WO2012148188A2

WO2012148188A2 - Utilization of an image analysis technique for an elastographic image of a uterine cervix in order to evaluate the condition of the uterine cervix of a pregnant woman

Info

Publication number: WO2012148188A2
Application number: PCT/KR2012/003235
Authority: WO
Inventors: 황한성
Original assignee: 건국대학교 산학협력단
Priority date: 2011-04-29
Filing date: 2012-04-26
Publication date: 2012-11-01
Also published as: KR20120122455A; KR101250456B1; WO2012148188A3; WO2012148188A9

Abstract

The present invention relates to the utilization of an image analysis technique for an elastographic image of a uterine cervix in order to evaluate the condition of the uterine cervix of a pregnant woman. More particularly, the present invention relates to the utilization of an image analysis technique, which involves performing elastography on a uterine cervix to evaluate the condition of the uterine cervix of a pregnant woman, and analyzing the elastographic imaging data obtained by means of the elastography using an image analyzer, to thereby obtain objective numerical data on the condition of the uterine cervix and enabling the subjective judgment of an observer to be disregarded.

Description

Application of Image Analysis to Cervical Elastic Images to Assess Maternal Cervical Status

The present invention relates to the use of an image analysis method for cervical elasticity images to evaluate the cervical state of the mother, and more specifically, to the cervix to evaluate the state of the cervix in the mother By performing elastography and analyzing the obtained elastographic imaging data using an imaging analyzer, it is possible to exclude the subjective judgment of the observer and to obtain objective numerical data on the state of the cervix. will be.

In general, mothers often need to analyze the condition of the cervix for a variety of reasons. Induction of labor, judgment of the analgesia, prediction of premature birth, determination of cesarean section operation, and determination of necessity of cervical constriction are all processes experienced by pregnant mothers. In this process, the condition of the uterine cervix is very important to be used to predict the health and prognosis of the fetus and mother. In fact, inducing birth to mothers is a very important part of the clinical area of obstetrics. This accounts for about 20% of all pregnancies and has a significant impact on caesarean delivery rates (Zhang J, Yancey MK, Henderson CE.US national trends in labor induction, 1989-1998. J Reprod Med 2002; 47: 120- 4).

This useful method of evaluating cervical condition is the Bishop score, first described in 1964, and has been used in clinical practice to date. It is also useful for evaluating cervical condition (Crane JM. Transvaginal ultrasound cervical length and successful labor). induction: a systematic review.Obstet Gynecol 2006; 107 (Suppl): 60S).

However, Bishop scores tend to be relatively inaccurate when scores are low, have high interobserver and intraobserver variability, and also cause discomfort in the balance test (Williams MC, Krammer J, O'Brien WF.The value of the cervical score in predicting successful outcome of labor induction.Obstet Gynecol 1997; 90: 784.9; Hendrix NW, Chauhan SP, Morrison JC, Magann EF, Martin JN Jr, Devoe LD.Bishop score: a poor diagnostic test to predict failed induction versus vaginal delivery. Southern Med J 1998; 91: 248.5; Tan PC, Vallikkannu N, Suguna S, Quek KF, Hassan J. Transvaginal sonographic measurement of cervical length vs. Bishop score in labor induction at term: tolerability and prediction of Cesarean delivery.Ultrasound Obstet Gynecol 2007 May; 29 (5): 568-73; Holcomb W Jr, Smeltzer JS.Cervical effacement: variations in belief among clinicians.Obstet Gynecol. 1991; / 78: / 437; Phelps JY, Higby K, Smyth MH, et. al.Accuracy and intraobserver variability o f simulated cervical dilatation measurements.Am J Obstet Gynecol. 1995; / 173: / 942).

In order to overcome this drawback, the use of suffocating ultrasound has recently been used to measure the length of the cervix and use it in clinical practice. Therefore, many studies have reported that the cervical length measured by choking ultrasound can be used to predict the success rate of inferior delivery of pregnant women, to predict cervical status, to predict the delivery of preterm birth, and to estimate the Bishop score. There were also reports of superiority (Daskalakis G, Thomakos N, Hatziioannou L, Mesogitis S, Papantoniou N, Antsaklis A. Sonographic cervical length measurement before labor induction in term nulliparous women.Fetal Diagn Ther 2006; 21: 34-38; Peregrine E , O'Brien P, Omar R, Jauniaux E. Clinical and ultrasound parameters to predict the risk of Cesarean delivery after induction of labor.Obstet Gynecol 2006; 107: 227-233; Rane SM, Guirgis RR, Higgins B, Nicolaides KH. J Matern Fetal Neonatal Med 2005; 17: 315-322; Gabriel R, Darnaud T, Gonzalez N, Leymarie F, Quereux C. Gynecol Obstet Fertil 2001; 29: 919-923).

