Abstract Background Our institution reported in 2019 on a multilayer artificial neural network (ANN) that predicted prostate cancer at biopsy in 334 patients. That model was trained with TensorFlow 1.x and scored at three fixed step counts, and hyperparameter selection was not kept separate from test evaluation. Here we revisit an expanded cohort from the same hospital using current machine-learning practice. Methods All biopsy episodes available in the institutional database were pooled (n = 526; 524 after one non-binary outcome code and one record with missing digital rectal examination [DRE] data were dropped). The seven predictors of the original report were retained: age, prior biopsy history, PSA, prostate volume, DRE, and MRI diffusion-weighted imaging findings in the peripheral and transition zones. Twenty-seven patients contributed more than one biopsy episode, so folds were built with patient-ID-grouped, stratified k-fold cross-validation (StratifiedGroupKFold; scikit-learn 1.8.0) at 3 and 5 folds, repeated over 10 random partitions, so that no patient had episodes in a training fold and a test fold at the same time. Four classifiers were compared: L2-regularized logistic regression, gradient boosting, random forest, and a shallow (single hidden layer) multilayer perceptron. Two outcomes were modeled: detection of any prostate cancer, and detection of clinically significant prostate cancer (Gleason score ≥ 7). Results Any-cancer prevalence was 55.7% (292/524); Gleason score ≥ 7 prevalence was 39.7% (208/524). Under repeated 5-fold cross-validation, gradient boosting discriminated best, both for any prostate cancer (mean AUC 0.826, 95% CI 0.823–0.830) and for Gleason score ≥ 7 (mean AUC 0.855, 95% CI 0.852–0.859). Random forest and logistic regression followed closely (AUC 0.81–0.85). The shallow multilayer perceptron did worse, and less consistently so (any-cancer AUC 0.671; Gleason score ≥ 7 AUC 0.742), falling below the deeper five-hidden-layer ANN of the 2019 report as well. Results with 3-fold cross-validation were essentially unchanged. Conclusions In the expanded cohort, regularized logistic regression, gradient boosting and random forest each discriminated prostate cancer at biopsy at least as well as the multilayer ANN reported earlier, with much simpler models and with hyperparameter tuning kept apart from performance estimation. At this sample size the shallow network gained nothing over the simpler alternatives. This is a preprint; the study has not undergone external peer review.

To determine whether a pre-trained deep neural network (ResNet50) can non-invasively evaluate oocyte quality and predict subsequent embryonic development by extracting cytoplasmic features from static bright-field images acquired immediately before piezo-intracytoplasmic sperm injection. This prospective cohort study included 306 metaphase II oocytes obtained from 63 patients (73 cycles). Immediately before piezo-intracytoplasmic sperm injection, spindle visibility was scored using PolScope, and static bright-field images were acquired. The cytoplasmic regions were automatically segmented and fed into a pre-trained ResNet50 model to extract intermediate layer features from layers 1 and 3. The resulting feature vectors were reduced to 20 components via principal component analysis. Logistic regression and support vector machines with stratified five-fold cross-validation were used to evaluate classification performance for spindle presence, normal fertilization, blastocyst formation, and high-quality blastocyst rates. Single ResNet-derived features were assessed individually. Logistic regression principal component analysis of 20 components achieved an area under the curve of 0.679 for spindle detection (sensitivity: 84.2

Purpose:To evaluate the clinical efficacy of ERPeakSM-guided personalized embryo transfer (pET) in recurrent implantation failure (RIF) and non-RIF patients, and assess the temporal stability of the window of implantation (WOI). Methods:This prospective cohort study included patients undergoing frozen embryo transfer (FET) with good-quality embryos. In hormone replacement therapy (HRT) cycles, RIF patients received pET (n = 199) or non-personalized ET (npET, n = 272); non-RIF patients received pET (n = 74) or npET (n = 351). Propensity score matching (PSM) adjusted for confounders. modified Natural cycle (mNC) FET was analyzed using inverse probability of treatment weighting (IPTW). WOI stability was evaluated via Kaplan-Meier analysis. Results:Following PSM in RIF patients, pET significantly improved live birth rates (LBR) compared to npET (41.0% vs. 23.6%, p < 0.001) and reduced miscarriage rates (17.4% vs. 36.2%, p = 0.014). In non-RIF patients, pET did not improve LBR (35.8% vs. 25.4%, p = 0.26). IPTW-adjusted mNC-FET analysis showed significantly higher LBR in the pET group (46.3% vs. 19.2%, p = 0.026). Kaplan-Meier analysis revealed WOI stability declines significantly faster in patients ≥ 38 years (median 30.0 months, p = 0.0059). Conclusions:ERPeakSM-guided pET significantly improves LBR in RIF patients undergoing HRT and mNC-FET, but lacks benefit in non-RIF patients. WOI is dynamic; reevaluation is recommended for older patients.

