insufficiency (AI). It indicates the inability
of the aortic valve. During diastole, the aortic valve allows blood flow in the reverse direction
from aorta into the left ventricle. Regurgitation is due to incompetence of the aortic valve or any
disorder of the valvular apparatus (e.g., leaflets, annulus of the aorta) ensuing in diastolic flow of
blood into the left ventricular chamber. AR force an accurate amount overload to the left
ventricle (LV) which results in dilation, eccentric hypertrophy and finally loss of function. The
integral part of aorta is aortic valve, which is tubular-like structure. During systole and diastole,
the valve apparatus includes three distinct leaflets with definitive passive motion. Due to
relatively high systolic and diastolic pressure the aortic valve is challenged with a relatively high
mechanical stress and, in terms of morphology, a consideration is to be made of the relation to
origin of the coronary arteries. During systole and diastole, the movement of the tube is
insufficient in all directions [18]. Echocardiography which is followed by MR imaging (or
contrast aortography) produces all valuable parameters. Additionally the key considerations are
the diameter of the AV orifice and of the neighboring ascending aorta segment diameter. Of late
in clinical cardiology an all-round estimation of Valvular Regurgitation is an essential objective
to be carried out by the cardiac surgery.
Evaluation of the severity of regurgitation is supreme to clinical decision making in
patients with AR., because patients with severe Aortic Regurgitation often need surgical
treatment. Semi-quantitative position of Aortic Regurgitation with color and spectral Doppler
echo or with angiography is generally used. But both the processes are delayed by some specific
restrictions or reasons. Invasive and Noninvasive quantitative evaluation of Regurgitant Volume
and Regurgitant Fraction are obtainable, but Regurgitant Volume and Regurgitation Fraction
depend on loading circumstances. It was proposed in recent times that noninvasive calculation of
aortic ERO was a measure of lesion severity in AR [63]. The EROA is a basic descriptor of
Aortic Regurgitation that confirms the effect of AR on the LV and offers information additional
to the volume overload measurements like Regurgitation Fraction. But the measurement of the
EROA by quantitative Doppler Echocardiography is not always possible to attain high degree of
reliability; a grouping of methods is desirable (very much suggested). In order to measure the
severity of valvular and congenital heart diseases the clinicians have great interest in PISA
methods only. Depending on the conservation of mass, for evaluating (or scheming) the effective
orifice area in Valvular Regurgitation the PISA method has been faithfully adopted [64]. In table
5.1, Echocardiographic and Doppler parameters used in the evaluation of Aortic Regurgitation
severity are described.
Table 5.1 Echocardiographic and Doppler parameters used to evaluate AR severity:
Utility, Advantages, and Limitations
Utility/Advantages Limitations
Left Ventricle size Enlargement sensitive for
chronic significant AR.
Normal size virtually excludes
significant chronic AR.
Enlargement seen in other
conditions. Normal in acute
significant AR.
Aortic cusps alterations Simple, usually abnormal in
severe AR; Flail valve denotes
severe AR.
Poor accuracy
Doppler parameters
Jet width or jet cross-
sectional area in
*LVOT- Color Flow
Simple, very sensitive, quick
screen for AR.
Expands unpredictably below
the orifice. Inaccurate for
eccentric jets.
Vena Contracta Width Simple, quantitative to identify
mild or severe AR.
Not useful for multiple AR
jets.
PISA method Quantification both EROA and
Rvol.
Feasibility limited by aortic
valve calcifications. Not valid
for multiple jets, less accurate
in eccentric jets.
Flow quantification-PW Quantitative, valid with
multiple jets and eccentric jets.
Provides both lesion severity
i.e. EROA, RF and volume
overload, Rvol.
Not valid for combined MR
and AR, unless pulmonic site
is used.
Jet density-CW Simple. Incomplete jet
compatible with mild AR.
Qualitative. Overlap between
moderate and severe AR.