Recently, however, other papers, including Meta-analysis, reported conflicting conclusions, and it is questioned that ultrasound measurement of cervical length using ultrasound is insufficient for routine use (Crane JM. Transvaginal ultrasound cervical length and successful labor induction: a systematic review.Obstet Gynecol 2006; 107 (Suppl): 60S.; Rozenberg P, Chevret S, Ville Y. (Comparison of pre-induction ultrasonographic cervical length and Bishop score in predicting risk of Cesarean section after labor induction with prostaglandins] .Gynecol Obstet Fertil 2005; 33: 17-22).

There are two important factors in the method of evaluating the cervix by earthquake: one is the extent of the cervix and the other is the hardness of the cervix. Both of these factors are so critical that they are included in the Bishop score and are widely used clinically to date. However, this can only be confirmed by a very uncomfortable test called pelvic examination, and the limits appear when the cervix is tightly closed.

Conventional measurement of cervical length using asphyxiation ultrasound has been much researched since it can be very helpful and more comfortable than earthquake resistance, which is more accurate than hand seismic measurement to measure the amount of cervical remaining. However, there is no way to express the strength of the cervix, earthquake-resistant and asphyxiated ultrasound is still mixed and used. In this regard, cervical length measurement by Bishop scoring or choking ultrasound has been insufficient. Therefore, another diagnostic method is required for more accurate evaluation.

Efforts to measure the strength of tissues or objects using ultrasound have already been used in many clinical domains under the name of elastography. It is a very safe test that has been performed on tissues 20 years ago, and in the clinical field, it was first applied to breasts (Garra BS, Cespedes EI, Ophir J, et al. Elastography of breast lesions: initial clinical). results.Radiology 1997; 202: 79e86).

The basic concept of elastic ultrasound (Elastography) is to measure that the harder structure, such as malignant mass, the smaller the elasticity of the tissue, the smaller the difference in the degree of degeneration of the tissue due to pressure. In particular, many studies have been conducted in the area of breast cancer or prostate cancer, and it has been reported that elastic ultrasound is a useful help in diagnosing cancer in each organ.

In order to evaluate the condition of the cervix, the size of the cervix (the remaining amount) and the degree of aging (hardness) should be evaluated. The existing Bishop score is based on this, but it is difficult to evaluate when the score is low and is very subjective. On the other hand, the cervical length is relatively objective, but because it only covers the length of the cervix, it can be said that it is difficult to evaluate the exact state.

The present invention has been made to solve the above problems and the object of the present invention is to provide a new means for evaluating the cervical condition of the mother.

In order to achieve the above object, the present invention comprises the steps of: a) obtaining an image of the cervix with suffocating ultrasound;

b) obtaining an elastic ultrasound image in a dual mode at the same phase as the image of the cervix;

c) selecting and drawing an area of the cervix in the dual mode image and overlapping the selected area in the ultrasound image; And

d) providing a method of providing information on the state of the cervix, including performing image analysis on the selected elastic ultrasound image region overlapped in the ultrasonic image.

In one embodiment of the present invention, the image of the cervix obtained by the choking ultrasound is preferably a black and white two-dimensional image, but is not limited thereto.

In another embodiment of the present invention, it is preferable that the elastic ultrasound image is gray, but the liquid is white and the solid is black.

In another embodiment of the present invention, the softest part of the ultrasound image is white and its numerical value is 255, the hard part is black and the numerical value preferably corresponds to 0. It is not limited.

In a preferred embodiment of the present invention, the image analysis step is to determine the area ratio of the area having a pixel value of less than 80 of the entire area of the selected ultrasound image overlapped in the ultrasound image as the hard part area of the cervix Preferably, but not limited to.

In another aspect, the present invention provides a method for providing information on cesarean section predictive predictive maternal delivery using information on the state of the cervix of the present invention.

In one embodiment of the invention, the information on the cervical state is preferably characterized in that the area of the hard area of the cervix,

In a preferred embodiment of the present invention, the area of the hard part of the cervix is more preferably set to a cut-off value (192.4mm ² ), but is not limited thereto.

In another aspect, the present invention provides a method for providing information on the induction of delivery using the information on the state of the cervix of the present invention and not delivering within 24 hours.

In one embodiment of the present invention, the information on the cervical state is preferably the area of the hard area of the cervix,

In a preferred embodiment of the present invention, the area of the hard part of the cervix is preferably 184.7 mm ² but the cut-off value (cut-off value) is not limited thereto.