This exploratory study evaluated whether a polygenic risk score (PRS) derived from a European genome-wide association study is transferable to a Japanese cohort, for both susceptibility to and severity of adult idiopathic hydrocele testis. Surgically confirmed hydrocele cases were compared with two independently ascertained control groups: prostate cancer patients with intraoperatively confirmed absence of hydrocele fluid (Control 1), and urolithiasis patients without malignancy (Stone Control). Multivariable logistic regression evaluated disease occurrence; multiple linear regression restricted to hydrocele cases evaluated the association between PRS and fluid volume. A higher PRS was associated with increased susceptibility to hydrocele when cases were compared with Stone Controls (strictest threshold: odds ratio 1.98, 95% CI 1.01-3.88, p = 0.048; AUC 0.733), though this was not observed using Control 1. Among cases, PRS showed a consistent inverse association with fluid volume across all three thresholds tested (standardized β = -0.54, p = 0.010 at ≥10 mL). Given the modest sample size and limited statistical power, these findings should be regarded as preliminary and hypothesis-generating. They nonetheless suggest that genetic susceptibility to hydrocele and its severity may show different associations with polygenic risk, a pattern requiring confirmation in larger, independent, ideally East Asian-derived cohorts.

The judicious selection of ovulation inhibitors in ovarian stimulation protocols is crucial for the success of assisted reproductive technology (ART). Herein, we investigate the dose-dependent effects of chlormadinone acetate (CMA) and cetrorelix, two distinct ovulation inhibitors, on oocyte maturation in patients with normal ovarian reserve, using univariable and multivariable Poisson regression analyses. Patients undergoing progestin-primed ovarian stimulation (PPOS) with CMA (n = 299) or gonadotropin-releasing hormone antagonist (GnRH-ant) with cetrorelix (n = 605) during their initial in vitro fertilization cycle were enrolled at our center from March 2018 to October 2020 (N = 904). The primary and secondary outcomes were the oocyte maturation and fertilization rates, respectively. After adjusting for several covariates including age, anti-Müllerian hormone levels, total gonadotropin dose, and type of trigger, we calculated the dose-dependent adjusted relative risk (aRR) and 95% confidence interval (CI) for 1 mg of CMA or 0.25 mg of cetrorelix. In the PPOS group, the median age was 34.0 years, and the median total CMA dosage was 22 mg (interquartile range [IQR]: 18.0-32.0). In the GnRH-ant group, the median age was 35.0 years, and the median total cetrorelix dosage was 0.5 mg (IQR 0.5-0.5). The aRR of the maturation rate was 1.003 (95% CI: 0.999-1.007) with PPOS (p = 0.194) and 1.009 (95% CI: 0.962-1.059) with GnRH-ant (p = 0.717). The aRR of the fertilization rate was 1.002 (95% CI: 0.985-1.020) with PPOS (p = 0.783) and 1.022 (95% CI: 0.839-1.246) with GnRH-ant (p = 0.829). Collectively, these findings indicate that within the applied dosages, ovulation inhibitors do not significantly impact oocyte maturation or fertilization rates in patients with normal ovarian reserve. These valuable insights can be applied when designing ART protocols and may guide clinicians in optimizing infertility treatments.

