The measurement of endometrial volume and sub-endometrial vascularity to replace the traditional endometrial thickness as predictors of in-vitro fertilization success

An explainable ultrasound-based machine learning model for predicting reproductive outcomes after frozen embryo transfer

RESEARCH QUESTION

Can an optimal machine learning model be developed to predict reproductive outcomes following frozen embryo transfer (FET)?

DESIGN

This prospective study included 787 infertile females who underwent FET. The participants were split into a training cohort (n = 550) and a test cohort (n = 237) at a ratio of seven to three. Radiomics features were extracted from ultrasound images of the endometrium and junctional zone. A radiomics model was developed to generate the radiomics score (rad score). Logistic regression was applied to process the clinical data and create a clinical model. A fusion machine learning model was developed by integrating the rad score with independent clinical data using the XGboost algorithm. The performance of the models was compared using the area under the receiver operating characteristic curve (AUC). The SHapley Additive exPlanations (SHAP) method was used to interpret and visualize the contributions of features to the outcomes of FET.

RESULTS

The fusion model demonstrated superior performance, as indicated by an AUC of 0.861 (95% CI 0.829-0.890), in the training cohort, surpassing both the clinical model (AUC 0.680, 95% CI 0.635-0.722; P < 0.001) and the radiomics model (AUC 0.814, 95% CI 0.777-0.848; P < 0.001). The SHAP summary plot reveals the impacts of each feature on the predictive model, and the rad score was found to be the main feature. SHAP force plots provided explanations at the individual level.

CONCLUSION

An explainable machine learning model was established utilizing clinical data and ultrasound images to forecast the outcomes of FET. By utilizing the SHAP method, clinicians may better comprehend the contributors to the outcomes of FET in individual patients, and make better decisions before FET.

Prediction of clinical pregnancy after frozen embryo transfer based on ultrasound radiomics: an analysis based on the optimal periendometrial zone

BACKGROUND

To investigate the optimal periendometrial zone (PEZ) in ultrasound (US) images and assess the performance of ultrasound radiomics in predicting the outcome of frozen embryo transfer (FET).

METHODS

This prospective study had 422 female participants (training set: n = 358, external validation set: n = 64). We delineated the region of interest (ROI) of the endometrium (EN) from ultrasound images of the median sagittal surface of the uteri of patients. We determined the ROIs of PEZ on US images by automatically expanding 2.0, 4.0, 6.0, and 8.0 mm radii surrounding the EN. We determined the radiomics characteristics based on the ROIs of the endometrium and PEZ, then compared the different sizes of PEZ to determine the optimal PEZ. We constructed models of the EN and optimal PEZ using six machine learning algorithms. We developed a combined model using the radiomics characteristics of EN and the optimal PEZ. We evaluated the performance of the three models using the area under the curve (AUC).

RESULTS

The optimal PEZ was 4.0 mm with a maximum AUC of 0.715 (95% confidence interval (CI): 0.581 - 0.833) in the external validation set. The combined radiomics model (endometrium and PEZ4.0 mm) yielded the best predictive performance with AUC = 0.853 (95% CI: 0.811 - 0.890) for the training set and AUC = 0.809 (95% CI: 0.696 - 0.909) for the external validation set.

CONCLUSIONS

PEZ4.0 mm could be the optimal area for predicting clinical pregnancy after FET. An US-based radiomics model that combines EN and PEZ4.0 mm demonstrated strong potential in helping clinicians predict FET outcomes more accurately, thereby supporting informed decision-making before treatment.

Automated endometrial identification and volume calculation in normal uteri using a novel smart ERA technique