The specific numerical method for the image of the elastic ultrasonic wave used in the present invention is that the numerical technique by the image analysis of the elastic image is not yet used in all other fields and may be a technique that can be applied in other clinical fields as well as gynecology. It is not limited to the above two examples used to give the following examples.

Hereinafter, the present invention will be described.

Clinically, to determine the status of the cervix of pregnant women, many gynecologists, both domestic and foreign, use seismic, which is very uncomfortable for mothers. This is because the seismic intensity of the cervix as well as the magnitude of the cervix can be seen by the seismic feeling, because it is used as clinically very important information. Efforts have been made to replace this uncomfortable, subjective test with a scientific and objective test, but to date it is incomplete.

In particular, the measurement of cervical length using ultrasound has been studied because it provides highly reliable numerical information on the degree of cervical patency. Ultrasonography can be done so easily that it is incomparable with earthquakes. However, by measuring the length of the cervix using ultrasound alone, the strength of the cervix is not known, so it is not possible to replace the earthquake completely.

The ultrasound imaging analysis of the cervix presented in the present invention (elastographic analysis) can be called a breakthrough in this respect, and no one has tried yet. In addition, it is the first attempt not only to obtain an image of the intensity of the cervix but to analyze the image so that the intensity of the cervix can be obtained as an objective value without the subjective opinion of the doctor.

As can be seen from the present invention, the ultrasonographic analysis of the present invention is the first attempt in the cervix, and if this successfully reflects the state of the cervix,

1) It can be shown that the elastic ultrasound can be effectively used in the cervix,

2) Unlike the existing Bishop score, it can be shown that ultrasound can be used to effectively evaluate the condition of the cervix.

3) Being able to evaluate the cervical condition of the mother more accurately will help to evaluate the various mothers, to classify the high-risk mothers in advance, and to receive treatment.

Also

4) It will be able to take a part in the introduction and utilization of new medical technologies in the domestic medical market.

Figure 1 is a photograph and description of the elasticity (elasticity) and strain (Strain) that is the basic principle of elastography.

2 is a general vaginal ultrasonography picture of the gray scale of the cervix.

Figure 3 is an elastography picture of the cervix obtained in black and white.

Figure 4 is an elastography picture of the cervix obtained in color.

5 is a gray scale image (left) and a black and white elastography image (right).

6 is a diagram illustrating a process of overlapping the same region on the right elastoscan image after setting an area in the left cervical image.

7 is a diagram showing the hard area of the cervix measured by the image analyzer (red area) and the data for the area (lower part figure).

Figure 8 is a graph showing the correlation between cervical length (cervical length) and cervical area (cervical area).

Figure 9 is a graph showing the correlation between the cervical length (cervical length) and the hard area of the cervix (hard area of cervix).

10 is a graph showing the correlation between the Bishop score and the cervial area.

11 is a graph showing the correlation between the Bishop score and the hard area of cervix.

Figure 12 is a factor predicting that natural delivery among inpatient mothers failed to deliver by cesarean section for induction of labor, cervical area (CxArea), cervical hard area (CxHardArea), cervical length ( Cxlength) The four elements of the Bishop score are compared using the ROC curve.

Figure 13 is an example of applying a model that predicts not to deliver within 24 hours.

Hereinafter, the present invention will be described in more detail with reference to non-limiting examples.

However, the following examples are intended to illustrate the invention and the scope of the present invention is not to be construed as limited by the following examples.

The present invention is to use the elastography using asphyxiation ultrasound to determine the degree of aging of the cervix in the mother, and to objectively quantify the simple imaging data obtained by using an image analyzer for clinical use.

The present invention is a prospective study aimed at people who underwent suffocating ultrasound to measure cervical length in pregnant women who visited obstetrics and gynecology. Excludes cases of intrauterine fetal death, fetal malformation, conventional surgical uterine surgery or active labor. The gestational age is determined on the basis of the final normal menstrual date (LNMP), but is determined by ultrasound in the first trimester and compared with the crown-rump length (CRL).

1. Measurement of cervical length and elastography values

Cervical length and elastography were measured by two different professional ultrasound users who did not know the mother's Bishop score, and one ultrasound user obtained two ultrasound pictures of the same mother. The mother performs an ultrasound in the position of the dorsal lithotomy position after emptying the bladder.

Commonly known methods of measuring cervical length, internal os and external os are clearly visible in the sagittal view of the cervix, and the cervix is enlarged to occupy more than 75% of the screen, and then the cervical length is measured. Funneling is measured by defining the internal os of the cervix at least 5 mm open.

