- Research Article
- 10.1016/j.rbmo.2025.105362
Can machine learning models predict oocyte yield during assisted conception?: a systematic review.
- Apr 01, 2026
- Reproductive biomedicine online
- Jessica Wilkinson + 9 more +9
Publications from 2021 to 2026
Showing 10 of 92 papers
Can machine learning models predict oocyte yield during assisted conception?: a systematic review.
Global trends in the uptake of specialist diagnostic ultrasound and MRI scans for endometriosis: An international cross‐sectional survey
Endometriosis affects 1 in 10 women globally. We aimed to determine health provider involvement and preferred methods of obtaining an endometriosis diagnosis across international jurisdictions. A global cross‐sectional survey, distributed through formal and informal FIGO Reproductive Medicine, Endocrinology, and Infertility (REI) committee networks from September through October 2024. A total of 445 respondents from 76 countries began the survey, with 360 completions. Of the respondents, 63.9% were female, with most being gynecologists. Academics and/or researchers comprised 6.7%, with 39.2% involved in research in some capacity. Endometriosis was largely diagnosed by gynecologists (95.0%), with most respondents overall describing ultrasound as the most common method of diagnosis (77.8%). Of those surveyed, 93.6% stated clinicians in their country used non‐invasive imaging (predominantly ultrasound) to diagnose endometriosis before surgery. One third of respondents did not feel confident distinguishing between routine gynecological scans and specialist endometriosis imaging. Significant barriers for uptake of imaging diagnostics included access to training, cost, and geographical accessibility. Regarding adoption of non‐invasive imaging in clinical care, “game‐changing” effects were described, including improvements in surgical preoperative mapping, and patient‐centered care, with concerns that it could increase clinical workloads. Significantly, shifts in endometriosis diagnosis, from visualization of lesions at laparoscopy to non‐invasive imaging diagnostics, were observed internationally. Policies should aim to overcome barriers to uptake of imaging‐based diagnostics by supporting training initiatives, reducing associated costs, and improving geographical accessibility. Respondents predicted transformative changes in endometriosis care once non‐invasive endometriosis diagnostics are effectively adopted.
Read moreAlignment among the zygotic cleavage plane, pronuclear axis, and polar axis predicts live birth outcome of blastocyst
Abstract Purpose To investigate whether alignment among zygotic cleavage plane (CP), pronuclear (PN) axis and polar axis (PB) is associated with higher live birth rate (LBR) in single transferred blastocysts? Methods A total of 103 fresh and 138 frozen blastocysts created via either conventional IVF insemination or intracytoplasmic sperm injection (ICSI) using autologous oocytes at between September 2022 and December 2023 were retrospectively analyzed. Live birth rates were compared according to alignment amongst three zygotic axes. The PN axis was defined by the longitudinal axis before fading and the PB axis was determined by the position of second polar body. Alignment relative to the zygotic CP was classified as CPPN + or CPPN- for the PN axis, and CPPB + or CPPB- for the PB axis. Results In the fresh transfer dataset, CPPN + blastocysts had a significantly higher LBR than CPPN- blastocysts (61.8% vs 12.5%, P < 0.001), while CPPB + outperformed CPPB- (49.3% vs 13.3%, P < 0.001). The highest LBR was observed in CPPN+/CPPB + blastocysts (77.5%), compared with CPPN+/CPPB- (20.0%, P < 0.001), CPPN-/CPPB+ (15.2%, P < 0.001) and CPPN-/CPPB- (6.7%, P < 0.001). Similar patterns were observed in the frozen transfer dataset, where CPPN+/CPPB + blastocysts again had the highest LBR (48.3%), compared with CPPN+/CPPB- (16.1%, P = 0.003), CPPN-/CPPB+ (19.4%, P = 0.007) and CPPN-/CPPB2- (11.1%, P = 0.005). Logistic regression confirmed CPPN and CPPB alignment as independent predictors of live birth in both datasets, after adjusting for several potential confounders. Conclusion Our findings support zygotic alignment among CP, PN and PB alxes as a potential viability marker for blastocyst selection.