Research question: Does progestin-primed ovarian stimulation (PPOS) have a negative effect on reproductive outcomes compared with a gonadotrophin-releasing hormone (GnRH) antagonist protocol? Design: This retrospective cohort study included 907 patients aged <40 years with normal ovarian reserves undergoing either PPOS (n = 299) or a GnRH-antagonist protocol (n = 608) in their first IVF cycle between 2018 and 2020. An additional genetic analysis, single-cell RNA sequencing (scRNA-seq), was performed on the mural granulosa cells (mGC) of metaphase II oocyte follicles retrieved from 16 patients, with the above inclusion criteria, undergoing PPOS (n = 8) or a GnRH-antagonist protocol (n = 8) between 2021 and 2022. Inverse probability of treatment weighting (IPTW) was performed on the clinical data. Predetermined primary outcomes were the premature LH surge rate and the live birth rate of the first frozen embryo transfer cycle for the first and second IPTW analyses, respectively. Results: The premature LH surge rate was lower in the PPOS group compared with the GnRH-antagonist group (3.1% versus 20.1%, OR 0.13, 95% CI 0.07-0.23; P < 0.001) in the first IPTW analysis. The good-quality cleavage embryo rate was lower in the PPOS group compared with the GnRH-antagonist group (37.2% versus 49.1%; P < 0.001). The live birth rate was lower in the PPOS group compared with the GnRH-antagonist group (31.5% versus 42.3%, OR 0.63, 95% CI 0.46-0.86; P = 0.004) in the second IPTW analysis. The scRNA-seq analysis demonstrated higher expression of 12 mitochondrial DNA (mtDNA) genes in the PPOS group compared with the GnRH-antagonist group. Conclusion: PPOS suppressed the premature LH surge rate but was associated with a lower live birth rate compared with the GnRH-antagonist protocol. The elevated expression of mtDNA genes in mGC may also indicate a decline in oocyte quality with PPOS.

To evaluate the predictive ability of ooplasmic fluidity assessment during piezo-intracytoplasmic sperm injection (ICSI) for unintentional membrane rupture (UMR) and oocyte degeneration.We established an optical-flow analysis method that detects ooplasm movement in piezo-ICSI videos. Videos of 104 oocytes recorded during piezo-ICSI at our institute between January 2020 and January 2024 were retrospectively analyzed.To evaluate the apparent velocities in the ooplasmic region, we analyzed the vertical velocity component of the obtained optical flow between the start of puncture and the rupture of the oolemma (1st period) and between sperm injection and pipette withdrawal (2nd period). The velocity distributions between groups were compared using the 95th, 99th, and 99.5th percentiles. The 95th, 99th, and 99.5th percentiles of the survival group (n = 76) were significantly higher than those of the degeneration group (n = 28) during the 1st period (P < 0.001). Similarly, the 95th, 99th, and 99.5th percentiles of the survival group were significantly higher than those of the degenerative group during the 2nd period (P < 0.05). The 95th, 99th, and 99.5th percentiles without UMR (n = 52) were significantly higher than those with UMR (n = 50) during the 1st period (P < 0.001). Similarly, the 95th, 99th, and 99.5th percentiles without UMR were significantly higher than those with UMR during the 2nd period (P < 0.001). Greater ooplasmic fluidity was correlated with lower incidence of UMR and oocyte degeneration. Elucidation of the mechanisms underlying this phenomenon may help reduce UMR and oocyte degeneration.

Male factor infertility accounts for approximately 30% of infertility cases. When spermatozoa are extremely scarce, replicating the male gamete to fertilize a large cohort of oocytes is ideal. Additionally, patients with inherited disorders currently rely on pre-implantation genetic testing (PGT) to select healthy embryos, which raises ethical concerns owing to the generation of multiple embryos to select one healthy conceptus. Therefore, it would be beneficial to decode the genetics of a single sperm cell before conceptus generation. In this study, we demonstrated the feasibility of replicating the sperm genome via androgenesis and selecting the desired gamete before fertilization to preserve a specific paternal genotype, as confirmed by phenotypic observations and genetic testing in a murine model. We achieved satisfactory pre-implantation development rates with replicated male gametes and generated healthy offspring. Specifically, using 8-cell stage androgenetic embryos, a single spermatozoon can yield up to three conceptuses carrying an identical paternal haplotype.

