Archives of Gynaecology and Women Health

Abstract

Background: Frozen embryo transfer (FET) involves thawing cryopreserved IVF embryos for uterine transfer, and AI predictive models optimize endometrial preparation by integrating clinical, hormonal, and imaging data to improve implantation success. So, the aim of this minireview was to demonstrate the role of AI predictive model for endometrial preparation in FET cycles.

Methods: We used different researching sites as PubMed, Springer, Frontiersin, ELSEVIER, Scopus, Web of Science, and Google Scholar. Relevant articles published in peer-reviewed journals were identified using a combination of keywords related to artificial intelligence, endometrial preparation, and frozen embryo transfer. The search included studies published in English up to 2025. Peer-reviewed original articles, systematic reviews, and clinical studies relevant to AI applications in reproductive medicine were included. Conference abstracts, non-English publications, and articles without available full text were excluded. Reference lists of selected papers were also screened to identify additional relevant studies.

Findings: The reviewed studies consistently showed that AI models contribute to improved assessment of embryo quality and endometrial receptivity in frozen embryo transfer cycles. Predictive algorithms demonstrated potential in personalizing treatment, reducing failed transfers, and enhancing clinical decision-making. Several reports also highlighted the possible cost-effectiveness of AI tools. However, most studies were limited by small sample sizes and the need for external validation.

Conclusion: AI shows great promise in improving embryo and endometrial assessment, personalizing frozen embryo transfer, and reducing failed cycles. Still, large-scale validation is needed before it can be fully integrated into routine clinical practice.

Introduction

Frozen embryo transfer (FET) is a widely accepted procedure used for the storage and transfer of excess embryos produced in fresh in vitro fertilization (IVF) or intracytoplasmic sperm injection (ICSI) cycles [1]. Increasing live birth rates resulting from improvements in technology, as well as increasing demand for preimplantation genetics testing and fertility preservation, has led to a progressive increase in the amount of FET cycles over the past decade [2]. A receptive endometrium and the development of good-quality embryos with potential implantation are required for successful implantation [3]. Despite extensive research endeavors, one of the more enigmatic aspects of assisted reproduction technology (ART) remains the mechanism of embryo implantation [4]. Diverse approaches have been taken in research into embryonic evaluation, encompassing morphological assessments, time-lapse study, blastocyst classifications, chromosome screening and, recently, the integration of artificial intelligence (AI) into clinical settings as a proposed supplementary tool to enhance embryo selection, thereby increasing the likelihood of successful pregnancies [5-6]. Artificial intelligence is an emerging technology aimed at embryo selection, although it may also have alternative aims, such as determining the diagnostic accuracy of PGT [5]. In medical practice, artificial intelligence is becoming more accepted and promises the ability to analyze vast data sets at speeds not possible for humans to compute in the accepted timeframe of preimplantation culture [7]. The advanced mathematics lends itself well to a static image or video analysis of embryo development and can compute and analyses cryptic morphological features [8]. By nature, it is non-invasive and, in due course, may be affordable and more accurate than established approaches, as more data are evaluated and compared between different sources [5]. Embryo grading by conventional morphology evaluation and morphokinetic analysis may not be the best or most accurate way to predict implantation potential, but it is standardized and recognized globally [9]. These AI qualities could aid in improving the current embryo grading system used by embryologists, which could be non-reproducible and sub-optimal within the clinical practice [10]. An ideal AI for this purpose should be trained on objective, reproducible parameters such as PGT-a or a pregnancy outcome (implantation, presence of sacs, heartbeat, or live birth), [11]. This way, the AI would learn over a standardized and reproducible procedure, rather than on subjective observations. By this approach, and in addition to relevant clinical data from each patient and cycle, an AI could be comparable or improve on the embryologist’s performance [12].

Preparation of the Endometrium for Implantation

From donor-egg cycles in women with no ovarian function, it is obvious that only sequential estrogen and progesterone is required for the endometrium to be receptive to embryo implantation [13]. The methodology for achieving estrogen and progesterone levels adequate for implantation can be quite variable but all appear to be equally effective for pregnancy outcome [14]. During the estrogen phase, the endometrium demonstrates linear growth of endometrial glands and blood vessels resulting in the typical trilaminar appearance of the full-thickness endometrium on ultrasound [15]. The proliferation phase ends 2-3 days after ovulation, but there is continuing growth of endometrial glands and vessels under the influence of progesterone within the endometrium, resulting in coiling of the glands and vessels and glycogen secretion into the glandular lumen [16]. These changes are accompanied by increased proliferation of T-cells, macrophages, and lymphoid nodules, and all these effects result in a homogeneous hyperechoic pattern on ultrasound associated with increased endometrial density [17]. The importance of growth in endometrial thickness during the estrogen phase is quite clear. A previous publication by [18] demonstrated that optimal clinical and ongoing pregnancy rates after in vitro fertilization occurred with endometrial thickness of R8 mm at the end of the estrogen phase in more than 24,000 fresh embryo transfer cycles and R7 mm in more than 20,000 frozen-thawed embryo transfer (FET) cycles. There was a significant decline in pregnancy rates for each mm decrease of thickness below 8 mm in fresh embryo transfers and below 7 mm in the FETs. Therefore, it appears that any route of administration of estrogen that results in an endometrial thickness of R8 mm with a trilaminar appearance on ultrasound will be optimal for embryo implantation [19]. Assessment of the Window of Implantation A recent development in assessment of the window of implantation (WOI) is a multigene microarray together with bioinformatics that has been proposed to identify genetic alterations associated with endometrial receptivity in an endometrial biopsy [20]. The endometrial receptivity array (ERA) test examines the expression of 238 genes thought to be involved in implantation [21]. The goal of this test is to enable customized FET based on the determination of a personalized WOI. In a mock cycle, endometrial biopsy is performed on the 7th day after an LH surge or on the 6th day of progesterone during a hormone replacement (HRT) cycle [22]. Results are expressed as perceptive, receptive, or postreceptive. If the result is non-receptive, for example, perceptive, the embryo replacement timing is delayed in a subsequent cycle, thereby enabling personalized embryo transfer [23]. Based on the current understanding of the endometrium and the implantation process, the following are the probable and interrelated five mechanisms regulating a receptive endometrium during the WOI (Figure 1) [20].

Figure 1: Potential Mechanisms Regulating the Optimal Window of Implantation. Based on current literature, five possible and interrelated mechanisms include: (A) Suitable synchrony between the endometrial cells. (B) Adequate synchrony between the endometrium and the embryo. (C) Standard progesterone-signaling and the endometrial responses to progesterone. (D) Silent genetic variations. (E) Typical morphological characteristics of the endometrial glands may together constitute the molecular basis of a receptive endometrium during the window of implantation [20].

Ultrasound Assessment

The problems with endometrial biopsy tests such as histologic dating, the ERA test, Bcl-6 and SIRT-1 measurement, and the endometrial function test are that they are invasive and the results need to be extrapolated to a subsequent cycle in which the embryo transfer will occur [24-25]. This brings us back to ultrasound, which is non-invasive and can be used in the cycle of interest. There is already evidence that endometrial thickness has good negative predictive ability. That is, an endometrium of < 8 mm thickness on the day of hCG trigger in fresh cycles, or < 7 mm at the end of the estrogen phase in FETs, is associated with an incremental decrease in ongoing pregnancy for each mm decrease in thickness [15]. In contrast, endometrial thickness measured in the progesterone phase before embryo transfer had no significant value in predicting pregnancy outcome [26]. However, the change in endometrial thickness between the day of the hCG trigger and the day of embryo transfer has not been examined. It was hypothesized that serial ultrasound tests to determine the change in endometrial thickness between the end of the estrogen phase and the time of embryo transfer may be more important to predict pregnancy outcome than the absolute measure of endometrial thickness at either time point alone [27]. Specifically, it was hypothesized that the endometrial thickness should decrease in the natural or artificial luteal phase as the endometrium becomes denser (hyperechoic on ultrasound) because of the secretory changes that are induced by progesterone [28].