Jet deceleration rate
(*PHT)- CW
Simple Qualitative; affected by
changes in LV and aortic
diastolic pressures.
*LVOT – Left Ventricle Outflow Tract, *PHT – Pressure Half Time
The Proximal Isovelocity Surface Area method [60, 26] depends on the continuity
principle and assumes that blood flow converging in the direction of a flat orifice forms
hemispherical isovelocity shells. It has been proved that the PISA method is precise and
reproducible. This method is frequently applied in medical science because the proximal
convergence method can be easily visualized and it is the only probable method currently easy to
have it. Despite the hypothetical advantages in comparison to Valvular Regurgitation the PISA is
seldom used routinely for the evaluation of AR severity. The chief purpose of the present
research is to give an efficient method based on image processing methods. And these can
exactly evaluate the EROA using Doppler Echocardiography image with the aid of PISA.
Considerable interest is shown in the PISA method to estimate the severity of valvular and
congenital heart diseases. In this Part I and part II contain the details of the methodology for
quantification of AR that has been carried out.
In Part I, in preprocessing stage the RGB color Doppler Echocardiography image has
been subjected to Wiener filtering. Then the filtered image has been quantized with the aid of
color quantization using NBS / ISCC color space which makes the image more precise. Besides
those the PISA method is used for calculating the quantitative parameters like VC, R vol, RF,
Effective regurgitant orifice (ERO) and mostly AR. The PFC method is pursued to quantify
Aortic Regurgitation by analyzing the converging flow field proximal to estimate the mildness
severity and eccentricity of an AR lesion.
Similarly in part II, an efficient method for quantifying the EROA in AR using clustering
based image segmentation processing methods on Doppler Echocardiographic image and helped
by PISA method has been presented. Considerable attention has been received by PISA method
by clinicians for assessing the severity of valvular and congenital heart diseases. In the
preprocessing stage the Gaussian filter is used to reduce the noise in color Doppler
echocardiographic image. Accordingly it has been improved with the support of image contrast
enhancement by using Contrast-Limited Adaptive Histogram Equalization (CLAHE). As a
sequel, the image is subjected to image segmentation based on Fuzzy k-means clustering to make
the quantification more precise. Clustering is a method for grouping an image into units that are
reliable to one or more characteristics. Besides those the PISA method is used for computing the
quantitative parameters like VC, R vol, RF, ERO and mostly AR. The PFC method is pursued to
quantify VR with the help of analyzing the converging flow field proximal to estimate the
mildness severity and eccentricity of an AR lesion. This research also offers a survey of
qualitative and quantitative parameters for rating AR severity, utility, advantages and limitations
of Echocardiography along with Doppler parameters which are made use of in the estimation of
MR severity.
5.2. Technical Terms
Aortic Regurgitation – It indicates the inability of the aortic valve. During diastole, the
aortic valve allows blood flow in the reverse direction from aorta into the left ventricle.
Wiener Filtering – It is one of most significant techniques. This filtering is used for
removing the blur in images because of linear motion or unfocussed optics.
Color Quantization – color image quantization is a method that reduces the number of
distinct colors in an image so that the new image will looks like the original image.
NBS/ISCC Color Space – ISCC–NBS System of Color Designation is a system used to
name the colors. The colors are named based on a set of 12 basic or fundamental color terms and
a few set of adjective modifiers.
Gaussian Filtering – It is one of the filters whose impulse response is a Gaussian
function. The main intention of this filter is to provide no overshoot to a step function input
while reducing the rise and fall time.
Image Enhancement – In the image enhancement, the perception of information in
images for human viewers are enhanced and it offers `better' input for further image processing
methods.
Clustering – It is one of the most significant unconfirmed learning problems; it deals with
finding a structure in a set of unlabeled data, since other problems are of this kind.