Read morePolicy solutions to improve access to fertility treatment and optimise patient care: consensus from an expert forum
BackgroundInfertility is an underrecognized disease which affects over 17% of the reproductive age population worldwide. However, availability of, and access to, assisted reproductive technology (ART) is variable across countries. There are significant challenges relating to awareness, financial and other barriers to care, cultural considerations, and the level of support provided to people undergoing care. Previous studies have explored these challenges, but less attention has been given to the policy implications. As the need for fertility care rises, we investigate the evidence that policy changes can be implemented to improve access to ART treatment.MethodsA review of literature was conducted on fertility policy challenges and developments, covering fertility recognition and awareness; cultural and religious considerations; and access to ART treatment, psycho-social care, and supplementary care. Nine medical and academic experts were invited to validate secondary research findings and provide their perspectives on policy implications. The experts covered different specialties and geographic expertise. Experts participated in individual 60-minute interviews, then a half-day Policy Forum discussion was held virtually in May 2023.ResultsLack of recognition of infertility as a disease, low financial coverage of fertility services, limited psychosocial support, and cultural considerations are substantial barriers to fertility services access. Some countries have limited reimbursement of services or offer only private care, significantly limiting treatment access. Others restrict reimbursement based on age, gender and family status, which creates access inequities. Policy action is needed to mitigate these challenges and to ensure timely and equitable access to fertility care. Decision-makers need to collectively recognize infertility as a disease, rather than just a social issue. Equity of access to infertility services should be ensured by expanding the availability of public funding, along with review and rationalisation of criteria for treatment reimbursement. To improve engagement in treatment and support through the fertility journey, access to psychosocial care should be expanded and included as a core service.ConclusionMajor obstacles to accessing ART treatment have been identified across regions globally, highlighting the urgent need for national policy action to enhance care quality by reviewing current legislation, improving patient and physician education, refining reimbursement procedures, and expanding psychosocial support services.
Read moreCell-free miRNA-150-5p serves as a biomarker and regulator of PROM-related preterm labor by targeting chorionic ADAM19.
The prediction and treatment of spontaneous preterm labor (sPTL) are critical challenges due to obscure etiology and lack of highly specific and sensitive testing methods. Here, we extended the application of noninvasive prenatal testing into the field of sPTL prediction using high-throughput small RNA sequencing to screen the potential biomarkers for sPTL in maternal peripheral blood. We found that hsa-miR-150-5p and hsa-miR-512-3p decreased in the patients affected by sPTL, compared with either term labor or term not labor patients. The change of hsa-miR-150-5p is validated using quantitative PCR with the area under the receiver operating characteristic curve (AUROC) of hsa-miR-150-5p around 0.8272. In the discovery set, hsa-miR-150-5p exhibited an AUROC of approximately 0.8508, which was validated in an independent cohort, accurately classifying preterm samples with an AUROC of 0.8010. Moreover, we showed miR-150-5p inhibited migration and invasion of chorionic cells by directly targeting a disintegrin and metalloproteinase 19 (ADAM19). The significant increase of ADAM19 in chorion from patients affected by sPTL further indicates its inverse correlationship with miR-150-5p. ADAM19 functions as a sheddase of membrane-bound TNF-α to release the TNF-α trimer into the extracellular environment, reciprocally inducing the expression of ADAM19 to form a regenerative cycle and augmenting the migration and invasion of fetal membrane cells. Remarkably, the predictive efficiency of miR-150-5p for premature rupture of membranes (PROMs)-related sPTL sharply rose to 94.21%, indicating its potential as a biomarker for preterm premature rupture of membranes (pPROMs). These findings establish miR-150-5p both as a promising noninvasive biomarker for identifying the risk of sPTL and a key regulator in pathogenesis of pPROM.NEW & NOTEWORTHY MiR-150-5p is identified as a potential noninvasive biomarker associated with sPTL and regulates chorionic cell migration, invasion, and adhesion by targeting ADAM19, a member of the ADAM family, contributing to premature rupture of fetal membranes. These findings provide new insights into the molecular mechanisms of sPTL and suggest that targeting the miR-150-5p/ADAM19 axis may offer novel therapeutic strategies.