Endometrial Preparation Methods

Endometrial preparation protocols prior to FET can generally be divided into three categories: natural cycles, which rely on endogenous hormonal production from a growing follicle; ovulation induction cycles, which use ovulation induction agents to promote follicular growth, primarily in anovulatory women, to mimic a natural cycle; and artificial cycles, in which exogenous estradiol and progesterone are administered. Figure 1 provides a schematic representation of the endometrial preparation methods and the recommended approaches [29].

Figure 2: Schematic representation of different endometrial preparation methods prior to frozen embryo transfer (FET). tNC, true natural cycle; mNC, managed natural cycle; CC, clomiphene citrate; NPP, natural proliferative phase; LPS, luteal phase support; iLPS, individualized LPS; P, progesterone; ET, endometrial thickness; E2, oestradiol [29].

Natural Cycle Frozen Embryo Transfer

Endometrial changes and preparation are driven by hormonal production associated with follicular growth, ovulation, and formation of the corpus luteum [30]. This process leads to proliferative changes, followed by secretory transformations of the endometrium, ultimately resulting in a receptive state. During the follicular phase, ultrasonographic evaluation of follicular growth and endometrial thickening is performed to assess an adequate endometrial lining, generally considered sufficient when it exceeds 6.5–7 mm [31]. Hormonal blood sampling may be useful to assess adequate follicular oestradiol production, and to detect premature ovulation characterized by an early progesterone rise, generally considered as more than 1.0 ng/ml [32]. True-NC (t-NC) For t-NC, transvaginal ultrasonography is performed on day 2 or 3 of menses to rule out any cyst or corpus luteum prevailing from the previous cycle [33]. Cycle cancellation is usually undertaken in cycles with serum P4 >1.5 ng/ml on day 2 or 3 of menses, even though this practice has been extrapolated from fresh embryo transfer cycles [34]. Transvaginal ultrasonographic monitoring is usually started on day 8-10 and endocrine monitoring is performed, using serum E2, LH and P4 measurements when the leading follicle attains a mean diameter of approximately 15 mm in diameter [35]. Following frequent endocrine and ultrasonographic monitoring, on alternate days or daily, the day of ovulation is precisely documented to schedule the timing of FET [36]. Modified-NC For modified-NC, the initial monitoring is the same as in t-NC; however, ovulation is triggered with hCG once the leading follicle reaches a mean diameter of 16-20 mm [37]. In modified-NC, hCG, not only induces ovulation, but also results in increased serum P4 production during the early and mid-luteal phase, thus, the hCG trigger works as an ovulation trigger as well as an early LPS. The place for endocrine monitoring in modified NC is controversial [38-39]. Whether monitoring of serum P4 and LH levels in modified-NC FET cycles has added clinical value needs to be explored [40]. Determination Of Ovulation Onset and Timing of Transfer Earlier studies yielded conflicting results regarding the efficacy of using a human chorionic gonadotrophin (HCG) trigger in natural cycle frozen embryo transfer (NC-FET) cycles [40-41]. More recent research by Mackens et al. (2020) and Ranisavljevic et al. (2025) [42-43] confirmed that the use of HCG in NC-FET results in comparable pregnancy outcomes, thereby allowing for more flexible planning. FET is typically timed based on either the onset of spontaneous ovulation referred to as ‘true NC-FET’ (tNC-FET) or by the administration of HCG termed ‘modified NC-FET’ (mNC-FET),[44]. Preliminary data from a recent RCT involving 604 patients compared pregnancy outcomes following single blastocyst transfers performed either 6 or 7 days after an HCG trigger, and found no significant difference in LBR (33.8% versus 34.5% for day 6 versus day 7 transfers, respectively; adjusted risk ratio 0.98, 95% CI 0.76–1.24), confirming a multiday window for optimal implantation [45]. Similarly, a retrospective cohort study evaluating euploid embryo transfers found higher LBR when transfers were performed 160 ± 4 h post- HCG administration, although LBR remained highly effective within a broader time range [46]. Luteal Phase Support in Natural Cycle Frozen Embryo Transfer In NC-FET, progesterone production is predominantly regulated by the corpus luteum, a transient endocrine gland formed after ovulation from residual granulosa and theca cells [47]. During the luteal phase, progesterone production by the corpus luteum is driven by LH, and in early pregnancy, this function is sustained by HCG produced by the trophoblast [48]. Progesterone production remains dependent on the corpus luteum until the placenta assumes this role around week 5–7 of gestation. An adequate mid-luteal progesterone concentration near the time of embryo transfer is crucial to achieve successful pregnancy outcomes [49]. A retrospective study by Gaggiotti-Marre et al. (2020) [50] demonstrated significantly higher LBR when the progesterone concentration exceeded 10 ng/ml on the day before blastocyst transfer in tNC-FET. Moreover, 37% of their study population exhibexhibited a low progesterone concentration, indicating that a substantial proportion of patients may experience luteal phase deficiency. A more recent RCT by Wånggren et al. (2022) [51] demonstrated that LPS initiated from the day of transfer in tNC-FET, regardless of progesterone concentration, was associated with improved LBR. In summary, LPS may be unnecessary in mNC-FET due to HCG-stimulated support of the corpus luteum, whereas it may benefit patients undergoing tNC-FET, particularly those with a low progesterone concentration on the day of blastocyst transfer or in cases of inaccurately diagnosed ovulation.

Ovulation Induction Cycles

Ovulation induction cycles are largely comparable with NCFET, with the primary distinction being that follicular growth is stimulated using ovulation induction agents such as aromatase inhibitors, selective oestrogen receptor modulators, or gonadotrophins [52]. This approach may be particularly advantageous for anovulatory patients, as it preserves natural ovulation and the protective functionality of the corpus luteum [29]. In patients with polycystic ovary syndrome, studies have demonstrated a lower risk of hypertensive disorders of pregnancy in letrozole-induced cycles compared with artificial cycles [53]. Furthermore, research has suggested that letrozole-induced cycles may yield more favourable pregnancy outcomes than artificial or even natural cycles in normo-ovulatory patients [54-55]. Additional investigations are necessary to determine whether letrozole-induced cycles are superior to other preparation methods, and whether the suppression of oestradiol production by letrozole leads to enhanced progesterone production during the luteal phase [29]. Artificial Cycle Frozen Embryo Transfer In artificial cycle frozen embryo transfer (A-FET), endometrial preparation is achieved through the sequential administration of oestradiol and progesterone to replicate the hormonal changes observed in a natural cycle. The administration of exogenous oestradiol suppresses follicular growth via negative feedback; as a result, most AC-FET are anovulatory [56]. Oestradiol Supplementation Oestradiol supplementation in AC-FET typically begins at the start of the menstrual cycle and continues for 14 days or longer, consistent with the standard duration of the follicular phase, until an endometrial lining of 6.5–7 mm is achieved [31]. However, prolonged oestradiol administration exceeding 28 days until FET was associated with a decrease in LBR, as shown in an observational cohort study of 1377 cycles [57]. These findings were not confirmed by a larger retrospective study by Rodríguez-Varela (2023) [58], which analyzed more than 7000 cycles and found no association between oestradiol duration and pregnancy rate. Mid-luteal oestradiol concentration measured on the day of blastocyst transfer may further influence pregnancy outcomes. Concentrations between 292 pg/ml and 409 pg/ml were associated with optimal outcomes, while concentrations outside this range correlated with a reduced pregnancy rate [59]. Progesterone Supplementation Progesterone supplementation can be administered following adequate oestradiol priming to induce secretory changes in the endometrium, thereby opening the window of implantation [40-60]. Individualized LPS has gained attention in AC-FET, largely due to the work of Labarta et al. (2021) [61], who identified reduced pregnancy outcomes when the progesterone concentration fell below 8.8 ng/ml on the day of transfer. Subsequent studies demonstrated that adding subcutaneous progesterone can rescue such cycles, resulting in comparable pregnancy outcomes [62]. Similarly, oral dydrogesterone has also shown potential as a rescue strategy (Mackens et al., 2023; Metello et al., 2024) [63]. Significant intraday variation in serum progesterone concentration has been documented, reflecting the pharmacokinetics of exogenously administered progesterone, as endogenous progesterone is typically absent in AC-FET [64]. Other routes of progesterone administration – such as oral dydrogesterone, and intramuscular and subcutaneous progesterone – are used less commonly but exhibit different pharmacokinetic profiles and corresponding cut-off levels for adequate progesterone during the mid-luteal phase [29]. Combination strategies using different routes of progesterone administration are gaining attention, with several studies demonstrating superiority over monotherapy. These combination approaches may offer a viable alternative when mid-luteal progesterone measurements and subsequent rescue strategies are not feasible [48-65].

