Investigating the Artificial intelligence in prediction and evaluation sperm and embryo quality in In vitro fertilization (IVF): A systematic review

Investigating the Artificial intelligence in prediction and evaluation sperm and embryo quality in In vitro fertilization (IVF): A systematic review | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Systematic Review Investigating the Artificial intelligence in prediction and evaluation sperm and embryo quality in In vitro fertilization (IVF): A systematic review shahrzad kaveh, Aida Ghafari, zahra khedri, solmaz sohrabei This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5504223/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 29 Jul, 2025 Read the published version in Discover Artificial Intelligence → Version 1 posted 10 You are reading this latest preprint version Abstract Importance: Assisted Reproductive Technologies (ART) have been developed to address infertility by improving embryo selection. Artificial intelligence (AI), using Time-Lapse Imaging (TLI), enhances predictions from fertilization to the blastocyst stage. Objective: Studies show AI can identify suitable embryos more effectively than specialists, improving in-vitro fertilization (IVF) success rates by enhancing transfer success and reducing miscarriage risks. With IVF success rates below 40%, it is essential to explore AI methods to boost outcomes . Findings: A systematic review in October 2024 searched databases like PubMed and Scopus using terms related to IVF and AI, excluding non-English and qualitative studies. Twenty-seven studies were reviewed; 17 predicted treatment responses with deep learning. Two studies used neural networks for successful treatment prediction, and eight employed ML methods such as NB, SVM, and RF, with an average AUC of 0.91. Models showed 90-96% accuracy, sensitivity, and precision. Conclusion: AI technologies, particularly NB and Reinforcement Learning, show promise in improving IVF outcomes by enhancing classification and diagnosis while saving time. Interdisciplinary approaches using micro and Nano-biotechnology can help overcome clinical challenges . Relevance: Examining the quality of sperm and egg separately using AI could further improve fertility testing and success in ART, optimizing clinical results. Figures Figure 1 Figure 2 Figure 3 Introduction Approximately 15% of couples worldwide suffer from infertility. In recent years, In Vitro Fertilization (IVF) has been a common treatment for infertility, though it may not be effective for all individuals. Given the substantial time and financial costs involved, with only a 40% success rate in achieving pregnancy, the use of advanced technologies to predict fertility outcomes is essential [ 1 ]. Artificial intelligence (AI) and modern engineering technologies are employed to select high-quality eggs for fertilization, offering a novel, cost-effective, and highly accurate solution for infertility treatment. IVF involves fertilizing eggs with sperm in a laboratory setting and then transferring one or more embryos, after several stages of cell division, into the uterus to continue natural embryonic development [ 2 ]. The use of cutting-edge technologies in medicine for treating infertility has become increasingly common. Fields such as nanotechnology, biotechnology, photonics, and AI have been integrated into modern infertility treatments. Before egg transfer, the egg must be assessed to ensure its suitability for fertilization. The selection of a poor-quality egg can result in IVF failure or complications in the uterus, posing risks to the patient [ 3 ]. Traditionally, geneticists examine the eggs under a microscope and, based on their expertise, select the egg with the best quality for fertilization. However, the success rate of this approach is approximately 30%. Thus, for IVF to be more successful combining advanced engineering methods and AI for precise egg selection is crucial. One innovative technique involves using a microchip with a microfluidic channel, approximately 100 microns in cross-section and a few millimeters in length, designed to match the egg's size [ 4 ]. The egg passes through this channel and is subjected to mechanical stresses, allowing AI-based imaging to evaluate the egg's quality based on various parameters. This method is more than twice as accurate as traditional observational techniques, as AI can assess all relevant variables related to the egg. The quality of immature eggs, which will be used in IVF after reaching maturity, has been determined with approximately 80% accuracy using this approach. Furthermore, new technologies employing electrical variables have been developed to assess mature eggs and select those suitable for cell division. These methods are more precise than conventional approaches and pose fewer challenges for integration into fertility clinics [ 5 ]. They can significantly reduce human error in the egg selection process, offering couples facing infertility a more efficient, lower-cost option for starting a family. Researchers have placed eggs in specific kits and, using AI-based computer analysis, can identify which eggs are best suited for IVF [ 6 , 7 ]. Thus, developing reliable methods to assess the quality of both eggs and sperm is crucial. It appears that evaluating egg morphology through micro-opto-fluidic channels, alongside assessments based on advanced engineering and AI techniques can provide more accurate and non-invasive methods for determining egg quality, significantly improving the success rate of IVF [ 8 , 9 ]. This technique is notably more precise and faster than previous methods, as AI-driven egg quality assessments do not rely on embryologists’ subjective judgments. Through an electronic engineering technique, it is possible to predict with nearly 60% accuracy which high-quality mature eggs will reach the blastocyst stage (ready for transfer to the mother's body) following successful fertilization [ 10 ]. Additionally, one of the current standards for infertility clinics is the use of Radio Frequency Identification (RFID) systems or AI technologies for sample identification during IVF. These systems prevent the simultaneous use of two samples, minimizing the risk of errors. Given the higher accuracy of AI techniques in evaluating eggs, sperm, and predicting IVF success, identifying precise techniques is of utmost importance. By utilizing these advanced methods in the design of clinical decision-support systems, physicians can achieve greater success in fertility treatments. Methods This study is a systematic review following the PRISMA 2020 guidelines [ 11 ]. In evidence-based medical research, formulating and designing research questions is considered the most critical part of these studies, as it shapes the study and investigation. 2.1. Eligibility Criteria SPICE is a useful tool, similar to PICO, for asking focused clinical questions and conducting qualitative reviews [ 18 ]. SPICE stands for Setting, Perspective, Intervention, Comparison, and Evaluation, offering a structured way to frame practice questions to find evidence in existing research. SPICE was more suitable for formulating our research questions: (i) Setting: all publications globally, (ii) Perspective: patients and healthcare providers, (iii) Intervention: AI techniques, (iv) Comparison: focusing solely on the infertility treatment community, (v) Evaluation: How do AI techniques perform in evaluating IVF data? 2.1.1. Inclusion Criteria Studies meeting all the following criteria were included in the review: (1) Studies related to IUI and IVF, (2) Utilization of AI techniques (machine learning, ensemble learning, deep learning), (3) Reporting performance based on AI evaluation metrics (accuracy, sensitivity, specificity), (4) Articles published in English. 2.1.2. Exclusion Criteria Exclusion criteria were as follows: (1) Studies that were systematic reviews or meta-analyses, (2) Book chapters or systematic review articles, (3) Studies using classical machine learning methods, (4) Articles without full-text availability in English. 2.2. Information Sources and Search Strategy A systematic search was conducted in electronic databases, including Web of Science, Medline (via PubMed), Scopus, and IEEE, to identify relevant studies published within a seven-year period from early 2017 to October 1, 2024. Additionally, the Embase database was searched until January 10, 2024. The search strategy included a combination of keywords related to "artificial intelligence," "IVF," and "male infertility." Table 1 lists the full set of keywords and terms used for the Scopus database search strategy. A reference management software (EndNote X8, Thomson Reuters) was used for collecting references and removing duplicates. 2.3. Study Selection Two reviewers (Z.Kh and S.S) independently screened the titles and abstracts of identified articles. The full text of retrieved articles was reviewed if both reviewers deemed them relevant. Any disagreement between the reviewers was resolved through discussion with a third researcher. The screening process is illustrated in the PRISMA 2020 flowchart (Flowchart 1). Two authors (Z.Kh and S.S) analyzed and synthesized the key characteristics of the selected studies and extracted the primary features. The first author (S.S) evaluated the extracted data and validated the key elements. 2.4. Data Collection Process The first reviewer (Z.Kh) collected the necessary data from the selected studies. The second reviewer (S.S) then verified the accuracy of the collected data. Any discrepancies were reviewed and resolved by two reviewers (S.S and Z.KH). The main data elements and characteristics of the selected studies are presented in Table 1 . 