Read moreHUWE1 is involved in Sertoli cell polarity establishment by ubiquitination mediated degradation of WT1
HUWE1, a member of HECT E3 ubiquitin ligase family, is implicated in a variety of cellular processes. Recent studies find that HUWE1 also plays critical roles in germ cell development and inactivation of HUWE1 causes germ cell loss in both male and female mice. In this study, we found that Huwe1 was also highly expressed in testicular Sertoli cells. Inactivation of Huwe1 in Sertoli cells resulted in loss of cell polarity, which in turn caused germ cells loss and male infertility. Further study revealed that dysregulation in the expression of cytoskeletal and adhesion-related molecules, as well as a significant increase in EMT-related trans-factors SNAI1&2 in Huwe1-deficient Sertoli cells. Intriguingly, the protein level of WT1 was significantly increased in Huwe1-deficient Sertoli cells, and overexpression of Wt1 in Sertoli cells also caused the defects in spermatogenesis which was consistent with Huwe1 CKO mouse model. Furthermore, the defect of spermatogenesis in Huwe1 CKO mice was partially rescued by deleting one allele of Wt1 gene. Mechanistic studies revealed that WT1 interacts with HUWE1 protein and it could be ubiquitinated by HUWE1. Our study demonstrates that HUWE1 is involved in the establishment of Sertoli cell polarity mainly by regulating the protein level of WT1 gene.
Read moreO-009 OvoVid Phase 2: Role of micronutrients and probiotics supplementation on reproductive outcomes in an oocyte donation program: multicentric, double-blind, randomized, controlled trial
Abstract Study question Does the administration of micronutrients and probiotics supplement during an oocyte donation program improve fertilization rate, blastocyst development, and pregnancy outcomes? Summary answer Micronutrients and probiotics supplementation in oocyte donors does not influence fertilization or blastocyst rates, but significantly improved pregnancy outcomes of recipients that used vitrified oocytes. What is known already Micronutrient supplementation could have a positive impact on IVF clinical results. Evidence about its role in improving oocytes’ quality and reproductive outcomes in normoresponders patients are still lacking. The OvoVid double-blind, randomized, controlled trial Phase 1 suggested that a micronutrients-probiotics supplementation enhances oocyte yield and quality in normoresponder oocyte donors. A significantly higher number of total oocytes retrieved and mature MII were achieved in the supplement group, which also showed a better follicular response to controlled ovarian hyperstimualtion (COH) than placebo in term of more follicles at the end of COH, despite a similar initial antral follicle count. Study design, size, duration Multicentric, double-blind, randomized controlled trial involving 130 oocyte donors and 107 recipients during a two-year period involving three Fertility Clinics in Spain. Oocyte donors were randomized in supplement or placebo group, taking one tablet a day at least for 30 days before and during COH. The micronutrients-probiotc supplement was a combination of Myoinositol+d-chiro-inositol (ratio 40:1), folic acid, melatonin, coenzyme Q10, vitamin D3, B6, B12, E, iron, zinc, Lactobacillus Gasseri and Crispatus. Participants/materials, setting, methods The Ovovid Phase 2 study analyzed the reproductive outcomes of 107 oocyte donation treatments in recipients, classified in supplement (n = 62) or placebo (n = 45) group depending on the oocytes’ origin, realized with donors’ oocytes of Ovovid Phase 1. Primary outcomes included fertilization and blastocyst development rates, while secondary outcomes encompassed pregnancy rates and oocyte survival after thawing. Statistical analysis was performed using Fisher’s exact test and Mann-Whitney U test, with significance set at p&lt;0.05. Main results and the role of chance No significant differences were observed in fertilization rate (76,7% vs. 78%, p &gt; 0,05) comparing supplement and placebo group, as well as in blastulation development (43,4% vs. 45,4%, p = 0,61), despite similar number of inseminated oocytes (6,7 ± 1,3 vs. 6,3 ± 1,2, p = 0,13). When vitrified oocytes were used, similar survival rate after thawing were described in the two groups (94,3% vs. 93,8%, p = 0,3). Pregnancy rate was similar regardless of multivitamin-probiotics supplement use (44,4% vs. 68,2%, p = 0,07). However, in the sub-analysis for type of oocyte, pregnancy rates with vitrified oocytes (Supplement n = 24; Placebo n = 18) were significantly higher in the supplement group compared to placebo (50% vs. 0%, p = 0,02). No significant differences were found for fresh oocytes (n = 65; Supplement n = 38; Placebo n = 27) outcomes. Limitations, reasons for caution The study was limited by the relatively small sample size for subgroup analyses and the lack of long-term follow-up for pregnancy outcomes. Further studies are warranted to confirm these observations and optimize supplementation strategies in oocyte donation programs and assisted reproductive technologies. Wider implications of the findings Micronutrients and probiotics supplementation during oocyte donors’ COH did not seem to improve overall fertilization or blastulation rates. However, it may be a valuable adjunct in improving recipients’ pregnancy outcomes with vitrified oocytes, suggesting a role in enhancing oocyte competence during vitrification, potentially optimizing outcomes for specific protocols. Trial registration number No
