Usefulness of embryo evaluation via artificial intelligence–based image analysis

Study question Do artificial intelligence (AI) models used to assess embryo viability (based on pregnancy outcomes) also correlate with known embryo quality measures such as Gardner score? Summary answer An AI for embryo viability assessment also correlated with Gardner score, further substantiating the use of AI for assessment and selection of good quality embryos. What is known already The Gardner score consists of three separate components of embryo morphology that are graded individually, then combined to give a final score describing Day 5 embryo (blastocyst) quality. Evidence suggests the Gardner score has some correlation with clinical pregnancy. We hypothesized that an AI model trained to evaluate likelihood of clinical pregnancy based on fetal heartbeat (in clinical use globally) would also correlate with components of the Gardner score itself. We also compared the ability of the AI and Gardner score to predict pregnancy outcomes. Study design, size, duration This study involved analysis of a prospectively collected dataset of single static Day 5 embryo images with associated Gardner scores and AI viability scores. The dataset comprised time-lapse images of 1,485 embryos (EmbryoScope) from 638 patients treated at a single in vitro fertilization (IVF) clinic between November 2019 and December 2020. The AI model was not trained on data from this clinic. Participants/materials, setting, methods Average patient age was 35.4 years. Embryologists manually graded each embryo using the Gardner method, then subsequently used the AI to obtain a score between 0 (predicted non-viable, unlikely to lead to a pregnancy) and 10 (predicted viable, likely to lead to a pregnancy). Correlation between the AI viability score and Gardner score was then assessed. Main results and the role of chance The average AI score was significantly correlated with the three components of the Gardner score: expansion grade, inner cell mass (ICM) grade, and trophectoderm grade. Average AI score generally increased with advancing blastocyst developmental stage. Blastocysts with expansion grades of ≥ 3 are generally considered suitable for transfer. This study showed that embryos with expansion grade 3 had lower AI scores than those with grades 4-6, consistent with a reduced pregnancy rate. AI correlation with trophectoderm grade was more significant than with ICM grade, consistent with studies demonstrating that trophectoderm grade is more important than ICM in determining clinical pregnancy likelihood. The AI predicted Gardner scores of ≥ 2BB with an accuracy of 71.7% (sensitivity 75.1%, specificity 45.9%), and an AUC of 0.68. However, when used to predict pregnancy outcome, the AI performed 27.9% better than the Gardner score (accuracies of 49.8% and 39.0% respectively). Even though the AI was highly correlated with the Gardner score, the improved efficacy for predicting pregnancy suggests that a) the AI provides an advantage in standardization of scoring over the manual and subjective Gardner method, and b) the AI is likely identifying and evaluating morphological features of embryo quality that are not captured by the Gardner method. Limitations, reasons for caution The Gardner score is not a linear score, creating challenges with setting a suitable threshold relating to the prediction of pregnancy. The 2BB treshold was chosen based on literature (Munné et al 2019) and verified by experienced embryologists. This correlative study may also require additional confirmatory studies on independent datasets. Wider implications of the findings The correlation between AI scores and known features of embryo quality (Gardner score) substantiates the use of the AI for embryo assessment. The AI score provides further insight into components of the Gardner score, and may detect morphological features related to clinical pregnancy beyond those evaluated by the Gardner method. Trial registration number Not applicable

Acknowledgements: We would like to thank the following for their contributions to the 2020 update. All staff at the editorial base of the Cochrane Gynaecology and Fertility Review Group; and in particular, Helen Nagels (Managing Editor) and Information Specialist Marian Showell, for help with the literature searches. Drs Jeanette MacKenzie, Evi Vogiatzi and Rik van Eekelen who provided peer review comments. We also thank all the authors who replied to our email queries about their study: Dr Elisabet Clua, Dr Maria Luisa López‐Regalado and Dr OM Abuzeid. We also thank the following. Zabeena Pandian, who was the lead author of the original review and two subsequent updates. Other co‐authors for their contribution in the previous versions: Allan Templeton, Ozkan Ozturk, Gamal Serour, and Jane Marjoribanks. Statistician Andy Vail (University of Manchester University, UK) for methodological advice.