2.5. Study Bias Risk The Joanna Briggs Institute (JBI) Critical Appraisal Checklist [ 12 ] for analytical cross-sectional studies was used to assess the risk of bias. The aim of this evaluation was to assess the methodological quality of the studies, and it included eight questions as follows: (1) were the inclusion criteria for the sample clearly defined? (2) Were the subjects and setting described in detail? (3) Was exposure measured in a valid and reliable manner? (4) Were objective and standard criteria used to measure conditions? (5) Were confounding factors identified? (6) Were strategies to deal with confounders stated? (7) Were the outcomes measured in a valid and reliable way? (8) Was appropriate statistical analysis used? These questions were answered with four options: (1) Yes, (2) No, (3) Unclear, and (4) Not applicable. Each "Yes" answer was given one point. If 70% or more of the questions were answered "Yes," the study was considered to have a low risk of bias. If 50%-69% of the questions were answered "Yes," the risk of bias was considered moderate, while less than 50% was considered high risk [ 16 ]. Two authors (Z.Kh and S.S) completed the checklist. Results As illustrated in Flowchart 1 , the database search resulted in the retrieval of 569 records up to September 2024. After removing duplicates and screening based on inclusion criteria, 27 articles were selected for review, with their characteristics detailed in Table 2 . The studies demonstrated that the data used for modeling through machine learning were highly diverse. While convolutional neural networks (CNNs) were the most frequently used deep learning models, ensemble learning had the least usage. The effectiveness of the selected AI methods across various studies was evaluated, with the results presented in Table 2 . The performance of these methods was assessed using various metrics, including accuracy, precision, sensitivity, specificity, and AUC (area under the curve). The reported metrics indicated a statistically significant performance of the AI techniques. Therefore, many algorithms used in these studies demonstrated AI’s capability in identifying and evaluating oocytes and sperm, as well as predicting treatment outcomes. AI techniques were predominantly applied to assess oocytes, followed by predicting pregnancy outcomes after IVF procedures. AI methods were also highly effective in sperm evaluation, with robust performance metrics (Fig. 1 ) . The analysis revealed that most studies used image data of oocytes and sperm for AI-based evaluations, while clinical data were primarily used to predict pregnancy chances and live birth outcomes. Figure 2 illustrates that the frequency of utilization of CNN models and Ensemble Learning is the highest, followed by the Random Forest (RF) model. The area under the curve for Artificial Neural Networks (ANN) and then RF exhibited the highest values, with the accuracy of RF and Ensemble Learning models achieving the most significant results in data analysis (Fig. 3 ). Table 3 indicates that the primary application domain for deep learning models in artificial intelligence is in oocyte selection and sperm evaluation. Additionally, techniques such as Support Vector Machines (SVM) and RF have been employed through clinical data analysis and laboratory assessments to predict successful pregnancy outcomes through in vitro fertilization (IVF). Table2: Study details on used of AI In IVF tech Names of authors, Country, year of publication Area used AI methods Data set Tools Out Put Badiola AC et al [13] 2023 USA Embryo evaluation CNNs A dataset of 592 blastocyst images collected from two IVF clinics for six consecutive months Python Acc=91% Sen=92% Goss .DM et al [14] 2023 Australia Sperm detection CNN 540 images, containing 5624 unique sperm instances Python Acc=92% Barnett-Itzhaki.Z et al [15] 2022 Canada Sperm detection Ensemble DL 2254 stain-free single sperm images PyTorch Acc=94% Pre=93% AUC=0.96 Ghasemian F et al [16] 2022 Iran Sperm detection CNN 309 specimens was collected from fertile and infertile men (age 22-38 years) who visited in Alzahra hospital (IVF center) Python Acc=94.65% Kothandaraman R et al [17] 2022 India Embryo evaluation Sperm detection EHIC ART dataset Python ROC=0.96 Güvenir HA et al [18] 2015 Turkey Prediction IVF Naïve Bayes and Random Forest 1456 woman has been compiled by the IVF Python ROC=0.95 Chen F et al [19] 2022 China Prediction of pregnancy outcomes SVM, RF, LR GEO, The methylation microarray (GSE144664) included 12 IVMconceived CC samples and 12 IVM-unconceived CC samples Python ROC=0.94, 0.88, 0.97 Handayani N et al [20] 2022 Prediction of pregnancy outcomes DT,RF, GXB data set comprised 1,669 clinically pregnant women and 2,901 nonpregnant women Python Recall=83% Acc=63% Wang CW et al [21] 2022 Taiwan Prediction of pregnancy outcomes RF, LR 100,000 IVF cycles, 2,000 ART cycles Python ROC=0.91 ROC=0.77 You JB et al [22] 2021 Canada Sperm selection SVM, DT,LSTM ,CNN sperm images and the DNA fragmentation Python Acc =88% Nguyen.D et al [23] 2021 Vietnam IVF/ICSI cycles CNN 1,135 day-3 embryo images from 188 IVF/ICS cycles TensorFlow Python deep learning library Acc=70.07% Sen=87.04% AUC=094 Barnes .J et al[24] 2023 USA embryo selection XGBoost, k-NN, SVM, ResNet18 CNN The dataset encompassed 10 378 human blastocysts (day 5 n=3994; day 6 n=6384) collected from 1385 patients from 2012 to 2017 at the Weill Cornell Medicine Center of Reproductive Medicine, New York, NY, USA. R (version 4.1.2), PyTorch (version 1.4.0) AUC=0.73, .071, .070 Acc=90% Sen=99% Wang.G et al [25] 2024 Chine embryo selection MLP 41,279 embryo images and 2,136 embryo time-lapse videos Python package of scikit-learn (version 0.22.1) AUC=0.85 Raef.B et al[26] 2019 Iran predict embryo transfer outcomes NB, SVM, NN, KNN, DT 500 patients and 1360 transferred embryos Python package Acc=90.4% ROC=0.937 Sato.T et al [27] 2022 Japan sperm selection YOLOv3 (CNN) high-quality images of 4625 unstained sperm, of Japanese men, evaluated on an RGB scale Python package Sen=88% Rafiul Hassan.M et al [28] 2018 Australia prediction of pregnancy outcome ANN, MLP, C4.5, RF, CART, SVM total dataset of 1048 IVF patients was divided into 10 nearly equal-sized mutually exclusive folds Python package AUC=0.992 Acc=97.3% Liu .X et al [29] 2023 China predictive performance on the live birth chance LR, RF, LGBM, and XGBoost 1,857 women undergoing the IVF cycle Python 3.9.12 Acc=70% Cao SS et al [30] 2024 China Predictive clinical pregnancies associated with surgical sperm retrieval XGBoost, SVM, CatBoost, GBDT 420 infertile couples who underwent surgical testicular sperm retrieval with ICSI for different etiologies at the IVF center of the Second Affiliated Hospital of Wenzhou Medical University between February 2020 and March 2023 were selected Python 3.11, R 4.3.1. The R packages ROC=0.85 Acc=80% Xi Q et al [31] 2021 China embryo selection XGBoost, LR 3383 individual IVF cases Python ROC=0.80 Ma BX et al [32] 2024 China embryo selection CNN 3448 biopsied blastocysts from 979 Time-lapse (TL)-PGT cycles Python ROC=0.66 Wang J et al [33] 2024 China sperm detection CNN dataset containing the 7,353 sperm Python 3.9, Pytorch version 1.7.1 Recall=90% Xiao YH et al [34] 2024 China embryo selection XGBoost, KNN 4,216 POR cycles who underwent in vitro fertilization (IVF) / intracytoplasmic sperm injection (ICSI) R software (version 4.3.1) ROC= 0.813 Goyal A et al [35] 2020 India predicts live‑birth occurrence before IVF RF,MLP, CNN, AdaBoost, Voting—hard classifer 495,630 patient records with 94 features on treatment cycles collected from various patient studies Python 3.9, Pytorch version 1.7.1 ROC=0.83 Recall=73% Mehrjerd A et al [36] 2024 Iran sperm quality evaluation Bagging, RF, Boosting 599 couples (courses) under IVF/ICSI 734 and 1197 under IUI Python and various frameworks, including Scikit-learn, Pandas, and NumPy ROC=0.79 Sene AA et al [37] 2021 Iran predict the intrauterine insemination success rate among infertile couples ANN 948 couples underwent IUI MATLAB software Acc=71% Spe=76% ROC=0.99 Sujata PN et al [38] 2020 India Success Rate in IVF CNN 525 embryo images Python 3.9 Acc=85% Canosa S et al [39] 2024 Italy early embryo morphokinetics identifies SVM , KNN 575 embryos Python 3.9 ROC=0.84 Table 3 Frequency AI methods IN IVF area AI Methods Area Used Frequency CNN Embryo evaluation, Sperm detection 6 SVM Prediction of pregnancy outcomes, Sperm selection, Embryo selection 5 RF Prediction of pregnancy outcomes, Predictive performance on live birth chance 5 XGBoost Embryo selection, Predictive clinical pregnancies associated with surgical sperm retrieval 4 Ensemble DL Sperm detection 1 EHIC Embryo evaluation, Sperm detection 1 Naïve Bayes Prediction IVF 1 LR Prediction of pregnancy outcomes 2 MLP Predictive performance on live birth chance, Success Rate in IVF 2 DT Prediction of pregnancy outcomes 2 KNN Predictive clinical pregnancies associated with surgical sperm retrieval, Early embryo morph kinetics 2 ANN Prediction of pregnancy outcomes 2 YOLOv3 (CNN) Sperm selection 1 Bagging Sperm quality evaluation 1 Boosting Sperm quality evaluation 1 Of the 24 studies included in this review, the risk of bias was deemed low for the majority. Only two studies were identified with a moderate risk of bias [ 28 , 30 ], and one study was classified with a high risk of bias [ 17 ] Table 4 . The questions "Were confounding factors identified?" and "Were strategies to address confounding factors outlined?" were not applicable in our included studies, as they were not experimental research. Discussion In this systematic review, 26 studies were evaluated and all of them had high performance indicators in the evaluation. The results of the studies presented in Table 2 show significant advances in the use of artificial intelligence (AI) in various fields related to in vitro fertilization (IVF). This research is particularly focused on embryo assessment, sperm detection, and pregnancy outcome prediction, and has introduced innovative methods to solve complex challenges in this field. Research shows that advanced algorithms such as Convolutional Neural Networks (CNN) and Ensemble Learning models provide extraordinary accuracy in analyzing fetal images. For example, Badiola AC et al (2023) and Barnes J et al. (2023) using CNN and hybrid models have been able to report a very high sensitivity and accuracy (up to 99%) for evaluating the quality of the created embryo. Using artificial intelligence and creating innovation in the field of techniques and building an optimal algorithm can improve the embryo selection process from traditional medical criteria to analyzes based on high-quality data. Algorithms based on artificial