Fuzzy k Means – Fuzzy K-Means is also known as Fuzzy C-Means. It is an extension of
K-Means, the popular simple clustering method .The K-Means discovers hard clusters, where
Fuzzy K-Means is more statistically formalized method than K-Means and discovers soft
clusters. Hard cluster denotes a point which belongs to only one cluster, but the soft cluster
denotes a particular point which belongs to more than one cluster with certain probability.
5.3. Review of Related Researches
A brief review of recent researches related to quantification of Aortic Regurgitation is
described below.
Thomas Wittlinger et al. [65] have presented a study that the MRI was related to both
angiography and echocardiography in estimating the RV and AR in patients. At 1.5 T, forty
patients were taken under examination. The regurgitant jet was located with the aid of a gradient-
echo sequence. To compute the left ventricular function, cine measurements were accomplished.
For flow estimation, a velocity-encoded breath-hold phase-difference magnetic resonance
sequence was made familiar. The severity of AR is calculated by making use of MRI accepted
with that of angiography in 28 of 40 patients, and with the Echocardiography result in 80%. As a
result the correlation between calculated stroke volume by the two methods that is magnetic
resonance cine and flow measurements was very excellent (r > 0.9).
Quantification of aortic regurgitation using Echocardiography is still stimulating. In spite
of rheological characteristics which, have been granted by Chen Li et al. [33], a novel
echocardiographic method called Vector Flow Mapping (VFM) directly measures blood flow
volume. They have in turn, projected to estimate the accuracy of VFM in quantifying chronic
AR. Therefore for AR quantification the clinicians feel that a highly reproducible parameter with
good precision has been discovered RegR evaluated by VFM-as a novel Doppler method that
allows quantitative analysis of FV despite the presence of turbulent flow.
Leopoldo Pérez de Isla et al. [66] have proposed detailed study so that the measurements
of LVOT area are arrived at by using 3D-echo which is highly reliable compared to those which
are made by using 2D-echo. Both the methods i.e. 2-D echo and 3-D echo are used to measure
the LVOT area and the circularity index was measured using 3D-echo exclusively. Moreover the
severity of valvular aortic stenosis was classified using both 2D-echo and 3D-echo. Therefore for
assessing the LVOT area, the 3D-echo may be a superior technique. The additional advantage is
that the LVOT is elliptical in shape, but not related to its circularity in size. Also for
distinguishing the severity of valvular aortic stenosis, the 3D-echo could be supportive.
Anne-catherine Pouleur, MD et al. [67] have presented a study to appraise the correctness
of multidetector CT compared with cine MR imaging, TTE and TEE in the measurement of
aortic valve area (AVA) of patients undergoing cardiac surgery. After examining 48 patients (
out of which 33 are men and 15 are women under ages 62 ±13), the accuracy of multidetector
CT for detection of aortic stenosis was compared with that of TTE. The results demonstrated
that, the multidetector CT could be an alternative to TTE in patients with poor acoustic windows.
The multidetector CT can be used to measure the AVA and detect aortic stenosis at the time of
noninvasive coronary imaging with accuracy similar to that of TTE and MRI.
An extensive attention is shown to Non-invasive management of AV disease. Actually a
lot of current publications have formerly reported its use in clinical practice. The major issue is
to obtain considerate pathophysiological processes and, most significantly, extensive
experimental activity. In addition to testing of various animal models, technical and material
features are also being intensively investigated. Clinicians are dubious about the applicability
and durability of this positive improvement whether it can be equated with the standard of
present cardiac surgery. Certainly it is justified that the full use of certain models as a tentative
measure to help to progress the circulatory status, may not permit the safe surgery. A tiny
analysis of the above mentioned issue has been granted by Sochman and Peregrin [18].