To evaluate the repeatability of a novel automated technique called Smart ERA (Smart Endometrial Receptivity Analysis) for the automated segmentation and volume calculation of the endometrium in patients with normal uteri,, and to compare the agreement of endometrial volume measurements between Smart ERA, the semi-automated Virtual Organ Computer-aided Analysis (VOCAL) technique and manual segmentation. This retrospective study evaluated endometrial volume measurement in infertile patients who underwent frozen-thawed embryo transfer (FET). Transvaginal three-dimensional ultrasound scans were performed using a Resona R9 ultrasound machine. Data was collected from patients between 2021 and 2022. Patients with normal uteri and optimal ultrasound images were included. Endometrial volumes were measured using Smart ERA, VOCAL at 15° rotation, and manual segmentation. Intra-observer repeatability and agreement between techniques were assessed using the intraclass correlation coefficient (ICC) and Bland-Altman analysis. A total of 407 female patients were evaluated (mean age 33.2 ± 4.7 years). The repeatability of Smart ERA showed an ICC of 0.983 (95% CI 0.984-0.991). The agreement between Smart ERA and the manual method, Smart ERA and VOCAL, and VOCAL and the manual method, as assessed by ICC, were 0.986 (95% CI 0.977-0.990), 0.943 (95% CI 0.934-0.963), and 0.951 (95% CI 0.918-0.969), respectively. The Smart ERA technique required approximately 3 s for endometrial volume calculation, while VOCAL took around 5 min and the manual segmentation method took approximately 50 min. The Smart-ERA software, which employs a novel three-dimensional segmentation algorithm, demonstrated excellent intra-observer repeatability and high agreement with both VOCAL and manual segmentation for endometrial volume measurement in women with normal uteri. However, these findings should be interpreted with caution, as the algorithm's performance may not be generalizable to populations with different uterine characteristic. Additionally, Smart ERA required significantly less time compared to VOCAL and manual segmentation.

An endometrial receptivity scoring system evaluated by ultrasonography in patients undergoing frozen-thawed embryo transfer: a prospective cohort study

Introduction

Ultrasound has become a routine method for endometrial receptivity (ER) evaluation. However, there is controversy over the independent evaluation values of various ultrasound indicators. Some researchers have designed multi-indicator prediction systems, but their prediction values are uneven. To further our understanding of ER, we conducted this prospective cohort study to estimate ER noninvasively and effectively.

Methods

Women who underwent the first frozen-thawed embryo transfer (FET) cycle from April 2019 to July 2021 were included in the study. On the day of transfer, transvaginal three-dimensional ultrasound examination was performed to evaluate ER, including endometrial thickness, morphology, volume, movement, blood flow and flow index. The clinical pregnancy rate was the primary outcome. Based on whether clinical pregnancy was achieved, enrolled patients were divided into pregnant and nonpregnant groups.

Results

This study analyzed 197 FET patients (139 pregnancies in total, 70.5%). The protective factors for clinical pregnancy included primary infertility [adjusted odds ratio (aOR), 1.98; 95% confidence interval (CI), 1.01-3.882; p = 0.047] and more frequent endometrial peristalsis (aOR, 1.33; 95% CI, 1.028-1.722; p = 0.03). Scores of 1-2 were assigned according to the relationship between different ultrasound indicators and the clinical pregnancy rate (CPR). The ER score of the patient was the sum of the scores of the 6 items. The ER score of the pregnant group was significantly higher than that of the nonpregnant group (7.40 ± 1.73 vs. 6.33 ± 1.99, p = 0.001). The CPR increased with an increasing ER score. The CPR in the ER < 6 group was significantly lower than that in the ER >6 group (45.5% vs. 75.6%, p = 0.001).

Conclusion

A noninvasive ultrasound scoring system for ER was proposed. This system may provide a non-invasive guidance perspective, in conjunction with invasive assessments currently used in clinical practice, to achieve more effective embryo transfer.

A predictive model for first-trimester pregnancy inception after IVF-ET based on multimodal ultrasound evaluation of endometrial receptivity

Background

In-vitro fertilization-embryo transfer (IVF-ET) is a commonly used assisted reproductive technology. Its success depends on many factors, including endometrial receptivity. Endometrial receptivity can be evaluated by ultrasound, endometrial biopsy, and magnetic resonance imaging. Compared with the latter two methods, ultrasound has the advantages of wide availability, non-invasiveness, and low cost. Three-dimensional (3D) ultrasound imaging examines endometrial thickness, morphology, and blood vessels, which are associated with the success of embryo implantation. However, there are no reports of endometrial receptivity assessment by 3D ultrasound. Therefore, we aimed to evaluate endometrial receptivity using 3D ultrasound and construct a predictive model for first-trimester pregnancy inception following IVF-ET.

Methods

We performed a prospective observational study on infertile women who underwent IVF-ET between December 2019 and February 2021. These women had 3D ultrasound evaluations, measuring endometrial thickness, volume, pattern, morphology, peristalsis, uterine artery blood flow index, sub-endometrial blood flow index, and distribution pattern. We recorded the occurrence of first-trimester pregnancies in these women. Using Akaike information criterion (AIC) and backward stepwise regression, a first-trimester pregnancy prediction model was constructed based on the minimum AIC value and validated internally and externally.