In infertility treatment, smooth endoplasmic reticulum clusters (sERCs) are occasionally observed in oocytes. The permissibility of transplanting sERC embryos is still under discussion, and its impact on pregnancy remains unclear. This study aims to clarify the relationship between sERC and abnormal fertilization. We investigate the differences between normally fertilized and abnormally fertilized oocytes containing sERC. If the differences are observed, it suggests that factors other than sERC may be related to abnormal fertilization. In this study, we conducted experiments using an AutoEncoder model. Images of oocytes taken during intracytoplasmic sperm injection (ICSI) were used and categorized into three stages: "not touch", "touch", and "puncture". The AutoEncoder model was trained on images of normally fertilized oocytes, while input images consisted of oocytes that resulted in abnormal fertilization. We calculated the brightness values of the difference images and performed a Wilcoxon rank-sum test. In the "not touch" stages, significant differences were confirmed. These findings indicate that morphological differences may exist prior to or during injection, suggesting the possibility of identifying oocytes with abnormal fertilization potential even before ICSI is performed.

Background In addition to genetic predisposition, occupational and environmental factors are important for the risk of prostate cancer. We investigated the effect of single nucleotide polymorphisms (SNPs) on the development of prostate cancer in Japan, including occupational and industrial history as confounding factors in addition to age, smoking, and alcohol drinking. Methods We enrolled 210 prostate cancer patients and 504 male control patients. We conducted four genome-wide association study (GWAS) patterns for prostate cancer development. In the association test, logistic regression models incorporated age, smoking history, alcohol consumption history, and each pattern of industrial/occupational classification. Results No SNPs satisfying the genome-wide significance level of 5×10-8 were detected in GWAS. SNPs with a suggestive association level of 1×10-6 were found near the long intergenic non-protein coding RNA 1824 (LINC01824) and tripartite motif family like 2 (TRIML2) genes in the GWAS using occupational history as a confounder and near the ribosomal protein S2 pseudogene 25 (RPS2P25) gene in the GWAS using industrial history as a confounder. No SNPs that met the suggestive association level were observed in the GWAS that did not include occupational and industrial history. Conclusion By adding occupational and industrial history to the confounding factors, there were SNPs detected in the GWAS for prostate cancer development. The consideration of occupational and industrial history may increase the usefulness of GWAS.

Research question: Would the use of the intracytoplasmic sperm injection (ICSI) position detector (IPD) make it possible to identify the optimal puncture position on oolemma during Piezo-ICSI and reduce oocyte degeneration and unintentional membrane rupture (UMR)? Design: This sibling oocyte study included 917 inseminated oocytes from 113 infertile patients undergoing Piezo-ICSI. Oocytes were randomly divided into two groups: with or without IPD. The rates of UMR, degeneration, fertilization and embryonic development were compared between the two groups. As a secondary analysis, non-IPD oocytes were retrospectively assessed as appropriate or non -appropriate injection sites and analysed alongside prospective 'appropriate' injections. Results: The rates of UMR (7.0% versus 12.9%, P = 0.004) and degeneration (2.4% versus 6.1%, P < 0.01 = 0.008) were significantly lower in the IPD group than in the non-IPD group. No significant differences, however, were observed in the rates of fertilization (two pronuclei, 83.8% versus 78.9%), blastocyst formation (48.5% versus 48.8%) or good -quality blastocysts (22.5% versus 20.5%). Additionally, no significant differences were observed in the rates of pregnancy (29.4% versus 35.1%) or live births (26.5% versus 29.7%) in a single embryo transfer setting with or without IPD. Comparing all 'appropriate' injections with 'non -appropriate' injections also showed a significantly decreased rate of UMR and degeneration (both P < 0.001). Conclusions: The present study demonstrated that a real-time image analysis during Piezo-ICSI markedly reduced oocyte degeneration by avoiding areas associated with a high risk of UMR. Therefore, IPD may increase the number of embryos available for treatment.