intelligence play an important role in detecting the exact quality of sperm and evaluating its condition. For example, Sato T et al (2022) using the YOLOv3 architecture has been able to provide high accuracy in detecting their quality by analyzing and evaluating uncolored sperm images. This innovative method reduces the time and cost associated with sample preparation. Also, Mehrjerd A et al (2024) was able to analyze sperm quality on a large scale with high accuracy using hybrid models such as RF. The review of studies in this field shows the use of various methods from classical models (such as Naïve Bayes) to advanced models (such as XGBoost). Chen F et al (2022) in his study using XGB and methylation microarray (GEO) data was able to achieve a significant level under the graph (ROC = 0.97). This index shows the high efficiency of the algorithm built in data analysis. which highlights the ability of XGB algorithm to analyze and combine genetic and clinical data. Also, Wang CW et al. (2022) have successfully predicted large-scale fertility using data related to IVF cycles and RF models. One of the prominent aspects of these studies is the use of a large and high-quality clinical and demographic data set in addition to the images. For example, Wang G et al. (2024) used time-lapse images of the fetus, which include the dynamic dimensions of growth in the evaluation. On the other hand, some researches such as Nguyen D et al (2021) have worked with more limited data, which has a negative effect on the performance of the models. Most studies have used the Python programming language and advanced libraries such as PyTorch and Scikit-learn. Also, the use of R software in some studies (such as Barnes J et al (2023)) shows the integration of statistical analysis and machine learning. A significant aspect in these studies is the adaptation of artificial intelligence architectures for specific applications. For example, the use of YOLOv3 for Sperm detection or CNN for analysis of time-lapse images of embryos demonstrates the flexibility of these methods related to IVF. Increasing the accuracy of assessments and predictions can have a direct effect on reducing costs and improving clinical outcomes It is necessary that these systems can help IVF applicants to adopt the best treatment process for them. Conclusion Building trust in AI within the medical field requires transparency and interpretability in the models used, enabling specialists to review and understand their functioning. In areas like IVF, where AI adoption is increasing, establishing a solid regulatory framework hinges on open discussions about the potential risks and benefits of AI to ensure its safe and effective use. This transparency is essential for fostering trust among physicians and patients, making them more confident in the reliability of AI-driven outcomes. A critical factor for AI success in clinical settings is data accessibility. Data sharing must be done in a way that safeguards confidentiality and fairness, necessitating the development of appropriate methods to streamline data processing. With over three million women worldwide undergoing IVF treatment annually, access to diverse and comprehensive datasets significantly enhances the accuracy and efficiency of AI models, ultimately leading to better treatment outcomes. Abbreviations IVF - In Vitro Fertilization TLI - Time-Lapse Imaging ART - Assisted Reproductive Technologies SPICE - Setting, Perspective, Intervention, Comparison, and Evaluation JBI - Joanna Briggs Institute PRISMA - Preferred Reporting Items for Systematic Reviews and Meta-Analyses RCT - Randomized Controlled Trial RF - Random Forest SVM - Support Vector Machine CNN - Convolutional Neural Network AUC - Area under the Curve NB - Naive Bayes RL - Reinforcement Learning AI : Artificial Intelligence ML : Machine Learning DL : Deep Learning Ensemble ML : Ensemble Machine Learning CNN : Convolutional Neural Network SVM : Support Vector Machine RF : Random Forest XGBoost : Extreme Gradient Boosting LSTM : Long Short-Term Memory ANN : Artificial Neural Network FNN : Feedforward Neural Network K-NN : K-Nearest Neighbors LR : Logistic Regression RNN : Recurrent Neural Network Acc: Accuracy Sen: Sensitivity or Recall Spe: Specificity AUC: Area under the ROC Curve ROC: Receiver Operating Characteristic RCT - Randomized Controlled Trial PCOS - Polycystic Ovary Syndrome EHIC - Electronic Health Information Center (in context with embryo and sperm detection) Declarations Funding Statement : We don’t have an any funding Conflict of interest statement for all authors: The authors declare that there is no conflict of interest. Any institution has not financially supported this article or the authors have provided organization and all its financial resources. CRediT Authorship Contribution Statement: S.S. designed the study and wrote the main manuscript text, Z. Kh collected data and evaluated quality articles. Authors reviewed the manuscript Attestation statements (see below): I declare that the above statements are true and accurate to the best of my knowledge, information and belief. Data sharing statement (see below): The datasets used and/or analyses during the current study available from the corresponding author on reasonable request. References Hassan MR, Al-Insaif S, Hossain MI, Kamruzzaman J. A machine learning approach for prediction of pregnancy outcome following IVF treatment. Neural computing and applications. 2020 Apr;32(7):2283-97. Yiğit P, Bener A, Karabulut S. Comparison of machine learning classification techniques to predict implantation success in an IVF treatment cycle. Reproductive BioMedicine Online. 2022 Nov 1;45(5):923-34. Blank C, Wildeboer RR, DeCroo I, Tilleman K, Weyers B, De Sutter P, Mischi M, Schoot BC. Prediction of implantation after blastocyst transfer in in vitro fertilization: a machine-learning perspective. Fertility and sterility. 2019 Feb 1;111(2):318-26. Liu L, Jiao Y, Li X, Ouyang Y, Shi D. Machine learning algorithms to predict early pregnancy loss after in vitro fertilization-embryo transfer with fetal heart rate as a strong predictor. Computer Methods and Programs in Biomedicine. 2020 Nov 1;196:105624. Oseguera-López I, Ruiz-Díaz S, Ramos-Ibeas P, Pérez-Cerezales S. Novel techniques of sperm selection for improving IVF and ICSI outcomes. Frontiers in cell and developmental biology. 2019 Nov 29;7:298. Güvenir HA, Misirli G, Dilbaz S, Ozdegirmenci O, Demir B, Dilbaz B. Estimating the chance of success in IVF treatment using a ranking algorithm. Medical & biological engineering & computing. 2015 Sep;53:911-20. Güvenir HA, Misirli G, Dilbaz S, Ozdegirmenci O, Demir B, Dilbaz B. Estimating the chance of success in IVF treatment using a ranking algorithm. Medical & biological engineering & computing. 2015 Sep;53:911-20. Hariton E, Pavlovic Z, Fanton M, Jiang VS. Applications of artificial intelligence in ovarian stimulation: a tool for improving efficiency and outcomes. Fertility and sterility. 2023 Jul 1;120(1):8-16. Vinson DR, Rauchwerger AS, Karadi CA, Shan J, Warton EM, Zhang JY, Ballard DW, Mark DG, Hofmann ER, Cotton DM, Durant EJ. Clinical decision support to O ptimize C are of patients with A trial F ibrillation or flutter in the E mergency department: protocol of a stepped-wedge cluster randomized pragmatic trial (O’CAFÉ trial). Trials. 2023 Mar 31;24(1):246. Hanassab S, Abbara A, Yeung AC, Voliotis M, Tsaneva-Atanasova K, Kelsey TW, Trew GH, Nelson SM, Heinis T, Dhillo WS. The prospect of artificial intelligence to personalize assisted reproductive technology. npj Digital Medicine. 2024 Mar 1;7(1):55. Tugwell P, Tovey D. PRISMA 2020. Journal of Clinical Epidemiology. 2021 Jun 1;134:A5-6. Jordan Z, Lockwood C, Munn Z, Aromataris E. The updated Joanna Briggs Institute model of evidence-based healthcare. JBI Evidence Implementation. 2019 Mar 1;17(1):58-71. Badiola AC. Artificial Intelligence (AI) for Embryo Ranking and its Use in Human Assisted Reproduction (Doctoral dissertation, University of Kent (United Kingdom)). Goss DM, Vasilescu SA, Vasilescu PA, Cooke S, Kim SH, Sacks GP, Gardner DK, Warkiani ME. AI facilitated sperm detection in azoospermic samples for use in ICSI. medRxiv. 2023 Oct 25:2023-10. Barnett-Itzhaki Z, Elbaz M, Butterman R, Amar D, Amitay M, Racowsky C, Orvieto R, Hauser R, Baccarelli AA, Machtinger R. Machine learning vs. classic statistics for the prediction of IVF outcomes. Journal of assisted reproduction and genetics. 2020 Oct;37:2405-12. Ghasemian F, Bahadori MH, Hosseini Kolkooh SZ, Esmaeili M. Using deep learning algorithm: the study of sperm head vacuoles and its correlation with protamine mRNA ratio. Cell Journal (Yakhteh). 2022 Jan 1;24(1):7-14. Kothandaraman R, Andavar S, Raj RS. A Hybrid Feature Ranking Algorithm for Assisted Reproductive Technology Outcome Prediction. Brazilian Archives of Biology and Technology. 2022 Jun 27;65:e22210605. Güvenir HA, Misirli G, Dilbaz S, Ozdegirmenci O, Demir B, Dilbaz B. Estimating the chance of success in IVF treatment using a ranking algorithm. Medical & biological engineering & computing. 2015 Sep;53:911-20. Chen F, Chen Y, Mai Q. Multi-Omics Analysis and Machine Learning Prediction Model for Pregnancy Outcomes After Intracytoplasmic Sperm Injection–in vitro Fertilization. Frontiers in Public Health. 2022 Jun 30;10:924539. Handayani N, Louis CM, Erwin A, Aprilliana T, Polim AA, Sirait B, Boediono A, Sini I. Machine Learning Approach to Predict Clinical Pregnancy Potential in Women Undergoing IVF Program. Fertility & Reproduction. 2022 Jun 10;4(02):77-87. S Wang CW, Kuo CY, Chen CH, Hsieh YH, Su EC. Predicting clinical pregnancy using clinical features and machine learning algorithms in in vitro fertilization. Plos one. 2022 Jun 8;17(6):e0267554. You JB, McCallum C, Wang Y, Riordon J, Nosrati R, Sinton D. Machine learning for sperm selection. Nature Reviews Urology. 2021 Jul;18(7):387-403. Nguyen Thanh T, Nguyen DM, Dinh Le T, Ngoc Do L, Tien Nguyen S, Nguyen Minh P, Nguyen Van P, Minh Bui T, Thi Bui TT, Nguyen Dao H, Trung Nguyen K. The Relationship Between Smooth Endoplasmic Reticulum Clusters in Metaphase II Oocytes and Embryological and Birth Outcomes in Infertile Couples. International Journal of General Medicine. 