Mild-Ovarian Stimulation (Mild-OS)

Mild OS with an oral agent (CC or letrozole) and/or exogenous gonadotropins may be used to prime the endometrium for FET [40]. For this purpose, mild OS is performed with <150 IU urinary/recombinant follicle stimulating hormone (FSH)/day, letrozole at a dose of 2.5 – 5 mg/day or CC at a dose of 50-100 mg/day, starting on the 2nd or 3rd day of the cycle. The follicular response is monitored by frequent vaginal ultrasonography and/or serum endocrine assessment. Human chorionic gonadotropin is administered when the diameter of the leading follicle is greater than17 mm, endometrial thickness ≥7 mm and serum E2 level >150 pg/ml [66]. The timing of the FET is scheduled according to the day of embryo freezing; day-3 embryos are transferred on hCG+5 and day-5/6 embryos are transferred on hCG+7 [67]. The rationale for mild OS in regularly cycling women is to improve subtle defects in folliculogenesis and subsequent luteal phase, resulting in a better endometrial milieu for embryo implantation. In addition, mild-OS avoids the reported risks (e.g. thromboembolic events) associated with exogeneous E2 and P administration in HRT cycles. Letrozole is an aromatase inhibitor; it has a half-life of ~2 days compared to ~2 weeks of CC. Unlike CC, the hypothalamic-pituitary-ovarian axis is intact during letrozole use. Letrozole has no negative effect on the endometrium [35]. Pregnancy Outcomes Between Different Protocols When comparing pregnancy outcomes between NC-FET and AC-FET, the latest Cochrane review concluded that there is no evidence favoring one method over the other [68]. Additionally, a recent large RCT, which was powered to detect differences in LBR between natural, modified natural and artificial cycles, confirmed these findings, showing no significant differences in LBR in the intention-to-treat analysis (37%, 33% and 34%, respectively). Given that many patients switched from NC-FET or mNC-FET to AC-FET after a cancelled cycle, the per-protocol analysis is important to interpret the outcomes accurately. This analysis yielded similar results (LBR 33.7%, 31.0% and 34.0% for NC-FET, mNC-FET and AC-FET, respectively), [69]. However, a recently published large multicenter cohort study reported that AC-FET was associated with significantly higher rates of pregnancy loss (36.5%, 25.7% and 23.6% for AC-FET, ovulation induction FET and NC-FET, respectively) and lower LBR (16.9%, 18.8% and 19.3% for AC-FET, ovulation induction FET and NC-FET, respectively) compared with ovulation induction or natural cycles [70]. Using transvaginal ultrasonography to measure the endometrial thickness is the most common clinical approach to evaluate endometrial receptivity [71]. A retrospective study compared two regimens of endometrial preparation in 2664 women with PCOS undergoing FET. The results showed that the endometrial thickness on the day of progesterone supplementation and on the day of embryo transfer was significantly thicker in patients receiving mild-OS with letrozole than in those receiving AC [72]. Additionally, after adjusting the related confounding factors, this study demonstrated that LBR was significantly higher, and the early pregnancy loss rate was lower in the letrozole group compared with the AC group [73]. The endometrial thickness was significantly thinner in mild-OS with CC than in AC [68]. Role of Artificial Intelligence in Reproductive Medicine AI functions today as an augmenting tool that extracts patterns from large, multimodal fertility datasets (images, time-lapse videos, ultrasound, hormone panels, and clinical records) to support diagnosis, prognosis and protocol personalization [74]. In reproductive medicine this has been translated into objective, reproducible assessments (for example of embryo morphology or endometrial features) that reduce inter-observer variability and can surface subtle signals humans miss enabling more data-driven counseling and research hypotheses [75]. It was emphasized that AI enables more objective and personalized fertility care, integrating multimodal data (imaging, hormonal profiles, patient history) into predictive systems [76].

AI-based Endometrial Analysis as a Predictor of ART Outcomes

A receptive endometrium and the development of good quality embryos with potential implantation are required for successful implantation [77]. Despite extensive research endeavors, one of the more enigmatic aspects of assisted reproduction technology (ART) remains the mechanism of embryo implantation. Diverse approaches have been taken in research into embryonic evaluation, encompassing morphological assessments, time-lapse studies, blastocyst classifications, chromosome screening, and, recently, the integration of artificial intelligence into clinical settings as a proposed supplementary tool to enhance embryo selection, thereby increasing the likelihood of successful pregnancies [5-8]. Higher live birth rates with endometrial thickness of 10–12 mm were demonstrated in cycles where a fresh embryo transfer was performed, and in frozen embryo transfer cycles, live birth rates plateaued after 7–10 mm endometrial thickness [31]. The intricate cellular composition of the endometrium has been explored by Greenwald et al. (2022) and Yamaguchi et al. (2021) [79-80] surpassing traditional parameters such as trilaminar patterns and overall thickness measurements in their studies. A novel “rhizome” structure, which is an intricate network of endometrial glands extending along the myometrium, has been identified in these studies through the utilization of 3D imaging [81]. This discovery provides a novel framework for comprehending the physiology of the endometrium concerning its receptivity to embryos [4]. Notably, AI has been widely used for embryo selection [12]. A recent study introduced a methodology that examines both the absolute and relative dimensions of the external layers of the endometrium. This approach departs significantly from conventional, obsolete and misleading paradigms. AI model was trained based on the findings related to the external layers of the endometrium. It was observed that when the external layers constitute 50% or more of the total endometrial composition in a trilaminar configuration, there is a substantial improvement in pregnancy rates. On the contrary, when the proportion of external layers falls below 50% of the endometrial thickness, a noticeable decline in pregnancy rates occurs [82]. Endometrial Evaluation and Classification System AI model was trained based on the Asch classification using an ultrasonographic endometrial assessment, as follows: 1) well-defined hyperechoic external layers, 2) thickness of the external layers, 3) echogenic mid line, 4) entirety thickness of the endometrium, 5) hypoechoic intermediate layer positioned between external layers and midline, and 6) the percentage of external layers relative to the total endometrial thickness (equal or greater than 50%, or less than 50%). The resultant classification scheme is summarized in Figure 3 [82].

Figure 3: Asch endometrial grading system. Based on the endometrium grading system and the likelihood of pregnancy, the images were categorized into good and bad [82].