5.4. The Proposed Methodology – Part I
A major goal of the clinical cardiology is quantifying the severity of AR and it is
extremely old clinical decision making. Very often the screening for the existence of AR is
preceded with the help of color Doppler flow mapping. Several echocardiographic methods have
been published to improve the quantification of valvular incompetence. Although the primary
method used for estimating AR severity has been found to be less precise than the latter
counterparts and therefore PFC method using color Doppler has been recognized as a precise
(exact) and consistent quantitative approach. For the quantification of aortic regurgitant, here we
present an efficient mode by uniting image processing method which exactly quantifies the
EROA which strengthen the PISA method to estimate the severity of AR lesion. More than this
here we deal with the consistency of the PISA method for computing ERO of AR. The present
approach mainly includes two modules:
i. Preprocessing
ii. Quantification using Proximal Isovelocity Surface Area (PISA)
5.4.1. Preprocessing
Initially in the preprocessing step the color Doppler Echocardiography AR image is
subjected to Wiener filtering, that reduces the noise quantity in the image in comparison with an
estimation of the preferred noiseless image.
Wiener Filtering: The purpose of this filter is to reduce the amount of noise present in a
signal. The Wiener filter is the Mean Squared Error optimal stationary linear filter for images
despoiled by additive noise and blurring. In order to calculate the Wiener filter, the signal and
noise processes are assumed to be second order stationary. Presuming stationary the nature of the
involved signals, the average squared distance between filter output and a preferred signal is
minimized by calculating the Wiener filter coefficients that can be made perfect in the frequency
domain producing easily:
))(/)(()( fPfPfW YYDY …(5.1)
Where )( fD is the desired signal, )()()( fYfWfS
is the Wiener filter output, )( fY the
Wiener filter input and )( fPYY , )( fPDY are the power spectrum of )( fY and the cross power
spectrum of )( fY , )( fD respectively.
The AR color flow Doppler TEE image in apical view showing Proximal Flow
Convergence is shown in figure 5.1(a), figure 5.1(b) shows its filtered output.
Figure 5.1 a) AR Color flow Doppler image (TEE) in apical view showing Proximal FC and b) its
filtered output
[Image Courtesy: Journal of American College of Cardiology, Clinical Studies article]
After that the filtered color Doppler Echocardiographic image is subjected to color
quantization. Then the color quantization fetches down the numerous colors used in an image,
usually with the intention that the new image has to be as visually similar as possible to the
original image.
Color Quantization: Color quantization is the technique used for minimizing the
number of colors in a digital image by changing them with a specific color selected from a
palette [68]. It is broadly used nowadays because it decreases the work load of huge image data
on storage and transmission bandwidth in several multimedia applications. Accordingly the main
inducement of color image quantization is for mapping the set of colors in the original color
image to a smaller set of colors in the quantized image so that this mapping reduces the variation
between the original and the quantized images, as mentioned earlier [45]. With the help of
NBS/ISCC color space color quantization is carried out as our contribution..
Imagine that the color Doppler Echocardiographic image as iI where, 10 ini .
Consequently, to accomplish the quantization, the RGB color space values of the image are
represented by vectors, that is,iRI ,
iGI andiBI correspondingly with size NM . Where NM is
resample into RR images (generally R =192). Therefore, from the color space imagesiRI ,
iGI
andiBI , the sampled image
iRS , iGS and
iBS are obtained. After that, with the help ofiRS ,
iGS ,
iBS and Color Look Up Table (CLUT), the color quantization is implemented. The matrix
representation of the CLUT can be specified as
111
222
111
000
nnn
lt
BGR
BGR
BGR
BGR
C
…(5.2)
From the above matrix representation, the CLUT has n different probable colors created
by various, xR , yG and zB transformations. After that by manipulating the Euclidean distance
between every pixel value of the re-sampled color space images and the ltC value, the database
images are color quantized. As a result the quantized image IQ is obtained. Then the quantized
image IQ consists of quantized RGB color space values RQ , GQ and BQ which are formulated
as
222 )),(()),(()),((min),( jBjGjRR BbaSGbaSRbaSbaQ …(5.3)
where, 10 nj , 10 ra and 10 rb . This is the same for the quantization
of other color space images GQ and BQ also (i.e. BGR QQQ ).