2024 Dec 31:3269-77. Barnes J, Brendel M, Gao VR, Rajendran S, Kim J, Li Q, Malmsten JE, Sierra JT, Zisimopoulos P, Sigaras A, Khosravi P. A non-invasive artificial intelligence approach for the prediction of human blastocyst ploidy: a retrospective model development and validation study. The Lancet Digital Health. 2023 Jan 1;5(1):e28-40.S Wang G, Wang K, Gao Y, Chen L, Gao T, Ma Y, Jiang Z, Yang G, Feng F, Zhang S, Gu Y. A generalized AI system for human embryo selection covering the entire IVF cycle via multi-modal contrastive learning. Patterns. 2024 May 2. Raef B, Maleki M, Ferdousi R. Computational prediction of implantation outcome after embryo transfer. Health informatics journal. 2020 Sep;26(3):1810-26. Sato T, Kishi H, Murakata S, Hayashi Y, Hattori T, Nakazawa S, Mori Y, Hidaka M, Kasahara Y, Kusuhara A, Hosoya K. A new deep‐learning model using YOLOv3 to support sperm selection during intracytoplasmic sperm injection procedure. Reproductive Medicine and Biology. 2022 Jan;21(1):e12454. Hassan MR, Al-Insaif S, Hossain MI, Kamruzzaman J. A machine learning approach for prediction of pregnancy outcome following IVF treatment. Neural computing and applications. 2020 Apr;32(7):2283-97. Liu X, Chen Z, Ji Y. Construction of the machine learning-based live birth prediction models for the first in vitro fertilization pregnant women. BMC Pregnancy and Childbirth. 2023 Jun 27;23(1):476. Cao SS, Liu XM, Song BT, Hu YY. Interpretable machine learning models for predicting clinical pregnancies associated with surgical sperm retrieval from testes of different etiologies: a retrospective study. BMC urology. 2024 Jul 29;24(1):156. Xi Q, Yang Q, Wang M, Huang B, Zhang B, Li Z, Liu S, Yang L, Zhu L, Jin L. Individualized embryo selection strategy developed by stacking machine learning model for better in vitro fertilization outcomes: an application study. Reproductive biology and endocrinology. 2021 Dec;19:1-0. Ma BX, Zhao GN, Yi ZF, Yang YL, Jin L, Huang B. Enhancing clinical utility: deep learning-based embryo scoring model for non-invasive aneuploidy prediction. Reproductive Biology and Endocrinology. 2024 May 22;22(1):58. Wang J, Jin Y, Jiang A, Chen W, Shan G, Gu Y, Ming Y, Li J, Yue C, Huang Z, Librach C. Testing the generalizability and effectiveness of deep learning models among clinics: sperm detection as a pilot study. Reproductive Biology and Endocrinology. 2024 May 22;22(1):59. Xiao YH, Hu YL, Lv XY, Huang LJ, Geng LH, Liao P, Ding YB, Niu CC. The construction of machine learning-based predictive models for high-quality embryo formation in poor ovarian response patients with progestin-primed ovarian stimulation. Reproductive Biology and Endocrinology. 2024 Jul 10;22(1):78. Goyal A, Kuchana M, Ayyagari KP. Machine learning predicts live-birth occurrence before in-vitro fertilization treatment. Scientific reports. 2020 Dec 1;10(1):20925. Mehrjerd A, Dehghani T, Jajroudi M, Eslami S, Rezaei H, Ghaebi NK. Ensemble machine learning models for sperm quality evaluation concerning success rate of clinical pregnancy in assisted reproductive techniques. Scientific Reports. 2024 Oct 16;14(1):24283. Sene AA, Zandieh Z, Soflaei M, Torshizi HM, Sheibani K. Using artificial intelligence to predict the intrauterine insemination success rate among infertile couples. Middle East Fertility Society Journal. 2021 Dec 15;26(1):46. Sujata PN, Madiwalar SM, Aparanji VM. Machine learning techniques to improve the success rate in in-vitro fertilization (IVF) procedure. InIOP Conference Series: Materials Science and Engineering 2020 Sep 1 (Vol. 925, No. 1, p. 012039). IOP Publishing. Canosa S, Licheri N, Bergandi L, Gennarelli G, Paschero C, Beccuti M, Cimadomo D, Coticchio G, Rienzi L, Benedetto C, Cordero F. A novel machine-learning framework based on early embryo morphokinetics identifies a feature signature associated with blastocyst development. Journal of Ovarian Research. 2024 Mar 15;17(1):63. Karami N, Iravani F, Bavarsad SB, Asadollahi S, Hoseini SM, Montazeri F, Kalantar SM. Comparing the advantages, disadvantages and diagnostic power of different non-invasive pre-implantation genetic testing: A literature review. International Journal of Reproductive BioMedicine (IJRM). 2024 May 12:177-90. Von Wolff M. The role of natural cycle IVF in assisted reproduction. Best practice & research Clinical endocrinology & metabolism. 2019 Feb 1;33(1):35-45. Giacobbe M, Conatti M, Gomes A, Bonetti TC, Monteleone PA. Effectivity of conventional in vitro fertilization (IVF) and intracytoplasmic sperm injection (ICSI) when male factor is absent: a perspective point of view. JBRA Assisted Reproduction. 2022 Jan;26(1):123. van den Hoven L, Hendriks JC, Verbeet JG, Westphal JR, Wetzels AM. Status of sperm morphology assessment: an evaluation of methodology and clinical value. Fertility and sterility. 2015 Jan 1;103(1):53-8. Kragh MF, Karstoft H. Embryo selection with artificial intelligence: how to evaluate and compare methods?. Journal of assisted reproduction and genetics. 2021 Jul;38(7):1675-89. VerMilyea M, Hall JM, Diakiw SM, Johnston A, Nguyen T, Perugini D, Miller A, Picou A, Murphy AP, Perugini M. Development of an artificial intelligence-based assessment model for prediction of embryo viability using static images captured by optical light microscopy during IVF. Human Reproduction. 2020 Apr 28;35(4):770-84. Table 4 Table 4 is available in the Supplementary Files section. Additional Declarations No competing interests reported. Supplementary Files Table4.docx Cite Share Download PDF Status: Published Journal Publication published 29 Jul, 2025 Read the published version in Discover Artificial Intelligence → Version 1 posted Editorial decision: Revision requested 19 Feb, 2025 Reviews received at journal 20 Jan, 2025 Reviewers agreed at journal 18 Jan, 2025 Reviews received at journal 02 Jan, 2025 Reviewers agreed at journal 27 Dec, 2024 Reviewers agreed at journal 26 Dec, 2024 Reviewers invited by journal 23 Dec, 2024 Editor assigned by journal 19 Dec, 2024 Submission checks completed at journal 18 Dec, 2024 First submitted to journal 22 Nov, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5504223","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Systematic Review","associatedPublications":[],"authors":[{"id":392255184,"identity":"8fc11100-a6f0-4321-bed1-67a532c5b537","order_by":0,"name":"shahrzad kaveh","email":"","orcid":"","institution":"Zanjan University of Medical Sciences","correspondingAuthor":false,"prefix":"","firstName":"shahrzad","middleName":"","lastName":"kaveh","suffix":""},{"id":392255185,"identity":"0fb8642e-7758-4626-b5c6-d32cf2626cf2","order_by":1,"name":"Aida Ghafari","email":"","orcid":"","institution":"Zanjan University of Medical Sciences","correspondingAuthor":false,"prefix":"","firstName":"Aida","middleName":"","lastName":"Ghafari","suffix":""},{"id":392255186,"identity":"1b9b929c-e44d-4558-a874-5c2ec1fa3049","order_by":2,"name":"zahra khedri","email":"","orcid":"","institution":"Zanjan University of Medical Sciences","correspondingAuthor":false,"prefix":"","firstName":"zahra","middleName":"","lastName":"khedri","suffix":""},{"id":392255187,"identity":"5fc4d774-0f5b-4a3b-b00d-a4edbc37df7b","order_by":3,"name":"solmaz sohrabei","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABCUlEQVRIiWNgGAWjYFAC5gYwZcDAwAakJORAnAMP8GphRNFiYQzWkkCClopEMBefFvn2xtYNP/7UyZlLH3724EOFRPr8sMMPgbbYyek2YNdicOZg280ensPGln1p5oYzzkjkbrydZgDUkmxsdgCHFonEths8EgcSN5xhMJPmbQNqmZ0A0nIgcRsOLfIzEttu/jGoA2ph/yb9959EuuHs9A94tTDcSGy7zZPADNTCYybN2CCRIC+dg98WkF9uyxw4bGxwhqfcsOeYhOEG6ZyCAwkGuP0i39587OYbYIgZnGHf9uBHTZ28/Oz0zR8+VNjJ4dKCxV6wSgNilYPtbSBF9SgYBaNgFIwEAACRIGaQ8wuLVgAAAABJRU5ErkJggg==","orcid":"","institution":"Shahid Beheshti University of Medical Sciences","correspondingAuthor":true,"prefix":"","firstName":"solmaz","middleName":"","lastName":"sohrabei","suffix":""}],"badges":[],"createdAt":"2024-11-22 11:23:36","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5504223/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5504223/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s44163-025-00420-8","type":"published","date":"2025-07-29T16:13:31+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":72335308,"identity":"d7246ece-78f7-49d4-8998-ed05a8f2b737","added_by":"auto","created_at":"2024-12-25 15:41:29","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":21775,"visible":true,"origin":"","legend":"\u003cp\u003earea use in AI\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-5504223/v1/a8b0bd22f0c7c5a9cad23054.png"},{"id":72335312,"identity":"b020f871-b439-4d37-ac7b-f1c93600195a","added_by":"auto","created_at":"2024-12-25 15:41:29","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":9257,"visible":true,"origin":"","legend":"\u003cp\u003eAI tech frequently\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-5504223/v1/5add6f4d5ff54b1166acfa40.png"},{"id":72335300,"identity":"469c3002-851b-4619-a334-8ee28d186b30","added_by":"auto","created_at":"2024-12-25 15:41:28","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":69924,"visible":true,"origin":"","legend":"\u003cp\u003eAI tech Acc \u0026amp; AUC\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-5504223/v1/d97fe65eb01b4dd9bc4be974.png"},{"id":88268222,"identity":"c7e708b5-d024-420e-a3f2-2014551e3b1d","added_by":"auto","created_at":"2025-08-04 16:50:12","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1512207,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5504223/v1/f938ef90-dfe5-4436-8a8b-b4f048fd9b05.pdf"},{"id":72335301,"identity":"c3765b1f-5031-4c9f-86bd-4220938fb9a8","added_by":"auto","created_at":"2024-12-25 15:41:29","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":215157,"visible":true,"origin":"","legend":"","description":"","filename":"Table4.docx","url":"https://assets-eu.researchsquare.com/files/rs-5504223/v1/c6c3655a86a76ba8f6916b46.