Results

111 women were enrolled, with 103 included in the analysis. Univariate and multiple logistic regression analyses showed that endometrial thickness and vascularization flow index (VFI) were independent factors associated with the occurrence of a pregnancy. The final prediction model corresponding to the minimum AIC value (65.166) was Y = − 6.131–0.182endometrial thickness + 0.542endometrial volume + 4.374VFI + 0.132age. In the test set, modeling cohort, and external validation cohort, the model showed satisfactory differentiation, with C index of 0.841 (95%CI 0.699–0.817), 0.727 (95%CI 0.619–0.815), and 0.745 (95%CI 0.671–0.840), respectively. The Hosmer–Lemeshow goodness of fit tests reported P = 0.865, 0.139, and 0.070, respectively, indicating a high agreement with the actual IVF-ET outcome. This model reached the highest diagnostic efficiency (sensitivity 88.9%, specificity 75%, Youden index 0.639) at a diagnostic cut-off value of ≥ 0.360.

Conclusions

The predictive model based on endometrial receptivity evaluations by 3D ultrasound had high diagnostic efficiency and could be a simple and effective tool to predict first-trimester pregnancy inception after IVF-ET.

An endometrial receptivity scoring system basing on the endometrial thickness, volume, echo, peristalsis, and blood flow evaluated by ultrasonography

BACKGROUND

Establishing a successful pregnancy depends on the endometrium and the embryo. It is estimated that suboptimal endometrial receptivity account for one-third of implantation failures. Despite the indepth understanding of the processes associated with embryo-endometrial cross-talk, little progress has been achieved for diagnosis and treatments for suboptimal endometrial receptivity.

METHODS

This retrospective study included women undergoing their first frozen-thawed embryo transfer (FET) cycles at our reproductive medicine center from March 2021 to August 2021. Transvaginal three-dimensional (3D) ultrasound was performed in the morning on the day of embryo transfer for all the thawed embryo transfer patients, to evaluate endometrial receptivity, including endometrial thickness, echogenicity, volume, movement and blood flow.

RESULTS

A total number of 562 patients of FET with 315 pregnancies (56.0%) was analyzed. It was found that only the echo of the endometrial central line was different between the pregnant group and non-pregnant group. Other parameters, such as endometrial thickness, volume, endometrial peristalsis, or the endometrial blood flow were not statistically different between the two groups. Then, according to the relationship between the different groups and the clinical pregnancy rate, a score of 0 to 2 was respectively scored. The sum of the scores for the six items was the patient's endometrial receptivity score. It showed that the clinical pregnancy rate increased as the endometrial receptivity score increased, and when the receptivity score reaches at least 5, the clinical pregnancy rate is significantly improved (63.7% versus 49.5%, P=0.001).

CONCLUSION

We developed an endometrial receptivity scoring system and demonstrated its validity. It may aid clinicians in choosing the useful marker in clinical practice and for informing further research.

Evaluating the role of endometrial colour Doppler dynamic tissue perfusion measurements in in vitro fertilisation success

The role of uterine receptivity and endometrial perfusion in in vitro fertilisation (IVF) remains unclear. In our prospective pilot study, we used a new dynamic tissue perfusion measurement method to evaluate this relationship in humans. A standard ovulation induction and embryo transfer method were applied to all the volunteers. On the day of embryo transfer, dynamic images of the uterus were recorded using colour Doppler ultrasound, and these images were uploaded onto PixelFlux Chameleon Software (GmbH, Münster, Germany). After determining the region of interest (ROI), the average velocity, intensity, and area values for all coloured pixels in the ROI, the tissue resistance index (RI) and the tissue pulsatility index were calculated. Endometrial thickness, morphology and dynamic endometrial perfusion parameters were compared between the clinically pregnant and non-pregnant groups. Endometrial thickness, morphology values and endometrial dynamic tissue perfusion measurements were similar between the groups. This study compared perfusion parameters between clinically pregnant and non-pregnant patients by accurately calculating endometrial tissue perfusion using standard software to establish its relationship with implantation success in IVF treatment.IMPACT STATEMENTWhat is already known about this subject? The relationship between IVF success, endometrial receptivity and perfusion is known. Clear valuations of endometrial receptivity require an endometrial biopsy which may cause endometrial damage to the actual IVF cycle. This problem has led researchers to conduct non-interventional studies. Studies have revealed the value of endometrial thickness, pattern and Doppler examination of endometrial uterine arteries in predicting the success of IVF treatment.What do the results of this study add? This prospective pilot study is the first one to use this programme in humans to evaluate uterine receptivity in IVF. Successful results can be obtained by using computer programmes in tissues where perfusion parameters cannot be measured using traditional colour Doppler ultrasonography. Revealing the relationship between tissue perfusion and IVF success will be more effective and accurate with the development of software technologies.What are the implications of these findings for clinical practice and/or further research? To increase the success of IVF treatment, current and new technological developments, as well as imaging methods should continue to be tested.