The instrument obtains a 2D image of the cervix using an ultrasonic device to measure the cervical length, and obtains an ultrasound image screen at the same position as the image. The elastographic image was grayed out, the amniotic fluid was white, and the hardest areas, such as bones, were measured in black.

When obtaining an image by elasticography (Elastography), the same method as when measuring the cervical length with an ultrasound device. In other words, the probe (ultrasonic rod) in close contact with the cervix, but not while pressing the measurement.

2. Image Analysis

Elastography uses elasticity, which is the ratio of strain to applied stress. However, the stress actually applied is difficult to measure in the clinic, but is mostly constant, and only the strain is measured and imaged.

The elastographic image obtained by the elastic ultrasound is not objective in expressing the intensity of the cervix as it is measured only by the image from white to black.

In addition, simply scoring 0, 1, 2, etc. is also very subjective. Therefore, in the present invention, more objective data will be obtained using image analysis.

1) Obtain an elastography image on the same phase as the normal ultrasound image obtained to measure cervical length.

2) Using these two images, select the area of the cervix and move it to the elastographic image.

3) The image analysis is performed by extracting only the image in the area selected by the cervix from the elastographic image obtained.

Various programs can be used for image analysis. Any program can be used. The analysis calculates the area of the area over the selected area that exhibits a certain intensity and measures how the clinical features change according to the area of the area.

Preferred Embodiment

1) It is an ultrasound image that is generally used in gray scale and is an ultrasound image that is still used in actual clinical practice. The cervix was observed by choking ultrasound, and the length of the cervix was measured at 2.56 cm (see FIG. 2).

2) Elastography is measured in the same phase as taking an image of a conventional cervix (Fig. 2). Images obtained by Elastography range from black and white to color (see FIGS. 3 and 4).

3) A gray scale cervical ultrasound image and a black and white elastography image are obtained in a dual mode (FIG. 5).

4) After selecting and drawing the region of the cervix on the left side in the dual mode image of FIG. 5, the region selected on the elastography image on the right side is overlapped (FIG. 6).

5) Imaging analysis is performed on the elastography region selected in FIG. The selected red area represents the area corresponding to the relatively hard area in the cervix.

Therefore, the area of the cervix and the area of the hard area are measured to obtain numerical data (FIG. 7).

6) The numerical values obtained in FIG. 7 were Area Fraction (17.4%) and Area ROI (1129.05 mm ² ). The softest part of the elastographic image is white, the numerical value is 255, the hardest part is black, and the numerical value is zero.

For comparison of the strengths, 80, which is 1/3 of the total strength range (0 to 255), was arbitrarily determined based on the comparison. Lower intensity values indicate stronger areas. Based on this, each value can be obtained. The calculated numerical values are as follows.

Area Fraction: The area ratio of the area having a pixel value of less than 80 among all the areas set in FIG. 6. In other words, 17.4% means the area ratio of the portion of the total set area (area of the cervix) having an intensity value of less than 80 (1/3 or more).

Area ROI: Area of the entire area set in FIG. 6

Hard area of cervix: Area ROI and Area Fraction can be used to calculate the area of a region consisting of pixels with intensity values less than 80 (ie, the area of a relatively hard area).

As can be seen in the examples below, the Area ROI calculated in FIG. 7 is 1129.02 mm ² and the Area Fraction is 17.4%. Therefore, the hard area of cervix is 1129.02 x 0.174 = 196.4 mm ² . Therefore, the area of the hard part of the cervix of the selected patient can be calculated as 196.4 mm ² .

Through the data obtained above,

1) Elastography calculates the area of cervix and the area of hard area and compares it with the length of cervix and Bishop score.

2) In the case of failure due to cesarean delivery due to failure of induction of labor, induction of induction and delivery within 24 hours is a clinically important moment for the mother. In fact, if the mother before the induction of labor can know in advance whether there is a possibility of surgery due to natural delivery failure, or if you know in advance that you need to give more than one day of induction, you can reduce the unnecessary treatment to the mother and cut the cost In addition to the benefits, you will be able to give birth safely.