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Investigating the Artificial intelligence in prediction and evaluation sperm and embryo quality in In vitro fertilization (IVF): A systematic review","fulltext":[{"header":"Introduction","content":"\u003cp\u003eApproximately 15% of couples worldwide suffer from infertility. In recent years, In Vitro Fertilization (IVF) has been a common treatment for infertility, though it may not be effective for all individuals. Given the substantial time and financial costs involved, with only a 40% success rate in achieving pregnancy, the use of advanced technologies to predict fertility outcomes is essential [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. Artificial intelligence (AI) and modern engineering technologies are employed to select high-quality eggs for fertilization, offering a novel, cost-effective, and highly accurate solution for infertility treatment. IVF involves fertilizing eggs with sperm in a laboratory setting and then transferring one or more embryos, after several stages of cell division, into the uterus to continue natural embryonic development [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. The use of cutting-edge technologies in medicine for treating infertility has become increasingly common. Fields such as nanotechnology, biotechnology, photonics, and AI have been integrated into modern infertility treatments. Before egg transfer, the egg must be assessed to ensure its suitability for fertilization. The selection of a poor-quality egg can result in IVF failure or complications in the uterus, posing risks to the patient [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. Traditionally, geneticists examine the eggs under a microscope and, based on their expertise, select the egg with the best quality for fertilization. However, the success rate of this approach is approximately 30%. Thus, for IVF to be more successful combining advanced engineering methods and AI for precise egg selection is crucial. One innovative technique involves using a microchip with a microfluidic channel, approximately 100 microns in cross-section and a few millimeters in length, designed to match the egg's size [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. The egg passes through this channel and is subjected to mechanical stresses, allowing AI-based imaging to evaluate the egg's quality based on various parameters. This method is more than twice as accurate as traditional observational techniques, as AI can assess all relevant variables related to the egg. The quality of immature eggs, which will be used in IVF after reaching maturity, has been determined with approximately 80% accuracy using this approach. Furthermore, new technologies employing electrical variables have been developed to assess mature eggs and select those suitable for cell division. These methods are more precise than conventional approaches and pose fewer challenges for integration into fertility clinics [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. They can significantly reduce human error in the egg selection process, offering couples facing infertility a more efficient, lower-cost option for starting a family. Researchers have placed eggs in specific kits and, using AI-based computer analysis, can identify which eggs are best suited for IVF [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Thus, developing reliable methods to assess the quality of both eggs and sperm is crucial. It appears that evaluating egg morphology through micro-opto-fluidic channels, alongside assessments based on advanced engineering and AI techniques can provide more accurate and non-invasive methods for determining egg quality, significantly improving the success rate of IVF [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. This technique is notably more precise and faster than previous methods, as AI-driven egg quality assessments do not rely on embryologists\u0026rsquo; subjective judgments. Through an electronic engineering technique, it is possible to predict with nearly 60% accuracy which high-quality mature eggs will reach the blastocyst stage (ready for transfer to the mother's body) following successful fertilization [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Additionally, one of the current standards for infertility clinics is the use of Radio Frequency Identification (RFID) systems or AI technologies for sample identification during IVF. These systems prevent the simultaneous use of two samples, minimizing the risk of errors. Given the higher accuracy of AI techniques in evaluating eggs, sperm, and predicting IVF success, identifying precise techniques is of utmost importance. By utilizing these advanced methods in the design of clinical decision-support systems, physicians can achieve greater success in fertility treatments.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003eThis study is a systematic review following the PRISMA 2020 guidelines [\u003cspan class=\"CitationRef\"\u003e11\u003c/span\u003e]. In evidence-based medical research, formulating and designing research questions is considered the most critical part of these studies, as it shapes the study and investigation.\u003c/p\u003e\n\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\n \u003ch2\u003e2.1. Eligibility Criteria\u003c/h2\u003e\n \u003cp\u003eSPICE is a useful tool, similar to PICO, for asking focused clinical questions and conducting qualitative reviews [\u003cspan class=\"CitationRef\"\u003e18\u003c/span\u003e]. SPICE stands for Setting, Perspective, Intervention, Comparison, and Evaluation, offering a structured way to frame practice questions to find evidence in existing research. SPICE was more suitable for formulating our research questions: (i) Setting: all publications globally, (ii) Perspective: patients and healthcare providers, (iii) Intervention: AI techniques, (iv) Comparison: focusing solely on the infertility treatment community, (v) Evaluation: How do AI techniques perform in evaluating IVF data?\u003c/p\u003e\n\u003c/div\u003e\n\u003ch3\u003e2.1.1. Inclusion Criteria\u003c/h3\u003e\n\u003cp\u003eStudies meeting all the following criteria were included in the review:\u003c/p\u003e\n\u003cp\u003e(1) Studies related to IUI and IVF,\u003c/p\u003e\n\u003cp\u003e(2) Utilization of AI techniques (machine learning, ensemble learning, deep learning),\u003c/p\u003e\n\u003cp\u003e(3) Reporting performance based on AI evaluation metrics (accuracy, sensitivity, specificity),\u003c/p\u003e\n\u003cp\u003e(4) Articles published in English.\u003c/p\u003e\n\u003ch3\u003e2.1.2. Exclusion Criteria\u003c/h3\u003e\n\u003cp\u003eExclusion criteria were as follows:\u003c/p\u003e\n\u003cp\u003e(1) Studies that were systematic reviews or meta-analyses,\u003c/p\u003e\n\u003cp\u003e(2) Book chapters or systematic review articles,\u003c/p\u003e\n\u003cp\u003e(3) Studies using classical machine learning methods,\u003c/p\u003e\n\u003cp\u003e(4) Articles without full-text availability in English.\u003c/p\u003e\n\u003ch3\u003e2.2. Information Sources and Search Strategy\u003c/h3\u003e\n\u003cp\u003eA systematic search was conducted in electronic databases, including Web of Science, Medline (via PubMed), Scopus, and IEEE, to identify relevant studies published within a seven-year period from early 2017 to October 1, 2024. Additionally, the Embase database was searched until January 10, 2024. The search strategy included a combination of keywords related to \u0026quot;artificial intelligence,\u0026quot; \u0026quot;IVF,\u0026quot; and \u0026quot;male infertility.\u0026quot; Table \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e lists the full set of keywords and terms used for the Scopus database search strategy. A reference management software (EndNote X8, Thomson Reuters) was used for collecting references and removing duplicates.\u003c/p\u003e\n\u003cp\u003e\u003cimg src=\"https://myfiles.space/user_files/122228_c8a1650c59388082/122228_custom_files/img1734713952.png\"\u003e\u003cbr\u003e\u003c/p\u003e\n\u003ch3\u003e2.3. Study Selection\u003c/h3\u003e\n\u003cp\u003eTwo reviewers (Z.Kh and S.S) independently screened the titles and abstracts of identified articles. The full text of retrieved articles was reviewed if both reviewers deemed them relevant. Any disagreement between the reviewers was resolved through discussion with a third researcher. The screening process is illustrated in the PRISMA 2020 flowchart (Flowchart 1). Two authors (Z.Kh and S.S) analyzed and synthesized the key characteristics of the selected studies and extracted the primary features. The first author (S.S) evaluated the extracted data and validated the key elements.\u003c/p\u003e\n\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\n \u003ch2\u003e2.4. Data Collection Process\u003c/h2\u003e\n \u003cp\u003eThe first reviewer (Z.Kh) collected the necessary data from the selected studies. The second reviewer (S.S) then verified the accuracy of the collected data. Any discrepancies were reviewed and resolved by two reviewers (S.S and Z.KH). The main data elements and characteristics of the selected studies are presented in Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e\n\u003c/div\u003e\n\u003ch3\u003e2.5. Study Bias Risk\u003c/h3\u003e\n\u003cp\u003eThe Joanna Briggs Institute (JBI) Critical Appraisal Checklist [\u003cspan class=\"CitationRef\"\u003e12\u003c/span\u003e] for analytical cross-sectional studies was used to assess the risk of bias. The aim of this evaluation was to assess the methodological quality of the studies, and it included eight questions as follows: (1) were the inclusion criteria for the sample clearly defined? (2) Were the subjects and setting described in detail? (3) Was exposure measured in a valid and reliable manner? (4) Were objective and standard criteria used to measure conditions? (5) Were confounding factors identified? (6) Were strategies to deal with confounders stated? (7) Were the outcomes measured in a valid and reliable way? (8) Was appropriate statistical analysis used? These questions were answered with four options: (1) Yes, (2) No, (3) Unclear, and (4) Not applicable. Each \u0026quot;Yes\u0026quot; answer was given one point. If 70% or more of the questions were answered \u0026quot;Yes,\u0026quot; the study was considered to have a low risk of bias. If 50%-69% of the questions were answered \u0026quot;Yes,\u0026quot; the risk of bias was considered moderate, while less than 50% was considered high risk [\u003cspan class=\"CitationRef\"\u003e16\u003c/span\u003e]. Two authors (Z.Kh and S.S) completed the checklist.