Figure 5.2 Color Quantization Outputs a) Quantized view b) Binary output c)
Segmented output
5.4.2. Quantification using Proximal Isovelocity Surface Area
After preprocessing stage the color Doppler Echocardiographic image is subjected to
efficient quantification of AR along with the help of PISA method. The PISA method which is
derived from the analysis of the Flow Convergence (FC) region proximal to the regurgitant
orifice and from the conservation of mass has already been previously illustrated in reference
sited [69 – 71]. The regurgitant jet size of LV cavity, LVOT width of the jet and the pressure
half-time are calculated by using CW Doppler. The Echocardiography calculates R vol or RF as
the total stroke volume through the AV is equal to the forward stroke volume plus R vol. Basing
on
these methods the degree of severity of AR can be understood as mild, moderate, severe or
eccentric. The Qualitative and quantitative standard parameters that are used in grading of
severity is illustrated in the table 5.2.
Table 5.2 Qualitative and quantitative parameters useful in grading Aortic Regurgitation
severity
Mild Moderate Severe
Structural Parameters
LA size Normal* Normal / dilated Generally dilated
Aortic leaflets Normal /
abnormal
Normal / abnormal Abnormal/flail,
or wide
coaptation defect
Doppler Parameters
Jet width in LVOT-Color
Flow ξ
Small in central
jets
Intermediate Large in central
jets; variable in
eccentric jets
Jet density-CW Incomplete Thick/Dense Thick/Dense
Jet deceleration rate-CW
(PHT, ms)ψ
Slow>500 Medium 500-200 Steep<200
Quantitative Parameters φ
VC width, cm ξ <0.3 0.3-0.60 >0.6
Jet width/LVOT width, % ξ <25 25-45 46-64 65
Jet CSA/LVOT CSA, % ξ <5 5-20 21-59 60
R vol, ml/beat <30 30-44 45-59 60
RF, % <30 30-39 40-49 50
EROA, cm2 <0.10 0.10-0.19 0.20-0.29 0.30
*Unless there are other reasons for LV dilation. Normally 2D values; LV minor axis
2.8cm/m2, LV end-diastolic volume 82ml/m
2(2).; ξ With a Nyquist limit of 50-60 cm/s. ψ
PHT is shortened with increasing LV diastolic pressure and vasodilator therapy, and may be
lengthened in chronic adaptation to severe AR. (AHA, ACC and ESC recommended values).
Understanding the hemispheric shape of the PISA, the diastolic aortic regurgitant flow
FlowR , is measured as
rFlow VrR 22 …(5.4)
Where in early diastole, the radius r of the FC is calculated and rV represents the
equivalent aliasing velocity. The aortic regurgitant ERO area is then calculated as
VelFlow RRPISAERO /)( …(5.5)
Here VelR represents the maximal velocity of the aortic regurgitant jet in early diastole
recorded with continuous wave Doppler Echocardiography from the apical, par apical or right
parasternal transducer position. Color-flow methods comprise the measurement of the maximal
anteroposterior diameter (height) of the regurgitant jet at the junction of the LV Outflow Tract
(LVOT) and the aortic annulus in parasternal long-axis view, and the maximum height of the LV
outflow tract at the same site. Continuous Doppler-wave imaging of AR allows quantification of
both the slope and pressures half-time. Regurgitant Volume ( VolR ) which is calculated as:
flow mitral - flow aorticVolR (5.6)
VelRA )785.0*(D flow ortic 2
LVOT (5.7)
Here LVOTD represents the diameter of LV Outflow Tract (LVOT). Regurgitant Fraction (
RF ) was calculated asflow aortic
R VolRF . A Regurgitant Fraction above 40% to 50 % accuses
more severe AR.
The quantitatively obtained values of mild and eccentric (figure 5.2) are shown in table
5.3.
