Current Updates on Involvement of Artificial Intelligence and Machine Learning in Semen Analysis

Predictors of Successful Testicular Sperm Extraction: A New Era for Men with Non-Obstructive Azoospermia

Background/Objectives: Non-obstructive azoospermia (NOA) is a severe form of male infertility characterized by the absence of sperm in the ejaculate due to impaired spermatogenesis. Testicular sperm extraction (TESE) combined with intracytoplasmic sperm injection is the primary treatment, but success rates are unpredictable, causing significant emotional and financial burdens. Traditional clinical and hormonal predictors have shown inconsistent reliability. This review aims to evaluate current and emerging non-invasive preoperative predictors of successful sperm retrieval in men with NOA, highlighting promising biomarkers and their potential clinical applications. Methods: A comprehensive literature review was conducted, examining studies on clinical and hormonal factors, imaging techniques, molecular biology biomarkers, and genetic testing related to TESE outcomes in NOA patients. The potential role of artificial intelligence and machine learning in enhancing predictive models was also explored. Results: Traditional predictors such as patient age, body mass index, infertility duration, testicular volume, and serum hormone levels (follicle-stimulating hormone, luteinizing hormone, inhibin B) have limited predictive value for TESE success. Emerging non-invasive biomarkers-including anti-Müllerian hormone levels, inhibin B to anti-Müllerian hormone ratio, specific microRNAs, long non-coding RNAs, circular RNAs, and germ-cell-specific proteins like TEX101-show promise in predicting successful sperm retrieval. Advanced imaging techniques like high-frequency ultrasound and functional magnetic resonance imaging offer potential but require further validation. Integrating molecular biomarkers with artificial intelligence and machine learning algorithms may enhance predictive accuracy. Conclusions: Predicting TESE outcomes in men with NOA remains challenging using conventional clinical and hormonal parameters. Emerging non-invasive biomarkers offer significant potential to improve predictive models but require validation through large-scale studies. Incorporating artificial intelligence and machine learning could further refine predictive accuracy, aiding clinical decision-making and improving patient counseling and treatment strategies in NOA.

Enhancing Male Fertility Through AI-Based Management of Varicoceles

REVIEW PURPOSE

The clinical management of subclinical and symptomatic varicoceles in male infertility remains challenging. Current guidelines focus on treating men with abnormal semen analyses, but a more precise approach to identify, stratify, and prognosticate men with varicoceles and fertility issues is essential.

RECENT FINDINGS

Multiple studies have utilized Artificial Intelligence (AI) to analyze clinical-demographic characteristics, semen analyses, pre-operative imaging findings, and intra-operative clinical data. These AI-driven approaches aim to discover novel biomarkers that can assess, stratify, and prognosticate men with subclinical and symptomatic varicoceles requiring early intervention. These sophisticated methodologies offer new insights and strategies for understanding normal spermatogenesis and the pathophysiology of varicocele-related male infertility. The application of AI strategies is expected to revolutionize varicocele management, enhancing male fertility and optimizing reproductive outcomes.

Diagnostic value of oxidation-reduction potential for male infertility: a systematic review and meta-analysis

Background

In the last few years, studies have initially confirmed the diagnostic significance of oxidation-reduction potential (ORP) in male infertility patients. In this article, we used meta-analysis to clarify the role of ORP in the diagnosis of male infertility.

Methods

PubMed, Embase, Web of Science, and Cochrane Library were searched by computer for relevant published literature. Quality assessment of the included literature was performed by Quality Assessment of Diagnostic Accuracy Studies (QUADAS) scale. Heterogeneity analysis of included studies was conducted using Metadisc 1.4 and Stata 12.0, and effective models for quantitative synthesis were selected based on heterogeneity results; the sensitivity and specificity of the synthesis were obtained using the software, and in order to reduce the effects of heterogeneity and thresholds, the information of sensitivity and specificity was integrated. We used the subject receiver operating characteristic (SROC) curve, area under the curve (AUC) and Q* index for comprehensive evaluation.

Results

Seven papers were eventually included in the study, and the results showed that ORP had a sensitivity of 0.81 [95% confidence interval (CI): 0.80-0.82] and specificity of 0.66 (95% CI: 0.63-0.69), an AUC of 0.8 and a Q* index of 0.74 for the diagnosis of male infertility.

Conclusions

ORP has high sensitivity and specificity for diagnosing male infertility.

ChatGPT's Efficacy in Queries Regarding Polycystic Ovary Syndrome and Treatment Strategies for Women Experiencing Infertility

This study assesses the efficacy of ChatGPT-4, an advanced artificial intelligence (AI) language model, in delivering precise and comprehensive answers to inquiries regarding managing polycystic ovary syndrome (PCOS)-related infertility. The research team, comprising experienced gynecologists, formulated 460 structured queries encompassing a wide range of common and intricate PCOS scenarios. The queries were: true/false (170), open-ended (165), and multiple-choice (125) and further classified as 'easy', 'moderate', and 'hard'. For true/false questions, ChatGPT-4 achieved a flawless accuracy rate of 100% initially and upon reassessment after 30 days. In the open-ended category, there was a noteworthy enhancement in accuracy, with scores increasing from 5.53 ± 0.89 initially to 5.88 ± 0.43 at the 30-day mark (p < 0.001). Completeness scores for open-ended queries also experienced a significant improvement, rising from 2.35 ± 0.58 to 2.92 ± 0.29 (p < 0.001). In the multiple-choice category, although the accuracy score exhibited a minor decline from 5.96 ± 0.44 to 5.92 ± 0.63 after 30 days (p > 0.05). Completeness scores for multiple-choice questions remained consistent, with initial and 30-day means of 2.98 ± 0.18 and 2.97 ± 0.25, respectively (p > 0.05). ChatGPT-4 demonstrated exceptional performance in true/false queries and significantly improved handling of open-ended questions during the 30 days. These findings emphasize the potential of AI, particularly ChatGPT-4, in enhancing decision-making support for healthcare professionals managing PCOS-related infertility.

Exploring altered bovine sperm trajectories by sperm tracking in unconfined conditions

Mammalian sperm motility is getting more relevant due to rising infertility rates worldwide, generating the need to improve conventional analysis and diagnostic approaches. Nowadays, computer assisted sperm analysis (CASA) technologies represent a popular alternative to manual examination which is generally performed by observing sperm motility in very confined geometries. However, under physiological conditions, sperm describe three-dimensional motility patterns which are not well reconstructed by the limited depth of standard acquisition chambers. Therefore, affordable and more versatile alternatives are needed. Here, a motility analysis in unconfined conditions is proposed. In details, the analysis is characterized by a significant longer duration -with respect to conventional systems- with the aim to observe eventually altered motility patterns. Brightfield acquisition in rectangular glass capillaries captured frozen-thawed bovine spermatozoa which were analyzed by means of a self-written tracking routine and classified in sub-populations, based on their curvilinear velocity. To test the versatility of our approach, cypermethrin -a commonly used pesticides- known to be responsible for changes in sperm motility was employed, assessing its effect at three different time-steps. Experimental results showed that such drug induces an increase in sperm velocity and progressiveness as well as circular pattern formation, likely independent of wall interactions. Moreover, this resulted in a redistribution of sperm with the rapid class declining in number with time, but still showing an overall velocity increase. The flexibility of the approach permits parameter modifications with the experimental needs, allowing us to conduct a comprehensive examination of sperm motility. This adaptability facilitated data acquisition which can be computed at different frame rates, extended time periods, and within deeper observation chambers. The suggested approach for sperm analysis exhibits potential as a valuable augmentation to current diagnostic instruments.