We aimed to investigate current expectations and clinical adoption of artificial intelligence (AI) software among neuroradiologists in Korea. In April 2022, a 30-item online survey was conducted by neuroradiologists from the Korean Society of Neuroradiology (KSNR) to assess current user experiences, perceptions, attitudes, and future expectations regarding AI for neuro-applications. Respondents with experience in AI software were further investigated in terms of the number and type of software used, period of use, clinical usefulness, and future scope. Results were compared between respondents with and without experience with AI software through multivariable logistic regression and mediation analyses. The survey was completed by 73 respondents, accounting for 21.9% (73/334) of the KSNR members; 72.6% (53/73) were familiar with AI and 58.9% (43/73) had used AI software, with approximately 86% (37/43) using 1-3 AI software programs and 51.2% (22/43) having up to one year of experience with AI software. Among AI software types, brain volumetry software was the most common (62.8% [27/43]). Although 52.1% (38/73) assumed that AI is currently useful in practice, 86.3% (63/73) expected it to be useful for clinical practice within 10 years. The main expected benefits were reducing the time spent on repetitive tasks (91.8% [67/73]) and improving reading accuracy and reducing errors (72.6% [53/73]). Those who experienced AI software were more familiar with AI (adjusted odds ratio, 7.1 [95% confidence interval, 1.81-27.81]; P = 0.005). More than half of the respondents with AI software experience (55.8% [24/43]) agreed that AI should be included in training curriculums, while almost all (95.3% [41/43]) believed that radiologists should coordinate to improve its performance. A majority of respondents experienced AI software and showed a proactive attitude toward adopting AI in clinical practice, suggesting that AI should be incorporated into training and active participation in AI development should be encouraged.

Day 2 embryo transfer (ET) and day 3 ET afford similar reproductive outcomes in the poor responder

Real-World Surveillance of FDA-Cleared Artificial Intelligence Models: Rationale and Logistics.

BackgroundThere has been a resurgence of Artificial Intelligence (AI) on a global scale in recent times, resulting in the development of cutting-edge AI solutions within hospitals. However, this has also led to the creation of isolated AI solutions that are not integrated into clinical IT. To tackle this issue, a clinical Artificial Intelligence (AI) platform that handles the entire development cycle of clinical AI models and is integrated into clinical IT is required. This research investigates the integration of a clinical AI platform into the clinical IT infrastructure. This is demonstrated by outlining the stages of the AI model development cycle within the clinical IT infrastructure, illustrating the interaction between different IT system landscapes within the hospital with BPMN diagrams.MethodsInitially, a thorough analysis of the requirements is conducted to refine the necessary aspects of the clinical AI platform with consideration of the individual aspects of clinical IT. Subsequently, processes representing the entire development cycle of an AI model are identified. To facilitate the architecture of the AI platform, BPMN diagrams of all the identified processes are created. Clinical use cases are used to evaluate the processes using the FEDS framework.ResultsOur BPMN process diagrams cover the entire development cycle of a clinical AI model within the clinical IT. The processes involved are Data Selection, Data Annotation, On-site Training and Testing, and Inference, with distinctions between (Semi-Automated) Batch Inference and Real-Time Inference. Three clinical use cases were assessed to evaluate the processes and demonstrate that this approach covers a wide range of clinical AI use cases.ConclusionsThe evaluations were executed successfully, which indicate the comprehensive nature of our approach. The results have shown that different clinical AI use cases are covered by the BPMN diagrams. Our clinical AI platform is ideally suited for the local development of AI models within clinical IT. This approach provides a basis for further developments, e.g., enabling the training and deployment of an AI model across multiple sites or the integration of security- and privacy-related aspects.

Monozygotic triplet pregnancy following egg donation and transfer of single frozen-thawed embryo