AI-Aided Prediction of Endometrial Receptivity

EndoClassify AI model The development of the AI model involved the de-identification of the images selected for the study, image augmentation processes to train the model with high-quality, diverse, and relevant data, followed by the evaluation and selection of segmentation (Attention U-Net) and classification (Inception GoogLeNet) models with a relatively modest computation cost22 (Figure 4), [83]. A two-tiered AI model (EndoClassify) was implemented using convolutional neural networks. The introduction of the EndoClassify AI model enhances the assessment of the endometrium by introducing a new method to evaluate endometrial conditions based on transvaginal ultrasound images [84]. Image quality is objectively evaluated by this model, and images are categorized as either ‘Good’ or ‘Bad’ according to a rigorous set of criteria [85]. Furthermore, valuable insights are provided by quantifying the percentage likelihood of pregnancy for each classification, furnishing clinicians with essential information for decision-making regarding embryo transfer or the postponement of the cycle [4].

Figure 4: Two-tiered AI model, segmentation and classification [4].

The decision to employ a two-step process was based on the benefits of applying segmentation before classification. Initially, segmentation was applied to isolate regions of interest (ROI) from the endometrial ultrasound images. Subsequently, classifications were applied to assign labels to de-identified ROIs by allowing features to be captured by the network at multiple scales and resolutions. This label indicated the degree of endometrial receptivity, expressed as a percentage of a good or bad endometrium, as defined by our novel endometrial classification system. Various AI classification models were evaluated to assess final model’s accuracy, precision, and ability to avoid false positive predictions (Figure 5)[82].

Figure 5: Manual Endometrium Grading System & EndoClassify AI model [82]

Advantages of AI Predictive Models in Clinical Decision-Making: AI-based predictive models provide three major advantages. First, personalization: they combine multiple predictors (e.g., age, endometrial thickness, embryo morphology, medical history) into individualized probability estimates, improving treatment tailoring. Second, efficiency and consistency: randomized trials demonstrated that AI embryo selection is comparable to expert embryologists while being significantly faster [86]. Finally, decision support: predictive tools assist clinicians in optimizing transfer timing, endometrial preparation, and patient counseling, which may reduce failed cycles and resource waste [1].

Clinical Implications and Benefits of AI in Frozen Embryo Transfer

Improved Personalization of Treatment AI models allow clinicians to integrate multiple clinical, hormonal, and imaging parameters to predict the optimal conditions for frozen embryo transfer [87]. By considering variables such as endometrial thickness, hormonal profiles, and embryo characteristics, AI supports truly individualized treatment planning. Recent studies report that AI-based endometrial receptivity analysis improves transfer timing, which can enhance implantation and pregnancy outcomes compared with conventional protocols [1]. Reduction of Failed Cycles A major challenge in ART is the high proportion of unsuccessful cycles. AI-driven predictive models can identify embryos with higher implantation potential and determine the optimal uterine environment, thereby reducing the likelihood of failed transfers. For example, machine learning approaches applied to time-lapse imaging have shown higher predictive accuracy for implantation compared with traditional morphology assessment, contributing to fewer failed cycles [88]. Potential Cost-Effectiveness Repeated failed cycles in IVF and FET increase financial and emotional burden for patients [89]. By improving success rates per transfer, AI systems may reduce the number of attempts needed to achieve pregnancy, lowering overall treatment costs. Preliminary cost-effectiveness analyses suggest that AI-guided embryo selection and transfer optimization could make ART more affordable in the long term, especially as models become widely available and require less manual intervention [90].

Conclusion

Frozen embryo transfer success depends on coordinated embryo quality and endometrial receptivity, yet predicting implantation remains a major challenge in assisted reproduction. Recent advances highlight the role of artificial in reproduction. Recent advances highlight the role of artificial intelligence as a non-invasive, data-driven tool that enhances embryo grading, endometrial evaluation, and transfer timing. Predictive models can integrate multiple variables such as embryo morphology, time-lapse imaging, endometrial thickness, and patient clinical data to provide individualized treatment strategies. By improving personalization of treatment, reducing failed cycles, and offering potential cost-effectiveness, AI holds promise as an adjunct in clinical decision-making. However, further validation through large-scale studies is essential before its integration into routine reproductive practice.