Read moreP-652 Association between obesity and leptin levels(serum and follicular fluid) in women seeking IVF and how they affect the reproductive outcomes- a prospective observational cohort study
Abstract Study question What is the association of BMI(Body-Mass-Index) with serum and follicular fluid leptin levels and their effect on reproductive outcomes? Summary answer Increasing-BMI has direct correlation with-leptin-levels. However, increasing leptin-levels did-not alter reproductive-outcomes. Use of leptin as biomarker in women of high-BMI to predict-reproductive-outcomes needs further evaluation. What is known already Obesity in women alters reproductive outcomes through menstrual and ovulatory disorders, impaired endometrial development and decreased embryo implantation. Numerous studies investigating the impact of elevated BMI on reproductive outcomes have shown conflicting results. Leptin is a 16-kDa protein hormone released from white adipose tissue. It regulates appetite and body weight. Therefore women with obesity have shown to have higher levels of Leptin. Leptin has been thought to exert its effects on reproduction via-HPO-axis-(Hypothalamo-Pituitary-Ovarian),but there might also be a direct ovarian effect. Literature suggests that high serum and/or follicular fluid leptin concentrations have correlated negatively with IVF(in-Vitro Fertilization) cycle outcomes Study design, size, duration A Prospective-Observational-Cohort-study conducted over a duration of 8months from May2023-December2023 at our center with prior approval from Institutional-Ethical-Committee(IEC)( IEC No: RECR/1312/Inst/TG/2019/021). The sample size of 98 women for this study, which was derived by assuming 80% power of the study and significance level at 5%. Women were divided into groups based on BMI and Leptin levels. Only self-gamete-cycles with female-age in-between 21-37yrs were considered for this study. Cycles with genetic-testing on embryos(PGT) were excluded. Participants/materials, setting, methods All participants underwent controlled-ovarian-stimulation as per-SOP. On the day of oocyte retrieval, their serum/follicular-fluid samples were collected and Leptin-levels were quantified using-Abkine-Leptin-ELISA-kit(Manufacturer’s protocol followed). Normal-Leptin-level-0.5-15.2 ng/ml Higher-Leptin-level-&gt;15.2 ng/ml All participants underwent Frozen Embryo Transfers(FET) with 1or2 vitrified-Warmed good-grade-Blastocysts that showed 100%survival. Participants were divided into the following-BMI-groups Normal BMI(n = 43)-18-24.9 Over-weight BMI(n = 33)-25-29.9 Obese BMI(n = 22) &gt;30 Outcome of their FET was recorded based on BMI and Leptin-levels.. Data was statistically analyzed using SAS software. Main results and the role of chance Effect of BMI on outcomes: Mean duration-of-stimulation across various BMI groups was comparable Mean-dosage-of-Gonadotrophins was higher in BMI&gt;30(pvalue-0.002- Significant)) Mean-oocytes-collected, maturity-rates, Fertilization-rates and Blastocyst-formation-rates were comparable across all BMI-groups. As BMI increased the pregnancy-outcomes decreased as shown below: Normal BMI Vs Overweight Vs Obese BMI-respectively Miscarriage Rate(MR)-0%(Vs)6.6%(Vs)17%(p = 0.001-Significant) IR-63%(Vs)35%(Vs)9%(p = 0.001-Significant) LBR-62.1%(Vs)42.4%(Vs)13.6%(p = 0.006- Significant) As per the data, BMI seems to have a negative-correlation with reproductive-outcomes. Mean levels of Leptin(serum/follicular fluid across BMI groups, Normal BMI Vs Overweight Vs Obese BMI respectively are described below: 14ng/ml(Vs)17.9ng/ml(Vs)19.6 ng/ml(p = 0.001- Significant) As the BMI increases, the serum/follicular-fluid-Leptin-levels also increase and the levels were similar and comparable in both the body-fluid samples(follicular & Serum). Data from this study suggests that serum Leptin levels can be a surrogate marker for Follicular-Leptin-levels Association of Leptin levels and Reproductive outcomes: Normal Leptin levels(Vs)Raised-Leptin respectively shown below: IR-42%(Vs)38%(p = 0.68-NS) MR-14.3%(Vs)12.1%(p = 0.20-NS) LBR-46.2%(Vs)40.3%(p = 0.76-NS) There seemed a poor correlation/association of Leptin levels and reproductive outcomes. The outcomes were comparable between normal and raised-Leptin-levels. Though data from this study infers that there is a direct correlation between BMI and Leptin-levels, there seemed no correlation between Leptin-levels and reproductive-outcomes. Utilising serum/follicular leptin-levels as a biomarker in obese-women to predict-reproductive-outcomes does not seem to help. Limitations, reasons for caution Larger sample size is required to make statistically significant conclusions regarding effect of serum leptin level on IVF success. Other causes of implantation failure like genetic abnormality in embryo, displaced window of implantation etc were not ruled out. Wider implications of the findings Further research needs to be done on the effectiveness of serum leptin level measurement as a predictor of IVF success in obese individuals. Trial registration number No
Read moreP-102 Can embryo biopsy technique in a PGT-A cycle affect foetal well-being in-utero and neo-natal health?