BackgroundStandard morphologic evaluation has been the most widely adopted approach to embryo selection, and remains the most common strategy.The objective of the study to determine the association between the morphologic grading and implantation rate of euploid blastocysts in single frozen-thawed embryo transfer (SET) cycles.MethodsA total of 271 patients aged 20–40 years undergoing euploid SET from January 2017 to December 2019 were included in retrospective cohort study.The cycles were divided into three groups based on their morphologic grading before cryopreservation: good-quality (n = 58), average-quality (n = 88) and poor-quality blastocysts (n = 125). The pregnancy outcome of the three morphologic groups were analyzed and a logistic regression of implantation rate was conducted.ResultsGood-quality blastocysts yielded statistically significantly higher implantation rates than poor-quality (79.31% vs. 48%; P<0.001). Planned subgroup analyses by age and the day of TE biopsy were conducted. Logistic regression analyses that adjusted for these variables identified higher implantation rates (adjusted odds ratio(aOR) = 4.083, 95% confidence interval (CI):1.836–9.082, P<0.001) for the good-quality blastocysts than for those that underwent poor-quality cycles in women aged < 35 years, but not in women aged ≥35 years (aOR = 6.074, 95% CI: 0.456–80.919, P = 0.172). The implantation rates were higher among women with good-quality blastocysts on both Day 5 and Day 6 of TE biopsy than among those with poor-quality blastocysts (Day 5, aOR = 3.294, 95% CI:1.260–8.616, P = 0.015; Day 6, aOR = 4.179, 95% CI:1.004 ~ 17.399, P = 0.049). Day 5 euploid blastocysts had no significant difference in implantation potential and early spontaneous abortion rate compared with similarly graded Day 6 euploid blastocysts.ConclusionsBlastocyst morphologic grading was associated with implantation rate for euploid embryo transfers after adjustment for potential confounders. These findings suggest that evaluating blastocyst morphology is critical when selecting the best euploid blastocyst.

Development and validation of an artificial intelligence system for triple-negative breast cancer identification and prognosis prediction: a multicentre retrospective study

Study objectivesNonrestorative sleep (NRS) has been identified as a potential risk factor for physical and mental well-being in adults, but limited research exists for children and adolescents. This study aimed to clarify the factors associated with NRS in Japanese junior high school students.MethodsThe participants were 529 Japanese junior high school students in grades 7 through 9. Participants were asked to respond to Google Forms, and responses were obtained from 392 students. Sleep habits, history of coronavirus disease (COVID-19), physical symptoms, social isolation, and the presence of the symptoms of restless legs syndrome (RLS) were identified. NRS, insomnia symptoms, and depressive symptoms were evaluated using the Restorative Sleep Questionnaire (RSQ), Athens Insomnia Scale (AIS), and Patient Health Questionnaire (PHQ)-9, respectively. The cut-off value for NRS determination by the RSQ score was the mean of the scores that maximized the sensitivity and specificity sum for detecting participants with AIS and PHQ-9 scores of ≥ 6 and ≥ 5, respectively. NRS-associated sleep parameters and those associated with depressive symptoms were evaluated using binominal logistic regression analysis. Multinomial logistic regression (MLR) analysis was used to confirm the reproducibility of the binomial logistic regression analysis results with lower RSQ scores.ResultsThe NRS group comprised 40.1% of participants and exhibited a higher prevalence of physical and depressive symptoms compared to those with restorative sleep. Logistic regression analysis, adjusted for sex, grade, and COVID-19 history, revealed the following odds ratios (95% confidence interval) for NRS: average total sleep time < 7 h 2.44 (1.16–4.33), AIS ≥ 6 2.74 (1.51–4.95), evening chronotype 2.58 (1.49–4.47), and RLS symptoms 2.21 (1.21–4.03). The same results were obtained using MLR as those obtained via binomial logistic regression analysis. Logistic regression analysis for depressive symptoms revealed that NRS displayed the highest odds ratio (95% CI) of 3.16 (1.90–5.27) among the sleep-related variables.ConclusionsThese findings suggest that NRS in Japanese junior high school students is associated with physical and mental health issues. Intervention and longitudinal studies are warranted to address NRS-associated sleep–wake problems in this age group.Supplementary InformationThe online version contains supplementary material available at 10.1186/s40101-025-00412-8.

