Study on the morphology of Mature ovarian follicles using Image Processing Toolbox


 Objective: This study aims to investigate the morphological characteristics of mature follicles with clinical value. Study Design:a retrospective studyMethods: The five morphological indexes of follicles in 72 natural ovulation cycles were measured. The shoot time of follicle photos was the day before ovulation. Measurement software was the MATLAB Image Processing Toolbox. Results: The measured average area, perimeter, equivalent circle diameter and roundness were 301.29 ± 8.52 mm2, 62.23 ± 6.82 mm, 20.21 ± 3.08 mm and 0.89 ± 0.05, respectively; and the maximal diameter was 23.12 ± 3.06 mm. Conclusion: Image Processing Toolbox is a simple and convenient measurement software. The standardization of the reference indexes of follicular morphology has an important clinical application value in reproductive medicine.Research Highlights and Graphical Abstract: In the present study, the morphological characteristics of these folliclesinclude the area, perimeter, maximal diameter, equivalent circle diameter and roundness were preliminarily studied using the related measurement software.


Introduction
Monitoring the growth and maturation of follicles by B ultrasound [1][2][3][4] has been widely used in the diagnosis and treatment of infertility. The morphological quality of the follicles [5][6][7] largely re ects the quality of the ova, is closely associated with pregnancy rate and delivery rate, and even affects the incidence of fetal malformation. However, there is no quantitative standard for the morphological characteristics of mature follicles in the domestic academic community. In the clinical diagnosis and treatment, a consensus has been achieved in only one index: the average diameter of mature follicles [1].
From the morphological perspective, the morphological indexes that re ect the quality of follicle include the area, perimeter, maximal diameter, equivalent circle diameter and roundness. In the present study, a certain number of follicular follicles were collected, and the morphological characteristics of these follicles were preliminarily studied using the related measurement software.

Materials
The source of the 72 samples used in this study was those right age women who established the birth service cards in the reproductive center of Huai'an First People's Hospital and became pregnant and delivered from 2015 to 2017. After obtained an informed consent from these women, the natural ovulation cycle during pregnancy was monitored by B ultrasound, and pictures of the follicles on day -1 (the day before ovulation) were taken.

Morphological measurement software
In this study, we used the Japanese ALOKASSD-1400 ultrasound diagnostic apparatus, 3.5-MHz convex array probe and 5.0-MHz transvaginal probe. Follicle photos were taken using 8-million pixel SONYSDSC-T20 digital camera.
The MATLAB Image Processing Toolbox was used to measure the morphological parameters of follicles. MATALB is one of a series of products of the MathWorks company, and Image Processing Toolbox is an image processing toolbox in MATALB. It can be applied in image acquisition [8,9] analysis [10][11][12] development of visualization and algorithm [13] and other services. Compared with traditional programming languages (such as C, C++ and Fortran), Image Processing Toolbox make solving the problem of technical computing more easy. MATALB has very low hardware and software requirements, its written programs can run in a stand-alone window system. Clinicians can obtain the measurement results of morphological indicators by simply using the graphical interface of MATALB.

Methods for quantitative determination of morphology
In this study, all of the follicle photos were taken at day -1 of the natural ovulation cycle in which the woman successfully conceived. Ultrasound images acquired on the ovulation date (day 0) showed that the dominant follicle disappeared. In addition to B ultrasound imaging features, the determination of ovulation day should also be with reference to basal body temperature (BBT), cervical score (CS) [14,15] and urine LH test paper, in order to ensure the accuracy of time determining [4].
The program written in MATALB Image Processing Toolbox [3] was used to quantitatively determine the morphology of each follicle on the photos. The determined morphological indexes included the area, perimeter, maximal diameter, equivalent circle diameter and roundness. The formula for calculating the degree of roundness of follicles is (A is the follicular area, P is the follicular perimeter). After the completion of all the measurement work, a total of 72 samples were obtained.

Statistical methods
The measurement data were analyzed using statistical software SPSS12.0,were expressed as mean ± standard deviation ( ± SD). The mean data were compared using t-test, and count data were compared using X 2 test.

Results
The average values of the parameters of the 72 samples are shown in Table 1. The ve morphological indexes of mature follicles were the follows: the average area was 301.29 ± 8.52 mm 2 , average perimeter was 62.23 ± 6.82 mm, maximal diameter was 23.12 ± 3.06 mm, average equivalent circle diameter was 20.21 ± 3.08 mm, and average roundness was 0.89 ± 0.05. 3. Discussion

Determination of sampling time of follicles in nature ovulation cycle
In a natural ovulation cycle, the growth and development of ovarian follicles are dynamic. A follicle produced in a natural cycle grows and develops at a certain speed, its morphology is different on every day during the cycle. In this study, the shoot times of photos of all samples were assigned on the day before ovulation in the natural ovulation cycle, in accordance with the requirements of statistical mathematics to sample selection. To accurately determine day -1, in addition to ultrasound image characteristics, other common monitoring methods, such as BBT, CS, LH, etc., are also required to use 3.2 Two aspects to re ect follicular quality When monitoring follicle using B ultrasound, although there is no de nite quantization standard for determining follicular quality in clinic, an empirical consensus has been achieved. It includes two aspects: external morphological characteristics and internal turbidity degree of follicles. This study preliminary discussed the former aspect.

Prospect and future
Human natural fertility is mainly single birth. A follicle is produced in a natural ovulation cycle in a women at childbearing age, then grows and develops to mature. A healthy follicle is fertilized after ovulation and grows until delivery. Morphological characteristics of follicle in the cycle is related to pregnancy, childbirth, and even fetal defect and deformity rate. But the understanding of this correlation is only in the level of empirical consensus. Its quantitative research is a complex systemic engineering, needs a number of large samples and long-term follow-ups. This study is just a small step for this systemic engineering.
In the eld of diagnosis and treatment of infertility, individuation of ovulation induction treatment, that is, developing the optimal option of medication according to the speci c situations of every infertility patients, has become the basic consensus of the clinical diagnosis and treatment. dynamic observation on the growth morphology of follicle in ovulation induction cycle of patients through B ultrasound monitoring is the basic auxiliary means of ovulation treatment, clinicians can determine the effect of the used ovulation-inducting drugs according to the B ultrasound images of ovarian follicles of patients, and adjust the medication in time. At present, however, this step is also just in an empirical level, that is, clinicians determine the advantages and disadvantages of the morphological quality of ovarian follicles based on their experience, lacking relatively uni ed and scienti c quantitative standard. This study made a preliminary attempt in this area.

Shortcomings
In addition to the sample size is not particularly large, the samples were all only followed up to the smooth delivery of the volunteers, where no obvious defect was observed in the fetal appearance. Then day -1 in its natural pregnancy cycle was deducted forward. The selected samples obviously could not exclude the possibility that fetus had recessive defects. Therefore, more adequate studies not only need more samples, also requires long-term follow-ups.

Declarations
Ethics approval and consent to participate This study was conducted in accordance with the declaration of Helsinki.
This study was conducted with approval from the Ethics Committee of Huaian First People's Hospital with Nanjing Medical University.
Written informed consent was obtained from all participants.

Competing Interest
The authors declare that they have no competing interests.
Author's Contribution Li-Feng Bao and Hui-Mei Li carried out the study .Xu Min drafted the manuscript.Li-Feng Bao and Hui-Mei Li participated in the design of the study. Hui-Mei Li conceived of the study, and participated in its design and helped to draft the manuscript. All authors read and approved the nal manuscript.
Abbreviations BBT basal body temperature CS cervical score