Abstract Study question Can embryo biopsy technique in a PGT-A (Pre-Implantation-Genetic-Testing-for-Aneuploidy) cycle affect foetal well-being in-utero and neo-natal health? Summary answer Embryo-biopsy technique in a PGT-A cycle doesn’t seem to affect foetal well-being in-utero and neo-natal health and is a safer intervention to offer when indicated. What is known already Embryo biopsy for PGT-A is one of the recent-advances applied in the field of assisted-reproduction (ART) to enhance reproductive outcomes. Currently PGT-A has been applied for screening of euploid embryos and also to select and rank embryos for transfer to enhance reproductive outcomes. The technique of biopsy involves removal of Trophectoderm with the help of diode-LASER. There have been conflicting inferences regarding the negative effects of this newer-invasive-technique on foetal well-being and neonatal health and safety. Hence, we have retrospectively looked at PGT-A data for foetal well-being and neonatal health and safety in the last ten years at our clinic Study design, size, duration We retrospectively looked at PGT-A data at our fertility clinic from 2013-2023. Study group -966 PGT-A cycles and transfer of single PGT screened Euploid Blastocyst Control group -3970 Non PGT Cycles and transfer of single non-PGT screened good grade Blastocyst Indication for PGT-A was RIF(Recurrent-Implantation-Failure), RPL(Recurrent-Pregnancy-Loss)& AMA(advanced-Maternal-Age). Female & male age between 27-45yrs. All women in both groups had elective single blastocyst transfer and only singleton pregnancies were evaluated. Participants/materials, setting, methods Study group – Blastocysts biopsied for Trophectoderm and tissue sent to genetic lab for NGS(Next- Generation-Sequencing) screening. Single Euploid blastocyst was transferred in a Frozen-Embryo-Replacement-Cycle(FET). Only embryos with 100% survival post warming were considered for FET. Control Group – ICSI was the choice of insemination and extended blastocyst culture with freeze all policy was adopted. Single good grade blastocyst with 100% survival post warming was transferred in FET cycle. Main results and the role of chance Outcomes measured: Implantation Rate(IR), Miscarriage Rate(MR), Cumulative Live Birth rate(cLBR), Incidence of Placental diseases like PIH(Pregnancy-Induced-Hypertension), IUGR(intra-Uterine-Growth-restriction), Pre-Term Deliveries (we considered only singleton pregnancies), Incidence of Low-birth-weight(&lt;2.5kgs)(LBW), NICU-admission and incidence of congenital-anomalies. Following were the outcomes between PGT and Non-PGT groups respectively: IR:69% Vs 50%(p = 0.0001-Significant) MR:11% Vs 12%(p = 0. 0.3801-Non-Significant) cLBR:61% Vs 54%(P = 0.0001-Significant) PIH:00% Vs 0.050%(P = 0.0001-Significant) IUGR:00% Vs 0.025%(P = 0.0001-Significant) Pre-Term-Deliveries:00% Vs 0.75%(p = 0.3487-Non-Significant) LBW:18% Vs 20%(p = 0.1537-Non-Significant) NICU-admissions:7.09% Vs 9%(p = 0.0504- Significant)( All babies in both arms discharged and sent home with no morbidity) Congenital-Anomalies:0% Vs 0.17%(p = 0.1929-Non-Significant) Chi-square test was used for statistical-analysis. PGT-A group showed higher IR and cLBR, which infers that PGT-A intervention is a beneficial intervention to enhance take home baby rates which is one of the main objective of any ART cycle. Considering that embryo biopsy involves removal of Trophectoderm tissue, there has been a concern that this might lead to poor placentation and might increase placental disease like PIH, IUGR and LBW. Data above indicates that embryo biopsy did not bring about higher incidence of placenta related issues. Lesser NICU admissions and absence of congenital anomalies in PGT-A group emphasize that this intervention in ART seems safe and efficient. PGT-A seems to be non-inferior to non-PGT ART-cycles. Limitations, reasons for caution Retrospective data analysis. Though the data of this study is showing PGT-A as a safe intervention, embryologists need to exercise care and caution while using LASER and perform biopsy with adequate skill and expertise. Hence, embryo biopsy and PGT-A should be offered only when indicated and not for all. Wider implications of the findings PGT-A seems to be beneficial and safe intervention to enhance reproductive outcomes and can be offered to couples where it is indicated. There have been concerns regarding invasiveness of the biopsy technique and use of LASER. Therefore, a well validated non-invasive PGT is the need of the hour. Trial registration number No