Objective: To demonstrate the effect of intraovarian injection of lyophilized platelet-rich plasma (PRP) in poor ovarian responders. Methods: This retrospective cohort study evaluated data for 262 women who underwent intraovarian injection of lyophilized PRP at two private fertility clinics. Women treated from December 2021 to September 2022 and who met the Bologna criteria were included. Intraovarian injection of lyophilized PRP prepared from 50[Formula: see text]mL of autologous peripheral blood was followed by two centrifugations and activation with 2% CaCl2, in which cells were removed with a filter and stored at room temperature. The differences in the follicle-stimulating hormone and antral follicle count (AFC), and the number of retrieved oocytes, mature oocytes, fertilized embryos, cleavage embryos, and blastocysts were measured before and 3 months after a single injection. A biological pregnancy with embryos generated after injection was also examined. Results: Patients’ median age and anti-Mullerian hormone (AMH) levels were 44 years and 0.38[Formula: see text]ng/mL, respectively. The number of retrieved oocytes or mature oocytes and fertilized embryos or cleavage embryos was higher after injection. The number of cleavage embryos in women [Formula: see text] years significantly increased after injection. Of the 262 patients who underwent intraovarian injection of lyophilized PRP, 71 underwent embryo transfers using embryos obtained 3 months after injection, and 14 had biological pregnancies. Seven of these patients were in their 40s. Out of 14, one had a live birth, six are ongoing pregnancies, six had chemical abortions, and one had a miscarriage. Conclusions: Intraovarian injection of lyophilized PRP improved egg retrieval results, especially in patients older than 40 years. Furthermore, this treatment may contribute to biological pregnancy in patient with poor ovarian response but needs to be further investigated in more cases in the future.

Objective: To assess the pregnancy outcomes, including live birth rate and pregnancy loss for women aged 45–49 years at the time of oocyte retrieval, who underwent frozen autologous transfer of embryos. Methods: Among frozen-thawed embryo transfers (ET) performed at our clinic from September 2013 to March 2021, 1,114 cycles of women aged 45–49 years at the time of oocyte retrieval were examined retrospectively. Positive pregnancy test, clinical pregnancy, pregnancy loss, and live birth rates per transfer among each age group were analyzed using one-way analysis of variance or the Fisher’s exact test. Results: The mean age at the time of oocyte retrieval and ET was 45.8 and 46.5 years, respectively. Positive pregnancy test rate was 7.6%, of which 72.9% resulted in pregnancy loss. The clinical pregnancy and live birth rates were 5.1% and 2.0%, respectively, with a live birth rate of 3.0% for women with oocytes retrieved at 45 years of age. Multivariate analysis revealed that younger oocyte retrieval age was associated with biochemical and clinical pregnancy; however, no such association was observed for cases with live birth. The oocyte retrieval age of patients with live births was 45 years in 16 cases, 46 years in six cases, and 47 years in one case. In one case of live birth (46 years old), 21 trisomy was observed in the child; however, congenital abnormalities were not observed in other cases. Chorionic tests were performed on the fetuses in 12 of the 33 clinical abortions, of which 11 showed chromosomal abnormalities. Conclusions: Fertility treatment with autologous oocytes in advanced-age women has a low success rate. Although prognosis is poor above 45 years of age, these results can help in providing information to aid patients in their decisions regarding whether or not to pursue treatment.

Sperm retrieval rates (SRR) and clinical outcomes after intracytoplasmic sperm injection (ICSI) in testicular sperm extraction (TESE) cases in relation to the etiology of azoospermia have not well been investigated yet. Here we report our latest nine-year experience in TESE. This study investigated SRR of conventional TESE in obstructive azoospermia (OA) and microdissection TESE in cryptozoospermia and non-obstructive azoospermia (NOA) patients between September 2013 and January 2023 (2217 TESE attempts with 2123 patients). The etiologies of NOA were categorized as unexplained, Klinefelter's syndrome (KS), post chemotherapy, post orchiopexy, and microdeletion of azoospermia factor (AZF) c on the Y chromosome. A total of 944 couples had 2259 TESE-ICSI cycles (257 couples and 610 cycles with OA and 446 couples and 1069 cycles with NOA) were evaluated with respect to fertilization, embryonic development and clinical pregnancy rates (CPR). SRR of patients with first TESE attempts (47.6%) was significantly higher than that of patients who previously failed sperm retrieval (38.6%) (P<0.001). In the first TESE cases, SSRs were 100% (256/256) in OA, 21.4% (160/748) in unexplained NOA, 51.7% (93/180) in KS, 44.1% (37/84) in post chemotherapy, 75.4% (46/61) in post orchiopexy, and 81.8% (45/55) in AZFc microdeletion, respectively. SRR of OA was significantly higher, while that of unexplained NOA was lower than any other groups. Normal fertilization rates in OA (62.6%) and post chemotherapy (59.4%) were significantly higher, but that of AZFc microdelecion (35.7%) was significantly lower than any other groups. Blastocyst development rate and good-quality blastcyst rate in AZFc microdeletion (30.6 and 11.9%) were significantly lower than in OA (52.0% and 23.7%), unexplained NOA (45.7% and 21.0%), post chemotherapy (50.8% and 24.0%), post orchiopexy (45.6% and 17.7%) and KS (42.0% of blastulation).CPRs per embryo transfer were lower in unexplained NOA (33.9%), AZFc microdeletion (31.0%), and post orchiopexy (29.4%) than in OA (41.9%). We have had a total of 719 newborns so far with comparable congenital anomaly rate comparing to those with ejaculated sperm-ICSI. The success of sperm recovery, fertilization and pre- and post-implantation development was significantly influenced by the etiology of azoospermia.