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eAs illustrated in \u003cstrong\u003eFlowchart 1\u003c/strong\u003e, the database search resulted in the retrieval of 569 records up to September 2024. After removing duplicates and screening based on inclusion criteria, 27 articles were selected for review, with their characteristics detailed in \u003cstrong\u003eTable\u0026nbsp;2\u003c/strong\u003e. The studies demonstrated that the data used for modeling through machine learning were highly diverse. While convolutional neural networks (CNNs) were the most frequently used deep learning models, ensemble learning had the least usage. The effectiveness of the selected AI methods across various studies was evaluated, with the results presented in \u003cstrong\u003eTable\u0026nbsp;2\u003c/strong\u003e. The performance of these methods was assessed using various metrics, including accuracy, precision, sensitivity, specificity, and AUC (area under the curve). The reported metrics indicated a statistically significant performance of the AI techniques. Therefore, many algorithms used in these studies demonstrated AI\u0026rsquo;s capability in identifying and evaluating oocytes and sperm, as well as predicting treatment outcomes.\u003c/p\u003e\n\u003cp\u003eAI techniques were predominantly applied to assess oocytes, followed by predicting pregnancy outcomes after IVF procedures. AI methods were also highly effective in sperm evaluation, with robust performance metrics (Fig.\u0026nbsp;\u003cspan\u003e1\u003c/span\u003e\u003cstrong\u003e)\u003c/strong\u003e. The analysis revealed that most studies used image data of oocytes and sperm for AI-based evaluations, while clinical data were primarily used to predict pregnancy chances and live birth outcomes.\u003c/p\u003e\n\u003cp\u003eFigure \u003cspan\u003e2\u003c/span\u003e illustrates that the frequency of utilization of CNN models and Ensemble Learning is the highest, followed by the Random Forest (RF) model. The area under the curve for Artificial Neural Networks (ANN) and then RF exhibited the highest values, with the accuracy of RF and Ensemble Learning models achieving the most significant results in data analysis (Fig.\u0026nbsp;\u003cspan\u003e3\u003c/span\u003e). Table\u0026nbsp;\u003cspan\u003e3\u003c/span\u003e indicates that the primary application domain for deep learning models in artificial intelligence is in oocyte selection and sperm evaluation. Additionally, techniques such as Support Vector Machines (SVM) and RF have been employed through clinical data analysis and laboratory assessments to predict successful pregnancy outcomes through in vitro fertilization (IVF).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable2: Study details on used of AI In IVF tech\u003c/strong\u003e\u003c/p\u003e\n\u003cdiv\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"756\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eNames of authors,\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eCountry,\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eyear of publication\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e\u003cspan dir=\"RTL\"\u003e\u0026nbsp;\u003c/span\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eArea used\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eAI methods\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cspan dir=\"RTL\"\u003e\u0026nbsp;\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 180px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eData set\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eTools\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eOut Put\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eBadiola AC et al [13]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2023\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eUSA\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eEmbryo evaluation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eCNNs\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 180px;\"\u003e\n \u003cp\u003eA dataset of 592 blastocyst images collected from two IVF clinics for six consecutive months\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e\u003cspan dir=\"RTL\"\u003e\u0026nbsp;\u003c/span\u003e\u003c/p\u003e\n \u003cp\u003ePython\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAcc=91%\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eSen=92%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eGoss .DM et al [14]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2023\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eAustralia\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eSperm detection\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eCNN\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 180px;\"\u003e\n \u003cp\u003e540 images, containing 5624 unique sperm instances\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e\u003cspan dir=\"RTL\"\u003e\u0026nbsp;\u003c/span\u003e\u003c/p\u003e\n \u003cp\u003ePython\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAcc=92%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBarnett-Itzhaki.Z et al [15]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2022\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eCanada\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eSperm detection\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eEnsemble DL\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 180px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e2254 stain-free single sperm images\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e\u003cspan dir=\"RTL\"\u003e\u0026nbsp;\u003c/span\u003e\u003c/p\u003e\n \u003cp\u003ePyTorch\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAcc=94%\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003ePre=93%\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eAUC=0.96\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eGhasemian F et al [16]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2022\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eIran\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eSperm detection\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eCNN\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 180px;\"\u003e\n \u003cp\u003e309 specimens was collected from fertile and\u003cbr\u003e\u0026nbsp;infertile men (age 22-38 years) who visited in Alzahra\u003cbr\u003e\u0026nbsp;hospital (IVF center)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e\u003cspan dir=\"RTL\"\u003e\u0026nbsp;\u003c/span\u003e\u003c/p\u003e\n \u003cp\u003ePython\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAcc=94.65%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eKothandaraman R et al [17]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2022\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eIndia\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eEmbryo evaluation\u003c/p\u003e\n \u003cp\u003eSperm detection\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eEHIC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 180px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eART dataset\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e\u003cspan dir=\"RTL\"\u003e\u0026nbsp;\u003c/span\u003e\u003c/p\u003e\n \u003cp\u003ePython\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eROC=0.96\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eG\u0026uuml;venir HA et al [18]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2015\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eTurkey\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003ePrediction IVF\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eNa\u0026iuml;ve Bayes and Random Forest\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 180px;\"\u003e\n \u003cp\u003e1456 woman has been compiled by the IVF\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e\u003cspan dir=\"RTL\"\u003e\u0026nbsp;\u003c/span\u003e\u003c/p\u003e\n \u003cp\u003ePython\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eROC=0.95\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eChen F et al [19]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2022\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eChina\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003ePrediction \u0026nbsp;of pregnancy outcomes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eSVM,\u0026nbsp;RF,\u0026nbsp;LR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 180px;\"\u003e\n \u003cp\u003eGEO,\u0026nbsp;The methylation microarray (GSE144664) included 12 IVMconceived CC samples and 12 IVM-unconceived CC samples\u003cbr\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e\u003cspan dir=\"RTL\"\u003e\u0026nbsp;\u003c/span\u003e\u003c/p\u003e\n \u003cp\u003ePython\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eROC=0.94, 0.88, 0.97\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eHandayani N et al [20]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2022\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003ePrediction \u0026nbsp;of pregnancy outcomes\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u003cspan dir=\"RTL\"\u003e\u0026nbsp;\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eDT,RF, GXB\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 180px;\"\u003e\n \u003cp\u003edata set comprised 1,669 clinically pregnant women and 2,901 nonpregnant\u003cbr\u003e\u0026nbsp;women\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e\u003cspan dir=\"RTL\"\u003e\u0026nbsp;\u003c/span\u003e\u003c/p\u003e\n \u003cp\u003ePython\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRecall=83%\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eAcc=63%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eWang CW et al [21]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2022\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eTaiwan\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003ePrediction \u0026nbsp;of pregnancy outcomes\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u003cspan dir=\"RTL\"\u003e\u0026nbsp;\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eRF, LR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 180px;\"\u003e\n \u003cp\u003e100,000 