Prevalence of T-shaped uterus among fertile women based on ESHRE/ESGE and Congenital Uterine Malformation by Experts (CUME) criteria

RESEARCH QUESTION

What is the prevalence of T-shaped uteri among fertile women based on ESHRE/ESGE and Congenital Uterine Malformation by Experts (CUME) criteria?

DESIGN

A prospective cohort study of 258 women of reproductive age with a history of at least one natural pregnancy resulting in live birth. Participants were recruited from the family planning clinic between January 2018 and March 2020. The ESHRE/ESGE classification of congenital anomalies of the female genital tract was used for describing abnormal findings. CUME criteria were also used for diagnosing T-shaped uterus. Uterine cavity volume was measured.

RESULTS

Mean age of participants was 35.4 ± 6.2 years. Participants were diagnosed with the following: congenital uterine abnormality (n = 9 [3.6%]); partial septate uterus (n = 5 [2.0%]) and hemiuterus (n = 2 [0.8%]). Two women (0.8%) were diagnosed with T-shaped uterus and borderline T-shaped uterus based on the ESHRE/ESGE criteria and CUME. Mean lateral indentation angle, lateral indentation depth and T-angle were 156.2° ± 9.53°, 2.85 ± 0.93 mm and 73.3° ± 9.85° in patients with normal uterine cavity. In patients with T-shaped and borderline T-shaped uteri, respective figures were 115° versus 121°, 10 mm versus 7.6 mm and 27.5° versus 70°. Median volume of the uterine cavity in patients with normal uterine cavity and T-shaped uterus was 3.71 ml (minimum 2.0 to maximum 9.03 ml, interquartile range 1.93) and 3.2 ml (2.9 and 3.62 ml), respectively.

CONCLUSIONS

The prevalence of T-shaped uteri in fertile women is low, which corresponds to previous reports of women with poor reproductive history.

Endometrial thickness measurements among Asherman syndrome patients prior to embryo transfer

STUDY QUESTION

Is there an association between endometrial thickness (EMT) measurement and clinical pregnancy rate among Asherman syndrome (AS) patients utilizing IVF and embryo transfer (ET)?

SUMMARY ANSWER

EMT measurements may not be associated with successful clinical pregnancy among AS patients undergoing IVF.

WHAT IS KNOWN ALREADY

Clinical pregnancy rate after IVF is significantly lower in patients with a thin endometrium, defined as a maximum EMT of <7 mm. However, AS patients often have a thin EMT measurement due to intrauterine scarring, with a paucity of data and no guidance on what EMT cutoff is appropriate when planning an ET among these patients.

STUDY DESIGN, SIZE, DURATION

This is a retrospective cohort study of 45 AS patients treated at a specialized advanced hysteroscopic clinic from 1 January 2015, to 1 March 2019.

PARTICIPANTS/MATERIALS, SETTING, METHODS

Review of EMT measurements prior to a total of 90 ETs, among 45 AS patients. The impact of the maximum EMT measurement prior to ET on clinical pregnancy rate was analyzed.