References

1. Wu, Hanglin, Ping Zhou, Xiaona Lin, Shasha Wang, and Songying Zhang. “Endometrial preparation for frozen–thawed embryo transfer cycles: a systematic review and network meta- analysis.” Journal of assisted reproduction and genetics 38, no. 8 (2021): 1913-1926
   View at Publisher | View at Google Scholar
2. Wyns, C., C. Bergh, C. Calhaz-Jorge, Ch De Geyter, M. S. Kupka, T. Motrenko, I. Rugescu, J. Smeenk, A. Tandler-Schneider, and S. Vidakovic. “ART in Europe, 2016: results generated from European registries by ESHRE.” Human reproduction open 2020, no. 3 (2020): hoaa032.
   View at Publisher | View at Google Scholar
3. Neykova, Konstantsa, Valentina Tosto, Irene Giardina, Valentina Tsibizova, and Georgi Vakrilov. “Endometrial receptivity and pregnancy outcome.” The Journal of Maternal-Fetal & Neonatal Medicine 35, no. 13 (2022): 2591-2605.
   View at Publisher | View at Google Scholar
4. Schuff, Ricardo H. Asch, Jorge Suarez, Nicolas Laugas, Marlene L. Zamora Ramirez, and Tamar Alkon. “Artificial intelligence model utilizing endometrial analysis to contribute as a predictor of assisted reproductive technology success.” Journal of IVF-Worldwide 2, no. 2 (2024): 1-8.
   View at Publisher | View at Google Scholar
5. Chavez-Badiola, Alejandro, Adolfo Flores-Saiffe-Farías, Gerardo Mendizabal-Ruiz, Andrew J. Drakeley, and Jacques Cohen. “Embryo Ranking Intelligent Classification Algorithm (ERICA): artificial intelligence clinical assistant predicting embryo ploidy and implantation.” Reproductive biomedicine online 41, no. 4 (2020): 585-593.
   View at Publisher | View at Google Scholar
6. Diakiw, Sonya M., Jonathan MM Hall, Matthew VerMilyea, Adelle YX Lim, Wiwat Quangkananurug, Sujin Chanchamroen, Brandon Bankowski et al. “An artificial intelligence model correlated with morphological and genetic features of blastocyst quality improves ranking of viable embryos.” Reproductive biomedicine online 45, no. 6 (2022): 1105-1117.
   View at Publisher | View at Google Scholar
7. Kakkar, Pragati, Shruti Gupta, Kasmiria Ioanna Paschopoulou, Ilias Paschopoulos, Ioannis Paschopoulos, Vassiliki Siafaka, and Orestis Tsonis. “The integration of artificial intelligence in assisted reproduction: a comprehensive review.” Frontiers in Reproductive Health 7 (2025): 1520919.
   View at Publisher | View at Google Scholar
8. Zaninovic, Nikica, and Zev Rosenwaks. “Artificial intelligence in human in vitro fertilization and embryology.” Fertility and Sterility 114, no. 5 (2020): 914-920.
   View at Publisher | View at Google Scholar
9. Bartolacci, Alessandro, Mariabeatrice Dal Canto, Maria Cristina Guglielmo, Laura Mura, Claudio Brigante, Mario Mignini Renzini, and Jose Buratini. “Early embryo morphokinetics is a better predictor of post-ICSI live birth than embryo morphology: speed is more important than beauty at the cleavage stage.” Zygote 29, no. 6 (2021): 495-502.
   View at Publisher | View at Google Scholar
10. Adolfsson, Emma, and Anna Nowosad Andershed. “Morphology vs morphokinetics: a retrospective comparison of inter-observer and intra-observer agreement between embryologists on blastocysts with known implantation outcome.” JBRA assisted reproduction 22, no. 3 (2018): 228
    View at Publisher | View at Google Scholar
11. Buldo-Licciardi, Julia, Michael J. Large, David H. McCulloh, Caroline McCaffrey, and James A. Grifo. “Utilization of standardized preimplantation genetic testing for aneuploidy (PGT-A) via artificial intelligence (AI) technology is correlated with improved pregnancy outcomes in single thawed euploid embryo transfer (STEET) cycles.” Journal of Assisted Reproduction and Genetics 40, no. 2 (2023): 289-299.
    View at Publisher | View at Google Scholar
12. Salih, M., Chris Austin, R. R. Warty, C. Tiktin, D. L. Rolnik, M. Momeni, H. Rezatofighi et al. “Embryo selection through artificial intelligence versus embryologists: a systematic review.” Human Reproduction Open 2023, no. 3 (2023): hoad031.
    View at Publisher | View at Google Scholar
13. Yu, Kun, Zheng-Yuan Huang, Xue-Ling Xu, Jun Li, Xiang-Wei Fu, and Shou-Long Deng. “Estrogen receptor function: impact on the human endometrium.” Frontiers in endocrinology 13 (2022): 827724.
    View at Publisher | View at Google Scholar
14. Ghobara, Tarek, Tarek A. Gelbaya, and Reuben Olugbenga Ayeleke. “Cycle regimens for frozen-thawed embryo transfer.” Cochrane Database of Systematic Reviews 7 (2017).
    View at Publisher | View at Google Scholar
15. Casper, Robert F. “Frozen embryo transfer: evidence-based markers for successful endometrial preparation.” Fertility and sterility 113, no. 2 (2020): 248-251.
    View at Publisher | View at Google Scholar
16. Dwiayu Larasati, Manggiasih, and Ita Ita. “Characteristics of endometrial cells and the factors that influence the implantation process.” Characteristics of Endometrial Cells and the Factors that Influence the Implantation Process 4, no. 1 (2021): 9-19.
    View at Publisher | View at Google Scholar
17. Youngster, Michal, Matan Mor, Alon Kedem, Itai Gat, Gil Yerushalmi, Yariv Gidoni, Jonathan Barkat et al. “Endometrial compaction is associated with increased clinical and ongoing pregnancy rates in unstimulated natural cycle frozen embryo transfers: a prospective cohort study.” Journal of Assisted Reproduction and Genetics 39, no. 8 (2022): 1909-1916.
    View at Publisher | View at Google Scholar
18. Liu, K. E., M. Hartman, A. Hartman, Z-C. Luo, and N. Mahutte. “The impact of a thin endometrial lining on fresh and frozen– thaw IVF outcomes: an analysis of over 40 000 embryo transfers.” Human Reproduction 33, no. 10 (2018): 1883-1888.
    View at Publisher | View at Google Scholar
19. Huang, Jialyu, Jiaying Lin, Hongyuan Gao, Jing Zhu, Xuefeng Lu, Ning Song, Renfei Cai, and Yanping Kuang. “Value of endometrial thickness change after human chorionic gonadotrophin administration in predicting pregnancy outcome following fresh transfer in vitro fertilization cycles.” Archives of Gynecology and Obstetrics 303, no. 2 (2021): 565-572.
    View at Publisher | View at Google Scholar
20. Sun, Bei, and John Yeh. “Non-invasive and mechanism-based molecular assessment of endometrial receptivity during the window of implantation: current concepts and future prospective testing directions.” Frontiers in Reproductive Health 4 (2022): 863173.
    View at Publisher | View at Google Scholar
21. Jia, Yan, Ya-Jun Dong, Yu-Lin Sha, Song-Chen Cai, Liang-Hui Diao, Zhu Qiu, Yan-Hua Guo, Yan Huang, Hong-Xia Ye, and Su Liu. “Effectiveness comparison between endometrial receptivity array, immune profiling and the combination in treating patients with multiple implantation failure.” American Journal of Reproductive Immunology 87, no. 3 (2022): e13513.
    View at Publisher | View at Google Scholar
22. Zhang, Wen-bi, Jue Li, Qing Li, Xiang Lu, Jun-ling Chen, Lu Li, Hua Chen et al. “Endometrial transcriptome profiling of patients with recurrent implantation failure during hormone replacement therapy cycles.” Frontiers in Endocrinology 14 (2024): 1292723.
    View at Publisher | View at Google Scholar
23. Dahiphale, Swati M., Deepika Dewani, Jayashree M. Dahiphale, Manjusha Agrawal, Apoorva Dave, Sandhya Pajai, Garapati Jyotsna, and S. M. Dahiphale. “A Comprehensive review of the endometrial receptivity array in embryo transfer: Advancements, applications, and clinical outcomes.” Cureus 16, no. 8 (2024).
    View at Publisher | View at Google Scholar
24. Sansone, Alison M., Brooke V. Hisrich, R. Brandt Young, William F. Abel, Zachary Bowens, Bailey B. Blair, Avery T. Funkhouser et al. “Evaluation of BCL6 and SIRT1 as non-invasive diagnostic markers of endometriosis.” Current Issues in Molecular Biology 43, no. 3 (2021): 1350-1360.