Table 5.3 Measured parameter values of Mild and Eccentric Aortic Regurgitation
Quantitative Parameters Mild Eccentric
Radius r (cm) 0.6085 1.32292
Vena Contracta Width (cm) 0.3175 0.635
Jet Width (cm) 0.9525 1.56104
LVOT width (cm) 1.2964 2.2754
Regurgitant Flow Rate (cm2) 111.6868 483.8353
EROA (cm2) 0.22337 0.96767
Aortic flow (cm3) 152.658 267.9303
Rvol (cm3) 21.456 57.9303
Regurgitant Fraction (%) 14.05 21.621
5.5. The Proposed Methodology – Part II
In part II also it is restated that the PFC method using color Doppler has been recognized
as a faithful and accurate quantitative approach for quantifying AR by combining different
image processing techniques which duly quantify the EROA in assessing the degree of severity
of an aortic regurgitant lesion. The approach which is offered for the quantification of AR has
three modules:
i. Preprocessing
ii. Image Segmentation
iii. Quantification using Proximal Isovelocity Surface Area (PISA)
5.5.1. Preprocessing
In preprocessing stage, primarily the color Doppler Echocardiography AR image is
subjected to Gaussian filtering which is used for reducing the noise present in the image. After
filtering make use of image enhancement for enhancing and improving the excellence i.e contrast
and brightness of the image for human viewing.
5.5.1.1. Gaussian Filtering
It is one of the linear filters used in different contexts of image processing. A Gaussian
filter merges both differential and low pass filtering which identify the edges or computing
orientation of features in digital images. If the Gaussian filter is used for low-pass filtering, then
the smooth impulse response is obtained which approximates the Gaussian derivative. The
weighted mean of the input values gives the output of the Gaussian filter at any time t . The
weights are specified by the formula
,...,1,0,1...,;2
exp).()(2
2
Cw (5.8)
Here represents distance in time lapsed from the current moment, is the Gaussian
filter parameter and the sum of all weights is made equal to unit value by the normalization
constant. )(C . The parameter represents the whole Gaussian filter.
Figure 5.3 shows a central severe Aortic Regurgitation image in short axes view and its filtered
output.
Figure 5.3 Aortic Regurgitation image along with their filtered output
[Image Courtesy : www.mpoullis.net (timed and dated 09:30, 04.03.2003)]
5.5.1.2. Image Enhancement
Image enhancement is used for the benefit of human viewers to understand or perceive
the information in images and it offers better input for further image processing methods. After
filtering, the image is subjected to image contrast enhancement. Contrast enhancement is
commonly done by converting an image to a color space where image intensity is one of the
main components. One such color space is *b*a*L [72]. Firstly, therefore, we have to create a
color transformation structure. The said structure decides the specific color space conversion by
using color transform functions to transform the image from RGB to *b*a*L color space, and
afterwards work on the luminosity layer 'L*' of the image.
The values of luminosity which can distance a wide range from 0 to 100 is scaled
between [0 1] as shown below,
100
0i
iLab
100)(I
L …(5.9)
Later the luminosity layer is changed with the processed data and is converted back to the
RGB color space with the help of Contrast-Limited Adaptive Histogram Equalization. This
CLAHE functions on small data regions instead of on the entire image unlike the histogram. The
contrast of each region is improved and so the histogram of each output region almost nearly
equals the specified histogram. (uniform distribution by default). In order to avoid the amplifying
noise that is present in the image, the contrast enhancement can be limited. Figure 5.4 shows
color space conversion output and image contrast enhancement output.
(a) (b)
Figure 5.4 a) Color Space Conversion Output b) Image contrast Enhancement Output
5.5.2. Image Segmentation using Clustering
Image segmentation is comprised as a significant part in low-level vision, image analysis,
and so on. More over the image segmentation decides the quality of the final results of image
analysis, and it is a hard and challenging task in image processing. Normally in image
segmentation, the processes of partitioning an image into several regions are uniform or similar.
