Beyond black-box models: explainable AI for embryo ploidy prediction and patient-centric consultation.

PURPOSE: To determine if an explainable artificial intelligence (XAI) model enhances the accuracy and transparency of predicting embryo ploidy status based on embryonic characteristics and clinical data. METHODS: This retrospective study utilized a dataset of 1908 blastocyst embryos. The dataset includes ploidy status, morphokinetic features, morphology grades, and 11 clinical variables. Six machine learning (ML) models including Random Forest (RF), Linear Discriminant Analysis (LDA), Logistic Regression (LR), Support Vector Machine (SVM), AdaBoost (ADA), and Light Gradient-Boosting Machine (LGBM) were trained to predict ploidy status probabilities across three distinct datasets: high-grade embryos (HGE, n = 1107), low-grade embryos (LGE, n = 364), and all-grade embryos (AGE, n = 1471). The model's performance was interpreted using XAI, including SHapley Additive exPlanations (SHAP) and Local Interpretable Model-agnostic Explanations (LIME) techniques. RESULTS: The mean maternal age was 38.5 ± 3.85 years. The Random Forest (RF) model exhibited superior performance compared to the other five ML models, achieving an accuracy of 0.749 and an AUC of 0.808 for AGE. In the external test set, the RF model achieved an accuracy of 0.714 and an AUC of 0.750 (95% CI, 0.702-0.796). SHAP's feature impact analysis highlighted that maternal age, paternal age, time to blastocyst (tB), and day 5 morphology grade significantly impacted the predictive model. In addition, LIME offered specific case-ploidy prediction probabilities, revealing the model's assigned values for each variable within a finite range. CONCLUSION: The model highlights the potential of using XAI algorithms to enhance ploidy prediction, optimize embryo selection as patient-centric consultation, and provides reliability and transparent insights into the decision-making process.

Feasibility of Weekly Integrated Palliative Care Outcome Scale (IPOS) Within a Stepped Care Model.

CONTEXT: Patients with advanced cancer may experience symptoms and concerns that are inadequately identified by the healthcare team, leading to calls for patient-reported symptom monitoring. OBJECTIVES: Assess the feasibility of administering weekly patient-reported online Integrated Palliative care Outcome Scale (IPOS) questionnaires within the context of a stepped care model in the outpatient care setting. METHODS: Analysis of intervention group data in a randomized controlled trial to assess the effectiveness of a stepped care model of palliative care for patients with advanced cancer. Patients in the intervention group were invited to complete the IPOS weekly for 16 weeks through a remotely-administered online questionnaire. At the end of the 16-week period, patients were invited to complete a feedback survey. Multivariable logistic regression was used to assess factors associated with more versus less than 70% completion of weekly questionnaires. RESULTS: Among 111 patients who survived more than 16 weeks, the mean number of questionnaires completed was 9.2/16 (58%). A total of 53 out of the 111 patients (48%) completed more than 70% of the questionnaires. Higher education level was found to be associated with higher completion of the questionnaires. A total of 79 out of 111 (71%) patients responded to the feedback survey, of which 67 (85%) felt comfortable with completing the online questionnaire and 11 (14%) felt it was troublesome to complete it on a weekly basis. CONCLUSION: In our study, there was suboptimal completion of patient-reported IPOS questionnaire. Further research is needed to improve the uptake of patient-reported outcomes in real-world clinical settings.

DNA Methylation of Window of Implantation Genes in Cervical Secretions Predicts Ongoing Pregnancy in Infertility Treatment.

Window of implantation (WOI) genes have been comprehensively identified at the single cell level. DNA methylation changes in cervical secretions are associated with in vitro fertilization embryo transfer (IVF-ET) outcomes. Using a machine learning (ML) approach, we aimed to determine which methylation changes in WOI genes from cervical secretions best predict ongoing pregnancy during embryo transfer. A total of 2708 promoter probes were extracted from mid-secretory phase cervical secretion methylomic profiles for 158 WOI genes, and 152 differentially methylated probes (DMPs) were selected. Fifteen DMPs in 14 genes (BMP2, CTSA, DEFB1, GRN, MTF1, SERPINE1, SERPINE2, SFRP1, STAT3, TAGLN2, TCF4, THBS1, ZBTB20, ZNF292) were identified as the most relevant to ongoing pregnancy status. These 15 DMPs yielded accuracy rates of 83.53%, 85.26%, 85.78%, and 76.44%, and areas under the receiver operating characteristic curves (AUCs) of 0.90, 0.91, 0.89, and 0.86 for prediction by random forest (RF), naïve Bayes (NB), support vector machine (SVM), and k-nearest neighbors (KNN), respectively. SERPINE1, SERPINE2, and TAGLN2 maintained their methylation difference trends in an independent set of cervical secretion samples, resulting in accuracy rates of 71.46%, 80.06%, 80.72%, and 80.68%, and AUCs of 0.79, 0.84, 0.83, and 0.82 for prediction by RF, NB, SVM, and KNN, respectively. Our findings demonstrate that methylation changes in WOI genes detected noninvasively from cervical secretions are potential markers for predicting IVF-ET outcomes. Further studies of cervical secretion of DNA methylation markers may provide a novel approach for precision embryo transfer.

