Computer software (SiD) assisted real-time single sperm selection correlates with fertilization and blastocyst formation

Evaluation of an artificial intelligence-facilitated sperm detection tool in azoospermic samples for use in ICSI

RESEARCH QUESTION

Can artificial intelligence (AI) improve the efficiency and efficacy of sperm searches in azoospermic samples?

DESIGN

This two-phase proof-of-concept study began with a training phase using eight azoospermic patients (>10,000 sperm images) to provide a variety of surgically collected samples for sperm morphology and debris variation to train a convolutional neural network to identify spermatozoa. Second, side-by-side testing was undertaken on two cohorts of non-obstructive azoospermia patient samples: an embryologist versus the AI identifying all the spermatozoa in the still images (cohort 1, n = 4), and a side-by-side test with a simulated clinical deployment of the AI model with an intracytoplasmic sperm injection microscope and the embryologist performing a search with and without the aid of the AI (cohort 2, n = 4).

RESULTS

In cohort 1, the AI model showed an improvement in the time taken to identify all the spermatozoa per field of view (0.02 ± 0.30  ×  10-5s versus 36.10 ± 1.18s, P < 0.0001) and improved recall (91.95 ± 0.81% versus 86.52 ± 1.34%, P < 0.001) compared with an embryologist. From a total of 2660 spermatozoa to find in all the samples combined, 1937 were found by an embryologist and 1997 were found by the AI in less than 1000th of the time. In cohort 2, the AI-aided embryologist took significantly less time per droplet (98.90 ± 3.19 s versus 168.7 ± 7.84 s, P < 0.0001) and found 1396 spermatozoa, while 1274 were found without AI, although no significant difference was observed.

CONCLUSIONS

AI-powered image analysis has the potential for seamless integration into laboratory workflows, to reduce the time to identify and isolate spermatozoa from surgical sperm samples from hours to minutes, thus increasing success rates from these treatments.

Automated Single-Sperm Selection Software (SiD) during ICSI: A Prospective Sibling Oocyte Evaluation

The computer-assisted program SiD was developed to assess and select sperm in real time based on motility characteristics. To date, there are limited studies examining the correlation between AI-assisted sperm selection and ICSI outcomes. To address this limit, a total of 646 sibling MII oocytes were randomly divided into two groups as follows: the ICSI group (n = 320): ICSI performed with sperm selected by the embryologist and the ICSI-SiD group (n = 326): ICSI performed with sperm selected using SiD software. Our results show a non-significant trend towards improved outcomes in the ICSI-SiD group across various biological parameters, including fertilization, cleavage, day 3 embryo development, blastocyst development, and quality on day 5. Similarly, we observed a non-significant increase in these outcomes when comparing both groups with sperm selection performed by a junior embryologist. Embryo development was monitored using a timelapse system. Some fertilization events happen significantly earlier when SiD is used for ICSI, but no significant difference was observed in the ICSI-SiD group for other timepoints. We observed comparable cumulative early and clinical pregnancy rates after ICSI-SiD. This preliminary investigation illustrated that employing the automated sperm selection software SiD leads to comparable biological outcomes, suggesting its efficacy in sperm selection.

The prospect of artificial intelligence to personalize assisted reproductive technology

Infertility affects 1-in-6 couples, with repeated intensive cycles of assisted reproductive technology (ART) required by many to achieve a desired live birth. In ART, typically, clinicians and laboratory staff consider patient characteristics, previous treatment responses, and ongoing monitoring to determine treatment decisions. However, the reproducibility, weighting, and interpretation of these characteristics are contentious, and highly operator-dependent, resulting in considerable reliance on clinical experience. Artificial intelligence (AI) is ideally suited to handle, process, and analyze large, dynamic, temporal datasets with multiple intermediary outcomes that are generated during an ART cycle. Here, we review how AI has demonstrated potential for optimization and personalization of key steps in a reproducible manner, including: drug selection and dosing, cycle monitoring, induction of oocyte maturation, and selection of the most competent gametes and embryos, to improve the overall efficacy and safety of ART.

Only the Best of the Bunch-Sperm Preparation Is Not Just about Numbers

In this Seminar, we present an overview of the current and emerging methods and technologies for optimizing the man and the sperm sample for fertility treatment. We argue that sperms are the secret to success, and that there are many avenues for improving both treatment and basic understanding of their role in outcomes. These outcomes encompass not just whether treatment is successful or not, but the wider intergenerational health of the offspring. We discuss outstanding challenges and opportunities of new technologies such as microfluidics and artificial intelligence, including potential pitfalls and advantages. This article aims to provide a comprehensive overview of the importance of sperm in fertility treatment and suggests future directions for research and innovation.