    View at Publisher | View at Google Scholar
25. Haas, Jigal, and Robert F. Casper. “Observations on clinical assessment of endometrial receptivity.” Fertility and sterility 118, no. 5 (2022): 828-831.
    View at Publisher | View at Google Scholar
26. Bu, Zhiqin, Xinhong Yang, Lin Song, Beijia Kang, and Yingpu Sun. “The impact of endometrial thickness change after progesterone administration on pregnancy outcome in patients transferred with single frozen-thawed blastocyst.” Reproductive Biology and Endocrinology 17, no. 1 (2019): 99.
    View at Publisher | View at Google Scholar
27. Kaye, Leah, Melody A. Rasouli, Angela Liu, Ankita Raman, Carrie Bedient, Forest C. Garner, and Bruce S. Shapiro. “The change in endometrial thickness following progesterone exposure correlates with in vitro fertilization outcome after transfer of vitrified-warmed blastocysts.” Journal of Assisted Reproduction and Genetics 38, no. 11 (2021): 2947-2953.
    View at Publisher | View at Google Scholar
28. Haas, Jigal, Ramsey Smith, Eran Zilberberg, Dan Nayot, James Meriano, Eran Barzilay, and Robert F. Casper. “Endometrial compaction (decreased thickness) in response to progesterone results in optimal pregnancy outcome in frozen- thawed embryo transfers.” Fertility and Sterility 112, no. 3 (2019): 503-509.
    View at Publisher | View at Google Scholar
29. Roelens, Caroline, and Christophe Blockeel. “Cycle management in frozen embryo transfer: the best of all worlds?.” Reproductive BioMedicine Online 50, no. 4 (2025).
    View at Publisher | View at Google Scholar
30. Kapper, Celine, Peter Oppelt, Clara Ganhör, Ayberk Alp Gyunesh, Barbara Arbeithuber, Patrick Stelzl, and Marlene Rezk-Füreder. “Minerals and the menstrual cycle: impacts on ovulation and endometrial health.” Nutrients 16, no. 7 (2024): 1008.
    View at Publisher | View at Google Scholar
31. Mahutte, Neal, Michael Hartman, Lynn Meng, Andrea Lanes, Zhong-Cheng Luo, and Kimberly E. Liu. “Optimal endometrial thickness in fresh and frozen-thaw in vitro fertilization cycles: an analysis of live birth rates from 96,000 autologous embryo transfers.” Fertility and sterility 117, no. 4 (2022): 792-800.
    View at Publisher | View at Google Scholar
32. Lawrenz, Barbara, Carol Coughlan, Laura Melado, and Human M. Fatemi. “The ART of frozen embryo transfer: back to nature!.” Gynecological Endocrinology 36, no. 6 (2020): 479-483.
    View at Publisher | View at Google Scholar
33. Mumusoglu, Sezcan, Murat Erden, Irem Yarali Ozbek, Onur Ince, Sandro C. Esteves, Peter Humaidan, and Hakan Yarali. “The true natural cycle frozen embryo transfer-impact of patient and follicular phase characteristics on serum progesterone levels one day prior to warmed blastocyst transfer.” Reproductive Biology and Endocrinology 21, no. 1 (2023): 86.
    View at Publisher | View at Google Scholar
34. Panaino, Tatiana R., Joyce B. da Silva, Maria Augusta T. de Lima, Paloma Lira, Patricia C. Arêas, Ana Cristina A. Mancebo, Marcelo M. de Souza, Roberto A. Antunes, and Maria do Carmo B. de Souza. “High Progesterone levels in the beginning of ICSI antagonist cycles and clinical pregnancy: still a concern?.” JBRA Assisted Reproduction 21, no. 1 (2017): 11.
    View at Publisher | View at Google Scholar
35. Mumusoglu, Sezcan, Mehtap Polat, Irem Yarali Ozbek, Gurkan Bozdag, Evangelos G. Papanikolaou, Sandro C. Esteves, Peter Humaidan, and Hakan Yarali. “Preparation of the endometrium for frozen embryo transfer: a systematic review.” Frontiers in endocrinology 12 (2021): 688237.
    View at Publisher | View at Google Scholar
36. Huang, Ting-Chi, William Hao-Yu Lee, Mei-Zen Huang, Kuan- Hao Tsui, Chuang-Yen Huang, Gwo-Jang Wu, Mei-Jou Chen et al. “Spontaneous ovulation, hormonal profiles, and the impact of progesterone timing variation on outcomes in natural proliferative phase frozen embryo transfer cycles with single euploid blastocyst transfer.” Journal of Ovarian Research 18, no. 1 (2025): 154.
    View at Publisher | View at Google Scholar
37. Liu, Xitong, Wentao Li, Wen Wen, Ting Wang, Tao Wang, Ting Sun, Na Zhang et al. “Natural cycle versus hormone replacement therapy as endometrial preparation in ovulatory women undergoing frozen-thawed embryo transfer: The compete open-label randomized controlled trial.” PLoS medicine 22, no. 6 (2025): e1004630.
    View at Publisher | View at Google Scholar
38. Groenewoud, Eva R., Nick S. Macklon, Ben J. Cohlen, Amani Al-Oraiby, Egbert A. Brinkhuis, Frank JM Broekmans, Jan-Peter de Bruin et al. “The effect of elevated progesterone levels before HCG triggering in modified natural cycle frozen-thawed embryo transfer cycles.” Reproductive biomedicine online 34, no. 5 (2017): 546-554.
    View at Publisher | View at Google Scholar
39. Kahraman, Semra, and Yucel Sahin. “Is there a critical LH level for hCG trigger after the detection of LH surge in modified natural frozen-thawed single blastocyst transfer cycles?.” Journal of Assisted Reproduction and Genetics 37, no. 12 (2020): 3025-3031.
    View at Publisher | View at Google Scholar
40. Hsueh, Ya-Wen, Chien-Chu Huang, Shuo-Wen Hung, Chia- Wei Chang, Hsi-Chen Hsu, Tung-Chuan Yang, Wu-Chou Lin, Shan-Yu Su, and Hsun-Ming Chang. “Finding of the optimal preparation and timing of endometrium in frozen-thawed embryo transfer: a literature review of clinical evidence.” Frontiers in Endocrinology 14 (2023): 1250847.
    View at Publisher | View at Google Scholar
41. Lee, Jacqueline C., Natalia S. Calzada-Jorge, Heather S. Hipp, and Jennifer F. Kawwass. “Natural cycle frozen embryo transfer: a survey of current assisted reproductive technology practices in the US.” Journal of Assisted Reproduction and Genetics 40, no. 4 (2023): 891-899.
    View at Publisher | View at Google Scholar
42. Mackens, S., A. Stubbe, S. Santos-Ribeiro, L. Van Landuyt, A. Racca, C. Roelens, M. Camus et al. “To trigger or not to trigger ovulation in a natural cycle for frozen embryo transfer: a randomized controlled trial.” Human Reproduction 35, no. 5 (2020): 1073-1081.
    View at Publisher | View at Google Scholar
43. Ranisavljevic, Noemie, Marine Bonneau, Nathalie Rougier, Samir Hamamah, Tal Anahory, Chris Serand, and Stephanie Huberlant. “Ovulation trigger versus spontaneous luteinizing hormone surge on live birth rate after frozen embryo transfer in a natural cycle: a randomized controlled trial.” Fertility and Sterility 123, no. 4 (2025): 718-720.
    View at Publisher | View at Google Scholar
44. Saupstad, Marte, Sara J. Bergenheim, Jeanette W. Bogstad, Morten R. Petersen, Anna Klajnbard, Lisbeth Prætorius, Nina lC Freiesleben et al. “Progesterone concentrations on blastocyst transfer day in modified natural cycle frozen embryo transfer cycles.” Reproductive BioMedicine Online 49, no. 1 (2024): 103862.
    View at Publisher | View at Google Scholar
45. Saupstad, M., S. Bergenheim, C. Colombo, J. W. Bogstad, A. Klajnbard, N. LC Freiesleben, B. Alsbjerg et al. “O-204 Optimal timing and endometrial preparation in modified natural cycle (mNC) frozen embryo transfer (FET) cycles: The FET OPTIMIZING randomised controlled trial.” Human Reproduction 39, no. Supplement_1 (2024): deae108-237.
    View at Publisher | View at Google Scholar
46. An, Belinda Gia Linh, Michael Chapman, Liza Tilia, and Christos Venetis. “Is there an optimal window of time for transferring single frozen-thawed euploid blastocysts? A cohort study of 1170 embryo transfers.” Human Reproduction 37, no. 12 (2022): 2797-2807.
    View at Publisher | View at Google Scholar