MAIN RESULTS AND THE ROLE OF CHANCE

A total of 25/45 (55.6%) AS patients ultimately went on to have ≥1 clinical pregnancy following a mean ± SD of 2.00 ± 1.26 ET attempts. There was a total of 90 ETs among the 45 AS patients, with 29/90 (32.2%) ETs resulting in a clinical pregnancy. Younger patient age (P = 0.05) and oocyte donation (P = 0.01) were the only variables identified to be significant predictors for a positive clinical pregnancy outcome on bivariate analysis. The mean EMT measurement prior to all ETs among AS patients was 7.5 ± 1.6 mm. EMT measurement prior to ET did not predict a positive clinical pregnancy on either bivariate (P = 0.84) or multivariable analysis (odds ratio 0.91, P = 0.60). 31.8% of EMT measurements measured <7.0 mm. In this small cohort, no difference in the clinical pregnancy rate was detected when comparing ETs with EMT measurements of <7.0 mm versus ≥7.0 mm (P = 0.83). The mean EMT measurement decreased with increasing AS disease severity; 8.0 ± 1.6 mm for mild disease, 7.0 ± 1.4 mm for moderate disease and 5.4 ± 0.1 mm for severe disease.

LIMITATIONS, REASONS FOR CAUTION

Our small sample size limits our ability to draw any definitive conclusions. In addition, patients utilized various infertility clinics. This limits our ability to evaluate the consistency of EMT measurements and the IVF care that was received.

WIDER IMPLICATIONS OF THE FINDINGS

EMT measurement cutoff values should be used with caution if canceling a scheduled ET in AS patients.

STUDY FUNDING/COMPETING INTEREST(S)

This study was not funded. K.I. reports personal fees from Karl Stroz and personal fees from Medtronics outside the submitted work. The other authors have no conflicts of interest.

TRIAL REGISTRATION NUMBER

N/A.

The endometrial proteomic profile around the time of embryo implantation†

Embryo implantation is an intricate process which requires competent embryo and receptive endometrium. The failure of endometrium to achieve receptivity is a recognized cause of infertility. However, due to multiplicity of events involved, the molecular mechanisms governing endometrial receptivity are still not fully understood. Traditional one-by-one approaches, including western blotting and histochemistry, are insufficient to examine the extensive changes of endometrial proteome. Although genomics and transcriptomics studies have identified several significant genes, the underlying mechanism remains to be uncovered owing to post-transcriptional and post-translational modifications. Proteomic technologies are high throughput in protein identification, and they are now intensively used to identify diagnostic and prognostic markers in the field of reproductive medicine. There is a series of studies analyzing endometrial proteomic profile, which has provided a mechanistic insight into implantation failure. These published studies mainly focused on the difference between pre-receptive and receptive stages of endometrium, as well as on the alternation of endometrial proteomics in women with reproductive failure. Here, we review recent data from proteomic analyses regarding endometrium around the time of embryo implantation and propose possible future research directions.

Effect of Endometrial Thickness Change in Response to Progesterone Administration on Pregnancy Outcomes in Frozen-Thawed Embryo Transfer: Analysis of 4465 Cycles

OBJECTIVE

To evaluate whether endometrial thickness (EMT) change in response to progesterone has an effect on pregnancy outcomes in frozen-thawed embryo transfer (FET) cycles.

DESIGN

Retrospective observational study.

SETTING

Tertiary-care academic medical center.

PARTICIPANTS

4465 infertile women undergoing their first FET between January 2010 and December 2015 in our center.

METHODS

This observational study included 4465 patients undergoing their first FET cycles between January 2010 and December 2015. EMT was measured by transvaginal ultrasound one day before progesterone administration and on the day of FET to observe EMT change.

MAIN OUTCOME MEASURES

Clinical pregnancy rate (CPR) and the live birthrate (LBR) was discussed.

RESULTS

Regardless of the endometrial preparation protocols such as artificial cycle, estrogen-progesterone replacement therapy (EP) or natural cycle (NC), EMT may increase, decrease or remain stable on the day of FET compared with that of one day before progesterone administration. CPR in EMT increase, decrease and stable groups were 48.4%, 51.3% and 50.7% in EP cycle versus 49.2%, 52.0% and 48.9% in NC cycle, showing no significant difference between the three groups in both cycles (P= 0.48, P= 0.49). LBR was 40.9%, 45.9% and 42.6% in EP cycle versus 44.2%, 44.8% and 42.1% in NC cycle, also showing no significant difference between the three groups in both cycles (P= 0.16, P= 0.66). In addition, CPR and LBR were not significantly associated with EMT increase.

CONCLUDES

EMT may increase, decrease or remain stable on the day of FET as compared with that of one day before progesterone administration. Whatever change in EMT that occurs after progesterone administration has no significant effect on CPR and LBR in FET cycles.