Read moreP-348 Artificial intelligence-based machine learning to diagnose and classify adenomyosis from ultrasound scans: a model development study
Abstract Study question Can an image recognition model, utilising automated deep learning, be trained to diagnose and classify the severity of adenomyosis on ultrasound? Summary answer The automated deep learning model developed in this study can accurately diagnose and classify the severity of adenomyosis from ultrasound images. What is known already Adenomyosis is prevalent across all stages of life, starting from adolescence to perimenopause, however it remains underdiagnosed. The diagnosis and classification of adenomyosis are essential for correlating with symptom severity and counselling about disease prognosis, management options and long-term impact. Diagnosing and classifying the severity is a time-consuming and complex process that depends on the operator’s experience. Machine learning (ML) has an enormous potential to assist healthcare professionals with image classification tasks. It can directly process and automatically learn mid to high-level abstract features acquired from images using a deep architecture model without requiring the manual definition of features. Study design, size, duration This observational cohort study utilised ultrasound images from four assisted conception units in the United Kingdom from February 2023 to February 2024. Labelled images were used as input for training and testing of the model. The complete dataset was manually split into 9:1 ratio; 90% of all images were used as training dataset (80% for training and 10% for validation) and 10% as test dataset. Ethics approval was obtained from the Institutional Review Board (IRB). Participants/materials, setting, methods An automated ML tool (Vertex AI Vision) was trained and tested to diagnose and classify adenomyosis severity. Two investigators, experienced in gynaecological ultrasound scanning, labelled two and three-dimensional images into four categories: normal uterus, mild, moderate, and severe adenomyotic uterus, as per standardised ultrasound criteria. The labelled images were formatted and uploaded on Google Cloud to train the neural network based on pattern recognition. The model’s performance was determined for an internal test dataset. Main results and the role of chance A total of 3,500 ultrasound images were obtained from 3,194 patients, and the internal test dataset consisted of 350 ultrasound images from 292 patients. An overall and per-category score threshold of 0.5 was adopted to provide a balance between reasonable precision and recall during the evaluation. The investigator’s categorization was used as the reference standard to evaluate the performance of the automated machine-learning model. The model’s overall accuracy for diagnosing and classifying adenomyosis, as determined by the area under the precision-recall curve, was 80.24% across all score thresholds. The overall precision was 83.00%, and recall was 65.75% at a confidence threshold of 0.5. The precision and recall were 91% and 78% for normal uterus, 67% and 53% for mild adenomyosis, 86% and 68% for moderate adenomyosis, and 88% and 64% for severe adenomyosis. The algorithm achieved a high accuracy for the internal dataset. The time taken by a trained healthcare professional to label and process 3,500 images into 4 categories was 70 hours. The time taken by an untrained automated ML tool to get trained, process, and label the images into 4 categories was 4 hours, saving 66 hours per healthcare professional. Limitations, reasons for caution Due to the inherent nature of non-invasive investigations for the population attending the assisted conception unit, histological confirmation of adenomyosis was not possible. In addition, the model’s performance has been tested using an internal test dataset and needs further validation using an external test dataset from a different population. Wider implications of the findings This novel study leverages the power of artificial intelligence to develop an automated machine learning model for diagnosing and classifying adenomyosis from ultrasound scans. This tool can potentially enhance the diagnostic capacity and efficiency of healthcare professionals and empower women by providing better care through early and uniform diagnosis. Trial registration number No
Read more