47. Pereira, María M., Monica Mainigi, and Jerome F. Strauss III. “Secretory products of the corpus luteum and preeclampsia.” Human Reproduction Update 27, no. 4 (2021): 651-672.
    View at Publisher | View at Google Scholar
48. Lawrenz, B., B. Ata, E. Kalafat, R. Del Gallego, S. Selim, J. Edades, and H. Fatemi. “Undetected, natural conception pregnancies in luteal phase stimulations—case series and review of literature.” Human Reproduction 39, no. 10 (2024): 2268- 2273.
    View at Publisher | View at Google Scholar
49. Basnayake, Surabhi Kumble, Michelle Volovsky, Luk Rombauts, Tiki Osianlis, Beverley Vollenhoven, and Martin Healey. “Progesterone concentrations and dosage with frozen embryo transfers–What’s best?.” Australian and New Zealand Journal of Obstetrics and Gynaecology 58, no. 5 (2018): 533-538.
    View at Publisher | View at Google Scholar
50. Gaggiotti-Marre, Sofia, Manuel Álvarez, Iñaki González-Foruria, Mònica Parriego, Sandra Garcia, Francisca Martínez, Pedro N. Barri, Nikolaos P. Polyzos, and Buenaventura Coroleu. “Low progesterone levels on the day before natural cycle frozen embryo transfer are negatively associated with live birth rates.” Human Reproduction 35, no. 7 (2020): 1623-1629.
    View at Publisher | View at Google Scholar
51. Wånggren, Kjell, M. Dahlgren Granbom, Stavros I. Iliadis, J. Gudmundsson, and Anneli Stavreus-Evers. “Progesterone supplementation in natural cycles improves live birth rates after embryo transfer of frozen-thawed embryos—a randomized controlled trial.” Human Reproduction 37, no. 10 (2022): 2366- 2374.
    View at Publisher | View at Google Scholar
52. Zaat, T. R., E. B. Kostova, P. Korsen, M. G. Showell, F. Mol, and M. Van Wely. “Obstetric and neonatal outcomes after natural versus artificial cycle frozen embryo transfer and the role of luteal phase support: a systematic review and meta-analysis.” Human reproduction update 29, no. 5 (2023): 634-654.
    View at Publisher | View at Google Scholar
53. Zhang, Jie, Mengjie Wei, Xuejiao Bian, Ling Wu, Shuo Zhang, Xiaoyan Mao, and Bian Wang. “Letrozole-induced frozen embryo transfer cycles are associated with a lower risk of hypertensive disorders of pregnancy among women with polycystic ovary syndrome.” American Journal of Obstetrics and Gynecology 225, no. 1 (2021): 59-e1.
    View at Publisher | View at Google Scholar
54. Godiwala, Prachi, Reeva Makhijani, Alison Bartolucci, Daniel Grow, John Nulsen, Claudio Benadiva, James Grady, and Lawrence Engmann. “Pregnancy outcomes after frozen-thawed embryo transfer using letrozole ovulation induction, natural, or programmed cycles.” Fertility and Sterility 118, no. 4 (2022): 690-698.
    View at Publisher | View at Google Scholar
55. Lou, Liqun, Yisong Xu, Mu Lv, Juanjuan Yu, Qimeng Xiao, Peiqin Chen, Mingzhu Bai, and Zhenbo Zhang. “Comparison of different endometrial preparation protocols on frozen embryo transfer pregnancy outcome in patients with normal ovulation.” Reproductive BioMedicine Online 45, no. 6 (2022): 1182-1187.
    View at Publisher | View at Google Scholar
56. Geysenbergh, B., S. Debrock, S. Lie Fong, A. Vanhie, C. Tomassetti, C. Bafort, L. Meeuwis et al. “O-186 Natural versus artificial cycle for endometrial preparation in frozen embryo transfer cycles in ovulatory women: a randomized controlled trial.” Human Reproduction 40, no. Supplement_1 (2025): deaf097-186.
    View at Publisher | View at Google Scholar
57. Bourdon, Mathilde, Pietro Santulli, Fleur Kefelian, Laurine Vienet- Legue, Chloé Maignien, Khaled Pocate-Cheriet, Jacques de Mouzon, Louis Marcellin, and Charles Chapron. “Prolonged estrogen (E2) treatment prior to frozen-blastocyst transfer decreases the live birth rate.” Human Reproduction 33, no. 5 (2018): 905-913.
    View at Publisher | View at Google Scholar
58. Rodríguez-Varela, Cristina, Maria Salvaleda-Mateu, and Elena Labarta. “Duration of oestrogen exposure does not affect reproductive outcome in artificial cycles: a retrospective analysis of more than 7000 hormonal replacement therapy cycles for an embryo transfer.” Frontiers in Endocrinology 14 (2023): 1233685.
    View at Publisher | View at Google Scholar
59. Alsbjerg, Birgit, Mette Brix Jensen, Helle Olesen Elbaek, Rita Laursen, Betina Boel Povlsen, Richard Anderson, Harkan Yarali, and Peter Humaidan. “Midluteal serum estradiol levels are associated with live birth rates in hormone replacement therapy frozen embryo transfer cycles: a cohort study.” Fertility and Sterility 121, no. 6 (2024): 1000-1009.
    View at Publisher | View at Google Scholar
60. Bulletti, Carlo, Francesco Maria Bulletti, Romualdo Sciorio, and Maurizio Guido. “Progesterone: the key factor of the beginning of life.” International journal of molecular sciences 23, no. 22 (2022): 14138.
    View at Publisher | View at Google Scholar
61. Labarta, Elena, Giulia Mariani, Stefania Paolelli, Cristina Rodriguez- Varela, Carmina Vidal, Juan Giles, José Bellver et al. “Impact of low serum progesterone levels on the day of embryo transfer on pregnancy outcome: a prospective cohort study in artificial cycles with vaginal progesterone.” Human Reproduction 36, no. 3 (2021): 683-692.
    View at Publisher | View at Google Scholar
62. Labarta, Elena, Giulia Mariani, Cristina Rodríguez-Varela, and Ernesto Bosch. “Individualized luteal phase support normalizes live birth rate in women with low progesterone levels on the day of embryo transfer in artificial endometrial preparation cycles.” Fertility and Sterility 117, no. 1 (2022): 96-103.
    View at Publisher | View at Google Scholar
63. Mackens, Shari, Francisca Pais, Panagiotis Drakopoulos, Samah Amghizar, Caroline Roelens, Lisbet Van Landuyt, Herman Tournaye, Michel De Vos, and Christophe Blockeel. “Individualized luteal phase support using additional oral dydrogesterone in artificially prepared frozen embryo transfer cycles: is it beneficial?.” Reproductive BioMedicine Online 46, no. 6 (2023): 939-945.
    View at Publisher | View at Google Scholar
64. Loreti, Sara, Caroline Roelens, Panagiotis Drakopoulos, Neelke De Munck, Herman Tournaye, Shari Mackens, and Christophe Blockeel. “Circadian serum progesterone variations on the day of frozen embryo transfer in artificially prepared cycles.” Reproductive BioMedicine Online 48, no. 1 (2024): 103601.
    View at Publisher | View at Google Scholar
65. Vidal, Angela, Carolin Dhakal, Nathalie Werth, Jürgen Michael Weiss, Dirk Lehnick, and Alexandra Sabrina Kohl Schwartz. “Supplementary dydrogesterone is beneficial as luteal phase support in artificial frozen-thawed embryo transfer cycles compared to micronized progesterone alone.” Frontiers in Endocrinology 14 (2023): 1128564.
    View at Publisher | View at Google Scholar
66. Yan, Meng-Han, Zhen-Gao Sun, and Jing-Yan Song. “Dual trigger for final oocyte maturation in expected normal responders with a high immature oocyte rate: a randomized controlled trial.” Frontiers in Medicine 10 (2023): 1254982.
    View at Publisher | View at Google Scholar
67. Li, Xiang, Youman Zeng, Juan He, Bowen Luo, Xiongcai Lu, Lingling Zhu, Zengyu Yang, Fuman Cai, Sheng-ao Chen, and Yudi Luo. “The optimal frozen embryo transfer strategy for the recurrent implantation failure patient without blastocyst freezing: thawing day 3 embryos and culturing to day 5 blastocysts.” Zygote 31, no. 6 (2023): 596-604.
    View at Publisher | View at Google Scholar
68. Glujovsky, Demián, Romina Pesce, Carlos Sueldo, Andrea Marta Quinteiro Retamar, Roger J. Hart, and Agustín Ciapponi. “Endometrial preparation for women undergoing embryo transfer with frozen embryos or embryos derived from donor oocytes.” Cochrane database of systematic reviews 10 (2020).
    View at Publisher | View at Google Scholar