ÖZET Amaç: Nadiren doğal gebeliklerde oluşan heterotopik gebelikte sıklıkla rüptür sonrası cerrahi tedavi gerekli olduğunda tanı konur. Bu sebeple erken tanı hasta için hayati önem taşımaktadır. Burada rüptüre tubal gebelik ve intrauterin canlı dikoryonik-diamniyotik ikiz gebelik ile prezente olan heterotopik gebelik vakası sunmayı amaçladık. Olgu: 30 yaşında, nullipar, son adet tarihine göre 6 hafta 3 günlük intruterin inseminasyon ile ( IUI ) gebeliği olan hasta, karın ağrısı ile kliniğimize başvurdu. Yapılan fizik muayenesinde akut batın bulguları mevcuttu. Transvaginal ultrasonografisinde intrauterin yerleşimli 6 hafta 5 gün ve 6 hafta 6 gün ile uyumlu fetal kalp atımı ( FKA ) pozitif olan dikoryonik - diamniyotik ikiz gebelik tespit edildi. Batın içi douglusta yaklaşık 6-7 cm serbest mayi izlendi. Sağ adneksial alanda ektopik odak olabilecek oluşum izlendi. Yapılan tetkiklerinde Hemoglobin: 10.5 gr/dl, β-hcg: 15.000 m IU / ml olarak gelen batın muayenesinde rebaund ve defansı olan hastaya laparotomi planlandı. Laparotomik eksplorasyonda sağ tubal rüptüre ektopik odak izlendi. Batında ortalama 500-600 cc kan ve koagulum izlendi. Uterus ve her iki over normal görünüm ve cesamette izlendi. Sağ total salpenjektomi yapıldı, batın yıkanmasının ardından laparotomiye son verildi. Sağ tubanın patolojik değerlendirme sonucu ektopik gebelik, hematosalpenks olarak geldi. Postoperatif 2. gününde bakılan kontrol ultrason bakısında intrauterin dikoryonik-diamniyotik fetal kalp atımı olan 6 hafta 5 gün ve 6 hafta 6 gün ile uyumlu ikiz gebeliğin devam ettiği görüldü. Hasta postop 4. günde sorunsuz poliklinik kontrolüne gelmek üzere taburcu edildi. Sonuç: Heterotopik gebelik nadir görülmesi sebebi ile tanı ve tedavide gecikmeler olabilmektedir. Özellikle intrauterin gebelik ultrason bakısında görülen ek yakınması olamayan hastalarda ektopik gebelik atlanabilmektedir. Bu vakamızda buna özellikle dikkat çekmek istedik. Anahtar kelimeler: Heterotopik gebelik; ikiz gebelik; tubalektopik gebelik

Pregnancies derived from vitrified human zygotes

Trusted evidence. Informed decisions. Better health.

Blood Oxygenation Level-Dependent MRI for Predicting Lymphovascular Space Invasion in Cervical Squamous Cell Carcinoma Treated with Neoadjuvant Chemotherapy: A Prospective Study.

BackgroundIntramyocardial hemorrhage (IMH) after ST-segment elevation myocardial infarction (STEMI) reperfusion increases the risk of major adverse cardiovascular events (MACE). We aimed to integrate a coronary angiography-derived index of microcirculatory resistance (caIMR) with clinical features in a nomogram model for predicting IMH in patients with STEMI following primary percutaneous coronary intervention (PCI).MethodsA retrospective study included 309 STEMI patients admitted at Xuzhou Medical University for primary PCI from 2022 to 2023 in training and validation cohorts. Their caIMR was calculated from coronary angiography images and IMH was assessed by cardiac Magnetic Resonance (CMR). A nomogram was constructed through logistic regression analyses. Predictive accuracy, calibration, and clinical usefulness were validated by an area under the curve (AUC) of the receiver operating characteristic (ROC) curve, calibration curves, and decision curve analysis (DCA), respectively. All statistical tests were two-sided.ResultsA total of 247 patients were assigned to the training cohort and 62 to the validation cohort randomly. A nomogram was established using two independent predictors of IMH derived from multivariable analysis: caIMR and left ventricular ejection fraction (LVEF). The AUC for the nomogram was 0.844 (95% confidence interval (CI): 0.793–0.896) in the training cohort and 0.834 (95% CI: 0.724–0.944) in the validation cohort. The predicted and actual estimates were significantly correlated in both cohorts, indicating good calibration of the nomogram. The nomogram was clinically useful for the prediction of IMH, within a 10–88% and 10–74% threshold probability in the training and validation cohorts respectively, exhibited by DCA. Furthermore, the nomogram showed superior discriminative ability for MACE prediction compared to CMR-confirmed IMH (AUC 0.726 vs. 0.584, P < 0.001).ConclusionsThis novel caIMR-based nomogram resulted in high accuracy of prediction for IMH and MACE in STEMI patients undergoing primary PCI. It may become a convenient tool to predict IMH immediately after the PCI procedure, and help with prognostic judgment and early clinical intervention.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12872-025-05057-0.