The Internet of Things in assisted reproduction

The Internet of Things (IoT) is a network connecting physical objects with sensors, software and internet connectivity for data exchange. Integrating the IoT with medical devices shows promise in healthcare, particularly in IVF laboratories. By leveraging telecommunications, cybersecurity, data management and intelligent systems, the IoT can enable a data-driven laboratory with automation, improved conditions, personalized treatment and efficient workflows. The integration of 5G technology ensures fast and reliable connectivity for real-time data transmission, while blockchain technology secures patient data. Fog computing reduces latency and enables real-time analytics. Microelectromechanical systems enable wearable IoT and miniaturized monitoring devices for tracking IVF processes. However, challenges such as security risks and network issues must be addressed through cybersecurity measures and networking advancements. Clinical embryologists should maintain their expertise and knowledge for safety and oversight, even with IoT in the IVF laboratory.

Improving outcomes of assisted reproductive technologies using artificial intelligence for sperm selection

Within the field of assisted reproductive technology, artificial intelligence has become an attractive tool for potentially improving success rates. Recently, artificial intelligence-based tools for sperm evaluation and selection during intracytoplasmic sperm injection (ICSI) have been explored, mainly to improve fertilization outcomes and decrease variability within ICSI procedures. Although significant advances have been achieved in developing algorithms that track and rank single sperm in real-time during ICSI, the clinical benefits these might have in improving pregnancy rates from a single assisted reproductive technology cycle remain to be established.

Stain-Free Sperm Analysis and Selection for Intracytoplasmic Sperm Injection Complying with WHO Strict Normal Criteria

This multi-center study evaluated a novel microscope system capable of quantitative phase microscopy (QPM) for label-free sperm-cell selection for intracytoplasmic sperm injection (ICSI). Seventy-three patients were enrolled in four in vitro fertilization (IVF) units, where senior embryologists were asked to select 11 apparently normal and 11 overtly abnormal sperm cells, in accordance with current clinical practice, using a micromanipulator and 60× bright field microscopy. Following sperm selection and imaging via QPM, the individual sperm cell was chemically stained per World Health Organization (WHO) 2021 protocols and imaged via bright field microscopy for subsequent manual measurements by embryologists who were blinded to the QPM measurements. A comparison of the two modalities resulted in mean differences of 0.18 µm (CI -0.442-0.808 µm, 95%, STD-0.32 µm) for head length, -0.26 µm (CI -0.86-0.33 µm, 95%, STD-0.29 µm) for head width, 0.17 (CI -0.12-0.478, 95%, STD-0.15) for length-width ratio and 5.7 for acrosome-head area ratio (CI -12.81-24.33, 95%, STD-9.6). The repeatability of the measurements was significantly higher in the QPM modality. Surprisingly, only 19% of the subjectively pre-selected normal cells were found to be normal according to the WHO2021 criteria. The measurements of cells imaged stain-free through QPM were found to be in good agreement with the measurements performed on the reference method of stained cells imaged through bright field microscopy. QPM is non-toxic and non-invasive and can improve the clinical effectiveness of ICSI by choosing sperm cells that meet the strict criteria of the WHO2021.

Embryo selection through artificial intelligence versus embryologists: a systematic review

STUDY QUESTION

What is the present performance of artificial intelligence (AI) decision support during embryo selection compared to the standard embryo selection by embryologists?

SUMMARY ANSWER

AI consistently outperformed the clinical teams in all the studies focused on embryo morphology and clinical outcome prediction during embryo selection assessment.

WHAT IS KNOWN ALREADY

The ART success rate is ∼30%, with a worrying trend of increasing female age correlating with considerably worse results. As such, there have been ongoing efforts to address this low success rate through the development of new technologies. With the advent of AI, there is potential for machine learning to be applied in such a manner that areas limited by human subjectivity, such as embryo selection, can be enhanced through increased objectivity. Given the potential of AI to improve IVF success rates, it remains crucial to review the performance between AI and embryologists during embryo selection.