Background Water channel aquaporin 1 (AQP1) protein expression is enhanced in the tunica vaginalis of patients with adult-onset non-communicating hydrocele testis and may contribute to the development of non-communicating hydrocele testis. We performed genetic and epigenetic analyses of the AQP1 gene in the tunica vaginalis of patients with adult-onset non-communicating hydrocele testis to elucidate the cause of enhanced AQP1 protein expression. Methodology The genotype was determined for Tag single-nucleotide polymorphisms (SNPs) representing the AQP1 gene and SNPs in the 5'-upstream region of the AQP1 gene. Then, by performing association analysis, the applicability of various genetic models was investigated for each SNP. Moreover, the methylation rate of CpG sites was examined for the CpG island related to the AQP1 gene. Results There was no significant association between each SNP and hydrocele testis for any of the genetic models. The average methylation rate of the 17 CpG sites evaluated was not significantly different between controls and hydrocele testis, but the methylation rate was lower in hydrocele testis than in controls at one CpG site. Conclusions There was a significant decrease in the methylation rate at one of the CpG sites in the CpG island associated with the AQP1 gene in the tunica vaginalis of patients with non-communicating hydrocele testis. This may increase AQP1 protein expression and contribute to the formation of hydrocele testis. SNPs related to the AQP1 gene were not associated with hydrocele testis.

The ideal technique of sperm extraction from testicle for non-obstructive azoospermia (NOA) would be minimally invasive and avoid destruction of testicular function without compromising the chance of retrieval adequate numbers of spermatozoa to perform ISCI. The possibility to including the sperm of larger seminiferous tubules is bigger than sclerotic seminiferous tubules. Whether the sperm is able to be collected or impossibility is not understood if not operating. Additionally, the operation is costly and physically large burden for patients. Therefore, we need a system that noninvasively measures the thickness of the tubule testicular in the testicle before microdissection testicular sperm extraction (micro TESE) procedure. We proposed a seminiferous tubule evaluation system with a higher (5.0MHz) frequency ultrasonic single probe to improve space resolution. We used an array probe (ISL Inc., ISL1938) which was 1.0MHz center frequency with 32 elements and a pulser receiver (Eishin Kagaku Ltd., MC-64) which administrated 20 nanosecond sampling interval. We employed four different diameter nylon fishing lines as small tubules (165 and 95μm) and large tubules (330 and 285μm) in total 24 in parallel phantom and ball phantom and calculate the cumulative relative frequency of amplitude of acquisition data that passed two types of band pass filters. We measured 20 times of objects which displayed treads on plane and spherical objects and evaluated with the multiple regression analysis and fuzzy logic. The evaluated rate of large line increased according to the increment of the number of large lines. The correlation coefficient and absolute error was 0.93 and 8.40% for parallel objects, additionally 0.91 and 9.81% for ball objects, respectively. To compare evaluate results with result of the multiple regression analysis, fuzzy logic evaluated the large lines rate (8.34% and 8.62% for parallel and ball objects, respectively) with significant lower mean absolute error. The proposed method successfully evaluated the rate of the large lines. This experimental result suggested the high possibility to determine the larger diameter testicular tubule under the assumption that the characteristics of human testicular tubule are similar to that of the fishing line. This method has advantage of cost and tractability as well as accuracy as a predictor of sperm retrieval before micro TESE.