External validation of the post-varicocele repair semen analysis nomogram to predict total motile sperm count: A multicenter study

Current and Future Applications of Artificial Intelligence to Diagnose and Treat Male Infertility

Artificial intelligence (AI) models are being increasingly applied to modern medicine. Within the field of urology, reproductive urology specifically offers many opportunities to utilize this advanced computational technology for diagnostic and therapeutic benefit. While the use of AI models in diagnosing and treating male infertility remains in its early days, current and future applications of these models include automation of semen analysis testing; predicting semen quality; identifying subsets of infertile men most likely to benefit from surgical treatment (i.e., varicocelectomy, surgical sperm retrieval); identifying rare sperm from testis tissue; and selecting optimal sperm for in vitro fertilization (IVF) with intracytoplasmic sperm injection (ICSI). In this chapter, we review the current literature surrounding these applications and discuss opportunities for future research.

Effect of varicocele and varicocelectomy on sperm deoxyribonucleic acid fragmentation rates in infertile men with clinical varicocele

Varicocele is the leading cause of male infertility. It can affect sperm quantity and quality through various non-mutually exclusive pathophysiological mechanisms, mainly oxidative stress. Excessive production of reactive oxygen species may overwhelm the sperm's defenses against oxidative stress and harm the sperm's DNA. Excessive sperm DNA breaks, so-called sperm DNA fragmentation, result from the oxidative stress cascade and are commonly found in the ejaculates of men with varicocele and fertility-related issues. Measuring sperm DNA fragmentation can provide valuable information on the extent of harm and might help select candidates for surgical treatment. Varicocelectomy is beneficial for alleviating oxidative stress-associated infertility and improving sperm DNA integrity. However, reproductive outcomes of infertile men with elevated sperm DNA fragmentation rates and surgically treated varicoceles remain poorly studied, and there is a need for well-designed trials to determine the impact of sperm DNA fragmentation reduction on natural and medically assisted reproduction.

Nomograms for Predicting Postoperative Sperm Improvements in Varicocele Patients

Background

Varicocele is a condition that seriously affects male fertility. It can cause pathological changes in the testicles and affect their spermatogenesis and endocrine function.

Objective

To formulate nomograms to predict sperm improvements after microscopic varicocelectomy.

Design setting and participants

A retrospective analysis was conducted on varicocele patients who met the research criteria and were enrolled from March 2020 to June 2022. They were divided into a development and a validation cohort in a 2:1 ratio.

Outcome measurements and statistical analysis

Data on preoperative testicular atrophy index, bilateral testicular elastic modulus, testosterone, pre- and postoperative 6-mo total sperm count, sperm concentration, and sperm vitality were collected. An increase of ≥25% is considered a postoperative improvement in sperm parameters. Predictive nomograms were constructed through forward stepwise LR regression, based on independent risk factors filtered by univariate and multivariate logistic regression analyses. Receiver operating characteristic curve analysis, calibration curve, and decision curve analysis were employed to assess the performance of the models.

Results and limitations

The areas under the curve of nomograms for predicting the postoperative improvement of total sperm count, sperm concentration, and sperm vitality were 0.915, 0.986, and 0.924 respectively. The nomogram models demonstrated good predictive performance. The single-center sample size was a limitation of this study.

Conclusions

In this study, we developed effective predictive nomogram models for anticipating postoperative improvements in sperm quality among varicocele patients. These models offer a significant value in providing accurate predictions of surgical outcomes. However, it is crucial to conduct further external validation.

Patient summary

In this study, a predictive nomogram model was constructed for assessing the improvement of sperm quality in varicocele patients after surgery. The model offered satisfactory results.

Prediction of semen analysis parameter improvement after varicocoelectomy using 1 H NMR-based metabonomics assays

BACKGROUND

Varicocoele is the most common correctable cause of male infertility; however, predicting varicocoelectomy outcomes is difficult. "Omics" techniques have been increasingly used to develop new diagnostic and prognostics tools for several male infertility causes, and could be applied to study varicocoele.