But the union of any two adjacent regions is not homogenous (i.e., regions that belong to
individual surfaces by clustering pixels) [73]. In packages, numerous image segmentation
methods are used as per availability. Here based on the Fuzzy K means clustering, the image
segmentation is employed. Fuzzy K means clustering is a pixel-based segmentation that groups
objects which are more similar to each other.
The fuzzy clustering algorithm is one of the most important techniques which is used in
unsupervised pattern recognition. In this way the first step was taken by Ressini after that Zadeh
defined the fuzzy sets. A wide range of applications besides fuzzy control and machine vision
utilized this technique in recent times. K-means is the most famous technique out of the many
fuzzy clustering methods. The main aim of fuzzy clustering (or grouping) is to analyze the data
Sn RYY },{ 1 based on minimum distance measure criterion into a few c clusters as shown
below,
n
k
pc
i ikmikm vYvuJ
1 1|||| ),( …(5.10)
Here the fuzzification parameter is m which obtains a value greater than unity. iv is thi
cluster center. To each cluster, ]1,0[ik is the degree of membership of data and p is the
power of Euclidian distance. By use of optimization algorithm, the optimal values
}{ },,,2 ,1 ,{ miki UciVV are computed.
Figure 5.5 Segmentation Output
5.5.3. Quantification using Proximal Isovelocity Surface Area
For the efficient quantification of Aortic Regurgitation, the preprocessed color Doppler
image was utilized with the aid of PISA method. The evaluation of the flow convergence region
close to the regurgitant orifice and the theory of conservation of mass, lead to the growth of
PISA method [69–71]. The severity of AR was calculated
by means of M-mode, 2-D and Doppler echocardiography and it was graded reliable with the
recommendations of the American Society of Echocardiography (ASE). In Doppler
echocardiography, Aortic Regurgitation is approximated by the size of the regurgitant jet in the
LV cavity, the jet width in the LVOT and the pressure half-time measured by the continuous
wave Doppler. Calculation of R vol or RF is also possible by Echocardiography, because the
sum of forward stroke volume and R vol provides the total stroke volume through the aortic
valve. Basing on these methods, the clinicians can grade the severity of AR whether mild,
moderate, severe or eccentric (See detailed in section 5.4.2).
Table 5.4 Measured values of the above mentioned parameters of Mild and Eccentric
Aortic Regurgitation
Quantitative Parameters Mild Eccentric
Radius r (cm) 0.6085 1.3224
Aliasing Velocity (cm/s) 44 47
Vena Contracta Width (cm) 0.2915 0.6435
Jet Width (cm) 0.2263 0.9125
LVOT width (cm) 0.9652 1.1564
Jet width/LVOT width (%) 23.44 78
Regurgitant Flow Rate (cm2) 102.3136 516.1586
EROA (cm2) 0.0989 0.96767
Aortic flow (cm3) 152.658 158.1882
Rvol (cm3) 21.456 57.9303
Regurgitant Fraction (%) 14.05 36.621
5.6. Conclusions
In this chapter the researcher has presented an efficient quantification of Aortic
Regurgitation. For quantification the Doppler Echocardiographic image is processed by utilizing
image processing techniques. The greater precision was achieved rationally while quantifying
AR Doppler image. In part I, we have engaged Weiner filtering and color quantization in the
preprocessing stage and quantification of AR is done with the help of PISA method. Similarly in
part II, for improving the accuracy of AR to a greater extent we have used Gaussian filtering and
contrast image enhancement in the preprocessing stage. The Aortic Regurgitation quantification
is made very exact by using image segmentation. The preprocessed color Doppler
Echocardiographic image is segmented using fuzzy K-means clustering method. The
enhancements in the quantifying of Valvular Regurgitation are finally led based on
improvements in imaging technologies which assist to improve the measurement of flow
convergence, Vena Contracta and regurgitant jet. Therefore from our research we can conclude
that it is advantageous to find out cardiac output non-invasively by Doppler Echocardiography
with the assistance of flow convergence method. These experimental results are found correlated
with several other procedures that exist for measuring cardiac output.