STUDY DESIGN SIZE DURATION

The search was done across PubMed, EMBASE, Ovid Medline, and IEEE Xplore from 1 June 2005 up to and including 7 January 2022. Included articles were also restricted to those written in English. Search terms utilized across all databases for the study were: ('Artificial intelligence' OR 'Machine Learning' OR 'Deep learning' OR 'Neural network') AND ('IVF' OR 'in vitro fertili*' OR 'assisted reproductive techn*' OR 'embryo'), where the character '*' refers the search engine to include any auto completion of the search term.

PARTICIPANTS/MATERIALS SETTING METHODS

A literature search was conducted for literature relating to AI applications to IVF. Primary outcomes of interest were accuracy, sensitivity, and specificity of the embryo morphology grade assessments and the likelihood of clinical outcomes, such as clinical pregnancy after IVF treatments. Risk of bias was assessed using the Modified Down and Black Checklist.

MAIN RESULTS AND THE ROLE OF CHANCE

Twenty articles were included in this review. There was no specific embryo assessment day across the studies-Day 1 until Day 5/6 of embryo development was investigated. The types of input for training AI algorithms were images and time-lapse (10/20), clinical information (6/20), and both images and clinical information (4/20). Each AI model demonstrated promise when compared to an embryologist's visual assessment. On average, the models predicted the likelihood of successful clinical pregnancy with greater accuracy than clinical embryologists, signifying greater reliability when compared to human prediction. The AI models performed at a median accuracy of 75.5% (range 59-94%) on predicting embryo morphology grade. The correct prediction (Ground Truth) was defined through the use of embryo images according to post embryologists' assessment following local respective guidelines. Using blind test datasets, the embryologists' accuracy prediction was 65.4% (range 47-75%) with the same ground truth provided by the original local respective assessment. Similarly, AI models had a median accuracy of 77.8% (range 68-90%) in predicting clinical pregnancy through the use of patient clinical treatment information compared to 64% (range 58-76%) when performed by embryologists. When both images/time-lapse and clinical information inputs were combined, the median accuracy by the AI models was higher at 81.5% (range 67-98%), while clinical embryologists had a median accuracy of 51% (range 43-59%).

LIMITATIONS REASONS FOR CAUTION

The findings of this review are based on studies that have not been prospectively evaluated in a clinical setting. Additionally, a fair comparison of all the studies were deemed unfeasible owing to the heterogeneity of the studies, development of the AI models, database employed and the study design and quality.

WIDER IMPLICATIONS OF THE FINDINGS

AI provides considerable promise to the IVF field and embryo selection. However, there needs to be a shift in developers' perception of the clinical outcome from successful implantation towards ongoing pregnancy or live birth. Additionally, existing models focus on locally generated databases and many lack external validation.

STUDY FUNDING/COMPETING INTERESTS

This study was funded by Monash Data Future Institute. All authors have no conflicts of interest to declare.

REGISTRATION NUMBER

CRD42021256333.

Deep Learning-Based Robust Automated System for Predicting Human Sperm DNA Fragmentation Index

Background

Determining the DNA fragmentation index (DFI) by the sperm chromatin dispersion (SCD) test involves manual counting of stained sperms with halo and no halo.

Aims

The aim of this study is to build a robust artificial intelligence-based solution to predict the DFI.

Settings and Design

This is a retrospective experimental study conducted in a secondary in vitro fertilisation setup.

Materials and Methods

We obtained 24,415 images from 30 patients after the SCD test using a phase-contrast microscope. We classified the dataset into two, binary (halo/no halo) and multiclass (big/medium/small halo/degraded (DEG)/dust). Our approach consists of a training and prediction phase. The 30 patients' images were divided into training (24) and prediction (6) sets. A pre-processing method M was developed to automatically segment the images to detect sperm-like regions and was annotated by three embryologists.

Statistical Analysis Used

To interpret the findings, the precision-recall curve and F1 score were utilised.

Results

Binary and multiclass datasets containing 8887 and 15,528 cropped sperm image regions showed an accuracy of 80.15% versus 75.25%. A precision-recall curve was determined and the binary and multiclass datasets obtained an F1 score of 0.81 versus 0.72. A confusion matrix was applied for predicted and actuals for the multiclass approach where small halo and medium halo confusion were found to be highest.

Conclusion

Our proposed machine learning model can standardise and aid in arriving at accurate results without using expensive software. It provides accurate information about healthy and DEG sperms in a given sample, thereby attaining better clinical outcomes. The binary approach performed better with our model than the multiclass approach. However, the multiclass approach can highlight the distribution of fragmented and non-fragmented sperms.