OBJECTIVES

The objective is to create metabolomics models capable of segregating men who improved semen analysis (SA) parameters or achieved natural pregnancy after microsurgical varicocoelectomy (MV) from those who did not, using hydrogen-1 nuclear magnetic resonance (1 H NMR) spectra of seminal plasma of pre-operative samples.

MATERIAL AND METHODS

We recruited 29 infertile men with palpable varicocoele. 1 H NMR spectra of seminal plasma were obtained from pre-operative samples and used to create metabonomics models. Improvement was defined as an increase in the total motile progressive sperm count (TMC) of the post-operative SA when compared to the baseline, and pregnancy was assessed for 24 months after MV.

RESULTS

Using linear discriminant analysis (LDA), we created a model that discriminated the men who improved SA from those who did not with accuracy of 93.1%. Another model segregated men who achieved natural pregnancy from men who did not. We identified seven metabolites that were important for group segregation: caprylate, isoleucine, N-acetyltyrosine, carnitine, N-acetylcarnitine, creatine, and threonine.

DISCUSSION

We described the use of metabonomics model to predict with high accuracy the outcomes of MV in infertile men with varicocoele. The most important metabolites for group segregation are involved in energy metabolism and oxidative stress response, highlighting the pivotal role of these mechanisms in the pathophysiology of varicocoele.

CONCLUSIONS

1 H NMR spectroscopy of seminal plasma can be used in conjunction with multivariate statistical tools to create metabonomics models useful to segregate men with varicocoele based on the reproductive outcomes of MV. These models may help counseling infertile men with varicocoele regarding their prognosis after surgery.

Artificial Intelligence Based Machine Learning Models Predict Sperm parameter Upgrading after Varicocele Repair: A Multi-Institutional Analysis

PURPOSE

Varicocele repair is recommended in the presence of a clinical varicocele together with at least one abnormal semen parameter, and male infertility. Unfortunately, up to 50% of men who meet criteria for repair will not see meaningful benefit in outcomes despite successful treatment. We developed an artificial intelligence (AI) model to predict which men with varicocele will benefit from treatment.

MATERIALS AND METHODS

We identified men with infertility, clinical varicocele, and at least one abnormal semen parameter from two large urology centers in North America (Miami and Toronto) between 2006 and 2020. We collected pre and post-operative clinical and hormonal data following treatment. Clinical upgrading was defined as an increase in sperm concentration that would allow a couple to access previously unavailable reproductive options. The tiers used for upgrading were: 1-5 million/mL (ICSI/IVF), 5-15 million/mL (IUI) and >15 million/mL (natural conception). Thus moving from ICSI/IVF to IUI, or from IUI to natural conception, would be considered an upgrade. AI models were trained and tested using R to predict which patients were likely to upgrade after surgery. The model sorted men into categories that defined how likely they were to upgrade after surgery (likely, equivocal, and unlikely).

RESULTS

Data from 240 men were included from both centers. A total of 45.6% of men experienced an upgrade in sperm concentration following surgery, 48.1% did not change, and 6.3% downgraded. The data from Miami were used to create a random forest model for predicting upgrade in sperm concentration. On external validation using Toronto data, the model accurately predicted upgrade in 87% of men deemed likely to improve, and in 49% and 36% of men who were equivocal and unlikely to improve, respectively. Overall, the personalized prediction for patients in the validation cohort was accurate (AUC 0.72).

CONCLUSIONS

A machine learning model performed well in predicting clinically meaningful post-varicocelectomy sperm parameters using pre-operative hormonal, clinical, and semen analysis data. To our knowledge, this is the first prediction model to show the utility of hormonal data, as well as the first to use machine learning models to predict clinically meaningful upgrading. This model will be published online as a clinical calculator that can be used in the preoperative counseling of patients.