Until now, published studies have measured the length of the cervix, measured the Bishop score by earthquake-proof, and predicted the likelihood of undergoing cesarean section during induction of labor and pain for more than 24 hours. . However, their sensitivity or specificity is not high (Daskalakis G, Thomakos N, Hatziioannou L, Mesogitis S, Papantoniou N, Antsaklis A. Sonographic cervical length measurement before labor induction in term nulliparous women.Fetal Diagn Ther 2006; 21: 34-38; Peregrine E, O'Brien P, Omar R, Jauniaux E. Clinical and ultrasound parameters to predict the risk of Cesarean delivery after induction of labor.Obstet Gynecol 2006; 107: 227-233; Rane SM, Guirgis RR, Higgins B, Nicolaides KH.Models for the prediction of successful induction of labor based on preinduction sonographic measurement of cervical length.J Matern Fetal Neonatal Med 2005; 17: 315-322; Gabriel R, Darnaud T, Gonzalez N, Leymarie F, Quereux C. Transvaginal ultrasonography of the uterine cervix before induction of labor.Gynecol Obstet Fertil 2001; 29: 919-923).

Therefore, in the present invention, the measurement of the cervical area and the hard part of the cervix was used to predict the successful delivery of mothers of induced labor and the likelihood of delivery within 24 hours, compared with the existing cervix and Bishop score. A total of 48 mothers were included, except for fertility mothers, pregnancy addiction, fetal malformations, maternal diabetes, and twins. All the procedures were carried out with the consent of the mother, and there was no mother's benefit from the test or the mother's disadvantage due to the test, and the procedure of the test was explained.

A. The factors predicting natural birth among inferior birth mothers to deliver by cesarean section are CxArea, CxHardArea, and Cxlength of cervical length. ) Four components of the Bishop score were compared using the ROC curve.

Factors predicting natural delivery among inbred mothers for induction of labor and delivery by cesarean section are CxArea, CxHardArea, Cxlength Bishop Four elements of the scores were compared using the ROC curve.

The AUCs obtained from these ROC curves and their probabilities are shown in Table 2 below, showing very high AUC of cervical area or hard area.

Table 1

variable	Area (AUC)	percentage
CxArea	0.773	0.005
CxHardArea	0.794	0.002
Cxlength	0.749	0.010
Bishop	0.767	0.006

Table 1 shows the Elastographic data and the predictive value of cesarean section for clinical variables.

The highest sensitivity and specificity for predicting caesarean section of CxHardArea of cervix with highest AUC was 192.4mm ² . When the CxHardArea was rearranged based on 192.4mm ² criteria, the sensitivity was 83.3%, specificity 75.0%, positive predictive value 52.6%, and negative predictive value 93.1%.

b. In the same way, a model that starts induction delivery and predicts delivery within 24 hours is as follows.

TABLE 2

variable	Area (AUC)	percentage
CxArea	0.751	0.004
CxHardArea	0.777	0.001
Cxlength	0.711	0.014
Bishop	0.684	0.032

Table 2 shows the Elastographic data and the predictive value of cesarean section for clinical variables.

The highest sensitivity and specificity for the prediction of cesarean section for CxHardArea of cervix with the highest AUC was 184.7 mm ² . When the CxHardArea was organized by 184.7 mm ² , sensitivity 73.7%, specificity 75.9%, positive predictive value 66.7%, and negative predictive value 81.5%.

The above results show that the elastographic data can be used effectively in the clinic. This objectively quantified elastographic data could be used in gynecology and other fields.

Claims

a) obtaining an image of the cervix with suffocating ultrasound;

b) obtaining an elastic ultrasound image in a dual mode at the same phase as the image of the cervix;

c) selecting and drawing an area of the cervix in the dual mode image and overlapping the selected area in the ultrasound image; And

and d) performing image analysis on the selected elastic ultrasound image region overlapped in the elastic ultrasound image.
The method of claim 1,

The image of the cervix obtained by the choking ultrasound is an information providing method for the state of the cervix, characterized in that the black and white two-dimensional image.
The method of claim 1,

The elastic ultrasound image is gray, the liquid is white, the bone is represented in black information providing method for the state of the cervix.
The method of claim 1,

The softest part of the elastic ultrasound image is white, the numerical value is 255, the hardest part (hard) is black, the numerical value of the information providing method for the cervix, characterized in that corresponding to zero.
The method of claim 1,

In the image analysis step, the cervical state is characterized by determining the area ratio of a portion having a pixel value of less than 80 of the entire region of the selected ultrasonic wave images overlapped in the ultrasonic wave image as a hard portion area of the cervix. How to Provide Information.
A method for providing information on cesarean section prediction in maternity of induction using information on the condition of the cervix of any one of claims 1 to 5.
The method of claim 6,

The information about the cervical state is a method of providing information on the prediction of cesarean section in induced delivery mother, characterized in that the hard area of the cervix.
A method for providing information about a delivery within 24 hours of starting delivery, using information on the condition of the cervix of any one of claims 1 to 5.
The method of claim 8,

And said information about the cervical condition provides information about not being delivered within 24 hours of induction of labor, characterized in that the area of the cervical hard area.