69. Ho, Vu NA, Toan D. Pham, Nam T. Nguyen, Rui Wang, Robert J. Norman, Ben W. Mol, Tuong M. Ho, and Lan N. Vuong. “Livebirth rate after one frozen embryo transfer in ovulatory women starting with natural, modified natural, or artificial endometrial preparation in Viet Nam: an open-label randomised controlled trial.” The Lancet 404, no. 10449 (2024): 266-275.
    View at Publisher | View at Google Scholar
70. Vinsonneau, L., J. Labrosse, G. Porcu-Buisson, N. Chevalier, J. Galey, N. Ahdad, J. P. Ayel et al. “Impact of endometrial preparation on early pregnancy loss and live birth rate after frozen embryo transfer: a large multicenter cohort study (14 421 frozen cycles).” Human Reproduction Open 2022, no. 2 (2022): hoac007.
    View at Publisher | View at Google Scholar
71. Craciunas, Laurentiu, Ioannis Gallos, Justin Chu, Tom Bourne, Siobhan Quenby, Jan J. Brosens, and Arri Coomarasamy. “Conventional and modern markers of endometrial receptivity: a systematic review and meta-analysis.” Human reproduction update 25, no. 2 (2019): 202-223.
    View at Publisher | View at Google Scholar
72. Zhang, Jie, Hongfang Liu, Yun Wang, Xiaoyan Mao, Qiuju Chen, Yong Fan, Yitao Xiao, and Yanping Kuang. “Letrozole use during frozen embryo transfer cycles in women with polycystic ovary syndrome.” Fertility and sterility 112, no. 2 (2019): 371-377.
    View at Publisher | View at Google Scholar
73. Wang, Xiaojuan, Yuan Li, Hongzhuan Tan, Sufen Cai, Shujuan Ma, Yangqin Peng, Hui Guo et al. “Letrozole-stimulated endometrial preparation protocol is a superior alternative to hormone replacement treatment for frozen embryo transfer in women with polycystic ovary syndrome, a cohort study.” Reproductive Biology and Endocrinology 21, no. 1 (2023): 101.
    View at Publisher | View at Google Scholar
74. Riegler, Michael Alexander, Mette Haug Stensen, Oliwia Witczak, Jorunn Marie Andersen, S. A. Hicks, Hugo Lewi Hammer, Erwan Delbarre et al. “Artificial intelligence in the fertility clinic: status, pitfalls and possibilities.” Human Reproduction 36, no. 9 (2021): 2429-2442.
    View at Publisher | View at Google Scholar
75. Ben-Meir, Assaf, and Natali Schachter-Safrai. “Artificial Intelligence in Reproductive Medicine.” In Hot Topics in Human Reproduction: Ethics, Law and Society, pp. 171-180. Cham: Springer International Publishing, 2023.
    View at Publisher | View at Google Scholar
76. Doultani, Shilpa, Prachi Sharma, Prateek Makwana, S. Patil, S. Layek, L. George, H. Highland, and K. Hadiya. “AI in Reproductive Biology: Transforming Fertility Assessment, ART, and Research.” Annual Research & Review in Biology 39, no. 9 (2024): 147-158.
    View at Publisher | View at Google Scholar
77. Lacconi, Valentina, Micol Massimiani, Ilenia Carriero, Claudia Bianco, Carlo Ticconi, Valentina Pavone, Alessandra Alteri et al. “When the embryo meets the endometrium: identifying the features required for successful embryo implantation.” International Journal of Molecular Sciences 25, no. 5 (2024): 2834.
    View at Publisher | View at Google Scholar
78. Bashiri, Zahra, Azita Afzali, Morteza Koruji, Hossein Torkashvand, Mehrdad Ghorbanlou, Nadia Sheibak, Zahra Zandieh, and Fatemehsadat Amjadi. “Advanced strategies for single embryo selection in assisted human reproduction: A review of clinical practice and research methods.” Clinical and Experimental Reproductive Medicine 52, no. 1 (2024): 8.
    View at Publisher | View at Google Scholar
79. Greenwald, Noah F., Geneva Miller, Erick Moen, Alex Kong, Adam Kagel, Thomas Dougherty, Christine Camacho Fullaway et al. “Whole-cell segmentation of tissue images with human- level performance using large-scale data annotation and deep learning.” Nature biotechnology 40, no. 4 (2022): 555- 565.
    View at Publisher | View at Google Scholar
80. Yamaguchi, Manako, Kosuke Yoshihara, Kazuaki Suda, Hirofumi Nakaoka, Nozomi Yachida, Haruka Ueda, Kentaro Sugino et al. “Three-dimensional understanding of the morphological complexity of the human uterine endometrium.” Iscience 24, no. 4 (2021).
    View at Publisher | View at Google Scholar
81. Gordts, S., G. Grimbizis, V. Tanos, P. Koninckx, and R. Campo. “Junctional zone thickening: an endo-myometrial unit disorder.” Facts, Views & Vision in Obgyn 15, no. 4 (2023): 309.
    View at Publisher | View at Google Scholar
82. Schuff, Ricardo H. Asch, Tamar Alkon, Marlene L. Zamora Ramirez, Jorge Suarez, and Nicolas Laugas. “Ultrasonographic endometrial classification in In Vitro Fertilization: a new approach.” Journal of IVF-Worldwide 1, no. 1-3 (2023).
    View at Publisher | View at Google Scholar
83. Singh, Satya P., Lipo Wang, Sukrit Gupta, Haveesh Goli, Parasuraman Padmanabhan, and Balázs Gulyás. “3D deep learning on medical images: a review.” Sensors 20, no. 18 (2020): 5097.
    View at Publisher | View at Google Scholar
84. Orovou, Eirini, Katerina D. Tzimourta, Maria Tzitiridou-Chatzopoulou, Antigoni Kakatosi, Antigoni Sarantaki, EIRINI OROVOU, and Antigoni D. Kakatosi. “Artificial intelligence in assisted reproductive technology: a new era in fertility treatment.” Cureus 17, no. 4 (2025).
    View at Publisher | View at Google Scholar
85. Suarez J. Advancements in Artificial Intelligence in the study of Endometrium. 2024 Aug 2(4):212-221.
    View at Publisher | View at Google Scholar
86. Wells, Cara, Cameron Hayden, Michael Rea, Ben Cook, and Russell Killingsworth. “Advancing bovine embryo evaluation: machine learning to assess embryos in routine embryo transfer practice.” Journal of IVF-Worldwide 3, no. 3 (2025): 32-44.
    View at Publisher | View at Google Scholar
87. Li, Junfeng, Hang Xing, Jing Zhao, Yuan Chen, Yuqing Zhang, Alix Hamon, Rongxiang Li, Shaozhe Yang, and Xiuhong Fu. “Enhancing frozen-thawed embryo transfer outcomes and treatment personalization through machine learning models.” Journal of Assisted Reproduction and Genetics (2025): 1-17.
    View at Publisher | View at Google Scholar
88. Ghaffari, Aida, and Solmaz Sohrabei. “Investigating artificial intelligence in predicting and evaluating sperm and embryo quality in the in vitro fertilization (IVF): a systematic review.” Discover Artificial Intelligence 5, no. 1 (2025): 1-14.
    View at Publisher | View at Google Scholar
89. Wei, Qun, Xia Hong, Yu He, Meiling Xu, Xuechun Jiang, Xuling Shen, and Zilian Wang. “A Qualitative Exploration of Emotional Experiences in Patients with Thin Endometrium Undergoing Repeated Cancellations of Frozen-Thawed Embryo Transfer Cycles.” International Journal of Women’s Health (2025): 127- 137.
    View at Publisher | View at Google Scholar
90. Kolokythas, Argyrios, and Michael H. Dahan. “Is Artificial Intelligence (AI) currently able to provide evidence-based scientific responses on methods that can improve the outcomes of embryo transfers? No.” JBRA Assisted Reproduction 28, no. 4 (2024): 629.
    View at Publisher | View at Google Scholar
91. Li, Li, Dan-Dan Gao, Yi Zhang, Jing-Yan Song, and Zhen-Gao Sun. “Comparison of stimulated cycles with low dose r-FSH versus hormone replacement cycles for endometrial preparation prior to frozen-thawed embryo transfer in young women with polycystic ovarian syndrome: a single-center retrospective cohort study from China.” Drug design, development and therapy (2021): 2805-2813.
    View at Publisher | View at Google Scholar
92. Vuong, Lan N., Toan D. Pham, Khanh TQ Le, Trung T. Ly, Ho L. Le, Diem TN Nguyen, Vu NA Ho et al. “Micronized progesterone plus dydrogesterone versus micronized progesterone alone for luteal phase support in frozen-thawed cycles (MIDRONE): a prospective cohort study.” Human Reproduction 36, no. 7 (2021): 1821-1831.
    View at Publisher | View at Google Scholar