Cervical Secretion Methylation Is Associated with the Pregnancy Outcome of Frozen-Thawed Embryo Transfer.

The causes of implantation failure remain a black box in reproductive medicine. The exact mechanism behind the regulation of endometrial receptivity is still unknown. Epigenetic modifications influence gene expression patterns and may alter the receptivity of human endometrium. Cervical secretions contain endometrial genetic material, which can be used as an indicator of the endometrial condition. This study evaluates the association between the cervical secretion gene methylation profile and pregnancy outcome in a frozen-thawed embryonic transfer (FET) cycle. Cervical secretions were collected from women who entered the FET cycle with a blastocyst transfer (36 pregnant and 36 non-pregnant women). The DNA methylation profiles of six candidate genes selected from the literature review were measured by quantitative methylation-specific PCR (qMSP). Bioinformatic analysis of six selected candidate genes showed significant differences in DNA methylation between receptive and pre-receptive endometrium. All candidate genes showed different degrees of correlation with the pregnancy outcomes in the logistic regression model. A machine learning approach showed that the combination of candidate genes' DNA methylation profiles could differentiate pregnant from non-pregnant samples with an accuracy as high as 86.67% and an AUC of 0.81. This study demonstrated the association between cervical secretion methylation profiles and pregnancy outcomes in an FET cycle and provides a basis for potential clinical application as a non-invasive method for implantation prediction.

Genome-wide analysis of cervical secretions obtained during embryo transfer reveals the association between deoxyribonucleic acid methylation and pregnancy outcomes.

OBJECTIVE: To study whether the methylation status of cervical secretions can reflect the ability of the endometrium to allow embryo implantation. DESIGN: Case-control study. SETTING: In vitro fertilization centers. PATIENT(S): Women undergoing embryo transfer cycles, in which at least 1 good-quality embryo was transferred. INTERVENTION(S): Collection of cervical secretions during the procedure of embryo transfer. MAIN OUTCOME MEASURE(S): Methylation profiles of cervical secretions in relation to pregnancy outcomes. RESULT(S): Genome-wide methylation profiles differ between cervical secretions from pregnancy and nonpregnancy cycles. Clustering analysis on the basis of the top 2,000 differentially methylated probes of cervical secretions from 28 pregnancy and 29 nonpregnancy cycles correctly categorized 86.0% of the samples in terms of conceptional status, which was verified in selected genes by quantitative methylation-specific polymerase chain reaction and validated in another independent sample set. The combination of selected genes was estimated to predict pregnancy outcomes with a maximal area under the receiver operating characteristic curve of 0.83. CONCLUSION(S): The methylation profiles of cervical secretions were associated with pregnancy outcomes in embryo transfer cycles. Although not clinically useful at present, deoxyribonucleic acid methylation in cervical secretions may shed new light on the less invasive assessment of endometrial receptivity.

Adjusted mitochondrial DNA quantification in human embryos may not be applicable as a biomarker of implantation potential.

OBJECTIVE: To evaluate the feasibility of adjusted mitochondrial DNA quantification in human embryos as a biomarker for implantation potential. DESIGN: Double-blind, observational, prospective analysis of an Asian population in a single university-affiliated in vitro fertilization center. A total of 1617 embryos derived from 380 infertile couples were collected. The DNA from blastomere biopsy (n = 99) or trophectoderm biopsy (n = 1518) were analyzed with next-generation sequencing. RESULTS: The adjusted mtDNA quantification followed a non-normal distribution in both types of the embryos. When stratified by ploidy status, the adjusted mtDNA quantification was significantly higher in aneuploid trophectoderm than in euploid cells, but not in blastomeres. The adjusted mtDNA quantification of embryos showed significant but very weak positive correlation in total trophectoderm cells with maternal age (Spearman's correlation, r = 0.095, p = 0.0028) but neither in blastomeres nor stratified by ploidy status. The median adjusted mtDNA quantification was also significantly higher in aneuploid blastocysts than in euploid ones while corrected with embryo morphology. Viable embryos did not contain significantly different quantities of adjusted mtDNA compared with nonviable embryos (implanted n = 103, non-implanted n = 164; median 0.00097 vs. 0.00088, p = 0.21) in 267 transferred blastocysts. CONCLUSION: Quantification of adjusted mitochondria DNA in human embryos was significantly lower in euploid blastocysts than in aneuploid blastocysts. However, no statistically significant differences regarding implantation outcome were evident. To our best knowledge, this study provides the largest scale and the first correlation data between mitochondria copy number and human embryo implantation potential in Asians.