IVF cycles,\u003c/p\u003e\n \u003cp\u003e2,000 ART cycles\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e\u003cspan dir=\"RTL\"\u003e\u0026nbsp;\u003c/span\u003e\u003c/p\u003e\n \u003cp\u003ePython\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eROC=0.91\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eROC=0.77\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eYou JB et al [22]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2021\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eCanada\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eSperm selection\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eSVM, DT,LSTM ,CNN\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 180px;\"\u003e\n \u003cp\u003esperm images and the\u003cbr\u003e\u0026nbsp;DNA fragmentation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e\u003cspan dir=\"RTL\"\u003e\u0026nbsp;\u003c/span\u003e\u003c/p\u003e\n \u003cp\u003ePython\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAcc\u003c/strong\u003e\u003cstrong\u003e=88%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eNguyen.D et al [23]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2021\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eVietnam\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eIVF/ICSI cycles\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eCNN\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 180px;\"\u003e\n \u003cp\u003e1,135 day-3 embryo images from 188 IVF/ICS cycles\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003eTensorFlow Python deep learning library\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eAcc=70.07%\u003c/p\u003e\n \u003cp\u003eSen=87.04%\u003c/p\u003e\n \u003cp\u003eAUC=094\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBarnes .J et al[24]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2023\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eUSA\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eembryo selection\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eXGBoost,\u003c/p\u003e\n \u003cp\u003ek-NN,\u003c/p\u003e\n \u003cp\u003eSVM,\u003c/p\u003e\n \u003cp\u003eResNet18 CNN\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 180px;\"\u003e\n \u003cp\u003eThe dataset encompassed 10 378 human blastocysts (day 5 n=3994; day 6 n=6384) collected from 1385 patients from 2012 to 2017 at the Weill Cornell Medicine Center of Reproductive Medicine, New York, NY, USA.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003eR (version 4.1.2),\u003c/p\u003e\n \u003cp\u003ePyTorch (version 1.4.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAUC=0.73,\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e.071,\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e.070\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eAcc=90%\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eSen=99%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eWang.G et al [25]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2024\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eChine\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eembryo selection\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMLP\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 180px;\"\u003e\n \u003cp\u003e41,279 embryo images and 2,136 embryo time-lapse videos\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003ePython package of scikit-learn (version 0.22.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAUC=0.85\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRaef.B et al[26]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2019\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eIran\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003epredict embryo transfer outcomes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eNB,\u003c/p\u003e\n \u003cp\u003eSVM,\u003c/p\u003e\n \u003cp\u003eNN,\u003c/p\u003e\n \u003cp\u003eKNN, DT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 180px;\"\u003e\n \u003cp\u003e500 patients and 1360 transferred embryos\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003ePython package\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAcc=90.4%\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eROC=0.937\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSato.T et al [27]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2022\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eJapan\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003esperm selection\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eYOLOv3\u003c/p\u003e\n \u003cp\u003e(CNN)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 180px;\"\u003e\n \u003cp\u003ehigh-quality images of 4625 unstained sperm, of Japanese men, evaluated on an RGB scale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003ePython package\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSen=88%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRafiul Hassan.M et al [28]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2018\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eAustralia\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eprediction of pregnancy outcome\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eANN,\u003c/p\u003e\n \u003cp\u003eMLP,\u003c/p\u003e\n \u003cp\u003eC4.5,\u003c/p\u003e\n \u003cp\u003eRF,\u003c/p\u003e\n \u003cp\u003eCART, SVM\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 180px;\"\u003e\n \u003cp\u003etotal dataset of 1048 IVF patients was divided into 10 nearly equal-sized mutually exclusive folds\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003ePython package\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAUC=0.992\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eAcc=97.3%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLiu .X et al [29]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2023\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eChina\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003epredictive performance on the live birth chance\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eLR, RF, LGBM, and XGBoost\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 180px;\"\u003e\n \u003cp\u003e1,857 women undergoing the IVF cycle\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003ePython 3.9.12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAcc=70%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCao SS et al [30]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2024\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eChina\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003ePredictive clinical pregnancies associated with surgical sperm retrieval\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eXGBoost, SVM,\u003c/p\u003e\n \u003cp\u003eCatBoost,\u003c/p\u003e\n \u003cp\u003eGBDT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 180px;\"\u003e\n \u003cp\u003e420 infertile couples who underwent surgical testicular sperm retrieval with ICSI for different etiologies at the IVF center of the Second Affiliated Hospital of Wenzhou Medical University between February 2020 and March 2023 were selected\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003ePython 3.11,\u003c/p\u003e\n \u003cp\u003eR 4.3.1. The R packages\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eROC=0.85\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eAcc=80%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eXi Q et al [31]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2021\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eChina\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eembryo selection\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eXGBoost, LR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 180px;\"\u003e\n \u003cp\u003e3383 individual IVF cases\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003ePython\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eROC=0.80\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMa BX et al [32]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2024\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eChina\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eembryo selection\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eCNN\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 180px;\"\u003e\n \u003cp\u003e3448 biopsied blastocysts from 979 Time-lapse (TL)-PGT cycles\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003ePython\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eROC=0.66\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eWang J et al [33]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2024\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eChina\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003esperm detection\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eCNN\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 180px;\"\u003e\n \u003cp\u003edataset containing the 7,353 sperm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003ePython 3.9, Pytorch version 1.7.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRecall=90%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eXiao YH et al [34]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2024\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eChina\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eembryo selection\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eXGBoost,\u003c/p\u003e\n \u003cp\u003eKNN\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 180px;\"\u003e\n \u003cp\u003e4,216 POR cycles who underwent in vitro fertilization (IVF) / intracytoplasmic sperm injection (ICSI)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003eR software (version 4.3.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eROC=\u003c/strong\u003e 0.813\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eGoyal A et al [35]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2020\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eIndia\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003epredicts live‑birth occurrence before IVF\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eRF,MLP,\u003c/p\u003e\n \u003cp\u003eCNN,\u003c/p\u003e\n \u003cp\u003eAdaBoost,\u003c/p\u003e\n \u003cp\u003eVoting\u0026mdash;hard classifer\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 180px;\"\u003e\n \u003cp\u003e495,630 patient records with 94 features on treatment cycles collected from various patient studies\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003ePython 3.9, Pytorch version 1.7.