Low Dose Growth Hormone Adjuvant Treatment With Ultra-Long Ovarian Stimulation Protocol in Poor Responders Showed Non-inferior Pregnancy Outcome Compared With Normal Responders.

Background: Growth hormone (GH) has long been used as adjuvant treatment in ovarian stimulation for in vitro fertilization (IVF), especially in poor responder (PR) patients. However, its clinical efficacy remains unclear, and most studies are underpowered owing to their small sample size with different regimens. Methods: Our study was divided into two parts. The first part was a parallel randomized, observational study in which 184 patients who fulfilled the criteria of poor ovarian response (POR) were enrolled and received ultra-long ovarian stimulation protocol with or without GH adjuvant therapy. For the second part, clinical data were retrospectively extracted from 163 patients classified as PRs who received 10 IU GH adjuvant therapy and 157 patients classified as normal responders (NRs) who received the same IVF protocol treatment without GH adjuvant therapy. Results: For the first part of the study, the ovarian response, the number of oocytes retrieved, and the number of available embryos transferred were all significantly higher in the GH (+) group than in the GH (-) group. The clinical pregnancy rate was significantly higher in the GH (+) group (31.9 vs. 16.7%, p = 0.0168). The miscarriage rate did not differ significantly between the groups. The ongoing pregnancy rate was also significantly higher in the GH (+) group than in the GH (-) group (26.6 vs. 14.4%, p = 0.0418). Logistic regression revealed that the chance of clinical pregnancy in the GH (+) group significant increased 2.34-fold in comparison with the GH (-) group (p = 0.018). Subgroup analysis showed that the chance of clinical pregnancy in the GH (+) group significantly increased 2.38-fold (p = 0.034). The second part of the study showed no statistical difference between the PR with GH and the NR without GH groups regarding the implantation rate (15.6 vs. 19.8%, p = 0.3254) and the clinical pregnancy rate (31.9 vs. 39.5%, p = 0.1565). The NR without GH group showed insignificantly higher chance of clinical pregnancy (OR = 1.39, p = 0.157) compared with the PR with GH group. Conclusion: Our results suggested that low-dose GH supplementation may improve ovarian response and pregnancy outcome in POR patients, particularly in patients younger than 40 years old. Moreover, the low-dose GH effect in POR patients resulted in non-inferior clinical pregnancy outcome compared with NRs.

Prediction of a rare chromosomal aberration simultaneously with next generation sequencing-based comprehensive chromosome screening in human preimplantation embryos for recurrent pregnancy loss.

Preimplantation genetic testing has been used widely in recent years as a part of assisted reproductive technology (ART) owing to the breakthrough development of deoxyribonucleic acid (DNA) sequencing. With the advancement of technology and increased resolution of next generation sequencing (NGS), extensive comprehensive chromosome screening along with small clinically significant deletions and duplications can possibly be performed simultaneously. Here, we present a case of rare chromosomal aberrations: 46,XY,dup(15)(q11.2q13),t(16;18)(q23;p11.2), which resulted in a normally developed adult but abnormal gametes leading to recurrent pregnancy loss (RPL). To our best knowledge, this is the first report of t(16;18) translocation with such a small exchanged segment detected by NGS platform of MiSeq system in simultaneous 24-chromosome aneuploidy screening.

Ring chromosome 21 presenting with sacrococcygeal teratoma: prenatal diagnosis, molecular cytogenetic characterization and literature review.

We present perinatal findings and molecular cytogenetic characterization of a prenatally detected sacrococcygeal teratoma associated with mosaic r(21). This is the first report of mosaic r(21) presenting with a fetal sacrococcygeal teratoma. We discuss cytogenetic abnormalities associated with fetal sacrococcygeal teratomas.

WITHDRAWN: Corrigendum to "Ring chromosome 21 presenting with sacrococcygeal teratoma: Prenatal diagnosis, molecular cytogenetic characterization and literature review" [Gene 522 (2013) 111-116].

This article has been withdrawn at the request of the author(s) and/or editor. The Publisher apologizes for any inconvenience this may cause. The full Elsevier Policy on Article Withdrawal can be found at http://www.elsevier.com/locate/withdrawalpolicy.