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eROC=0.83\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eRecall=73%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMehrjerd A et al [36]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2024\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eIran\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003esperm quality evaluation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eBagging, RF,\u003c/p\u003e\n \u003cp\u003eBoosting\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 180px;\"\u003e\n \u003cp\u003e599 couples (courses) under IVF/ICSI 734 and 1197 under IUI\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003ePython and various frameworks, including Scikit-learn, Pandas, and NumPy\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eROC=0.79\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSene AA et al [37]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2021\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eIran\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003epredict the\u0026nbsp;intrauterine insemination success rate among\u0026nbsp;infertile couples\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eANN\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 180px;\"\u003e\n \u003cp\u003e948 couples underwent IUI\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003eMATLAB software\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAcc=71%\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eSpe=76%\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eROC=0.99\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSujata PN et al [38]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2020\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eIndia\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eSuccess Rate in IVF\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eCNN\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 180px;\"\u003e\n \u003cp\u003e525 embryo images\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003ePython 3.9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAcc=85%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCanosa S et al [39]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2024\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eItaly\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003eearly embryo morphokinetics identifies\u003cbr\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eSVM , KNN\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 180px;\"\u003e\n \u003cp\u003e575 embryos\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003ePython 3.9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eROC=0.84\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003cdiv align=\"char\"\u003e\u003cbr\u003e\u003c/div\u003e\n \u003ctable id=\"Tab2\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv\u003eTable 3\u003c/div\u003e\n \u003cdiv\u003e\n \u003cp\u003eFrequency AI methods IN IVF area\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eAI Methods\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eArea Used\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eFrequency\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCNN\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eEmbryo evaluation, Sperm detection\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003e6\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSVM\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePrediction of pregnancy outcomes, Sperm selection, Embryo selection\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003e5\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRF\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePrediction of pregnancy outcomes, Predictive performance on live birth chance\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003e5\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eXGBoost\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eEmbryo selection, Predictive clinical pregnancies associated with surgical sperm retrieval\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003e4\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eEnsemble DL\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSperm detection\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eEHIC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eEmbryo evaluation, Sperm detection\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNa\u0026iuml;ve Bayes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePrediction IVF\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePrediction of pregnancy outcomes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMLP\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePredictive performance on live birth chance, Success Rate in IVF\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eDT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePrediction of pregnancy outcomes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eKNN\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePredictive clinical pregnancies associated with surgical sperm retrieval, Early embryo morph kinetics\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eANN\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePrediction of pregnancy outcomes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eYOLOv3 (CNN)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSperm selection\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eBagging\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSperm quality evaluation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eBoosting\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSperm quality evaluation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003eOf the 24 studies included in this review, the risk of bias was deemed low for the majority. Only two studies were identified with a moderate risk of bias [\u003cspan\u003e28\u003c/span\u003e, \u003cspan\u003e30\u003c/span\u003e], and one study was classified with a high risk of bias [\u003cspan\u003e17\u003c/span\u003e] \u003cstrong\u003eTable\u0026nbsp;4\u003c/strong\u003e. The questions \u0026quot;Were confounding factors identified?\u0026quot; and \u0026quot;Were strategies to address confounding factors outlined?\u0026quot; were not applicable in our included studies, as they were not experimental research.\u003c/p\u003e\n"},{"header":"Discussion","content":"\u003cp\u003eIn this systematic review, 26 studies were evaluated and all of them had high performance indicators in the evaluation. The results of the studies presented in Table\u0026nbsp;2 show significant advances in the use of artificial intelligence (AI) in various fields related to in vitro fertilization (IVF). This research is particularly focused on embryo assessment, sperm detection, and pregnancy outcome prediction, and has introduced innovative methods to solve complex challenges in this field.\u003c/p\u003e\u003cp\u003eResearch shows that advanced algorithms such as Convolutional Neural Networks (CNN) and Ensemble Learning models provide extraordinary accuracy in analyzing fetal images. For example, Badiola AC et al (2023) and Barnes J et al. (2023) using CNN and hybrid models have been able to report a very high sensitivity and accuracy (up to 99%) for evaluating the quality of the created embryo. Using artificial intelligence and creating innovation in the field of techniques and building an optimal algorithm can improve the embryo selection process from traditional medical criteria to analyzes based on high-quality data. Algorithms based on artificial intelligence play an important role in detecting the exact quality of sperm and evaluating its condition. For example, Sato T et al (2022) using the YOLOv3 architecture has been able to provide high accuracy in detecting their quality by analyzing and evaluating uncolored sperm images. This innovative method reduces the time and cost associated with sample preparation. Also, Mehrjerd A et al (2024) was able to analyze sperm quality on a large scale with high accuracy using hybrid models such as RF.\u003c/p\u003e\u003cp\u003eThe review of studies in this field shows the use of various methods from classical models (such as Naïve Bayes) to advanced models (such as XGBoost). Chen F et al (2022) in his study using XGB and methylation microarray (GEO) data was able to achieve a significant level under the graph (ROC = 0.97). This index shows the high efficiency of the algorithm built in data analysis. which highlights the ability of XGB algorithm to analyze and combine genetic and clinical data. Also, Wang CW et al. (2022) have successfully predicted large-scale fertility using data related to IVF cycles and RF models. One of the prominent aspects of these studies is the use of a large and high-quality clinical and demographic data set in addition to the images. For example, Wang G et al. (2024) used time-lapse images of the fetus, which include the dynamic dimensions of growth in the evaluation. On the other hand, some researches such as Nguyen D et al (2021) have worked with more limited data, which has a negative effect on the performance of the models. Most studies have used the Python programming language and advanced libraries such as PyTorch and Scikit-learn. Also, the use of R software in some studies (such as Barnes J et al (2023)) shows the integration of statistical analysis and machine learning. A significant aspect in these studies is the adaptation of artificial intelligence architectures for specific applications. For example, the use of YOLOv3 for Sperm detection or CNN for analysis of time-lapse images of embryos demonstrates the flexibility of these methods related to IVF. Increasing the accuracy of assessments and predictions can have a direct effect on reducing costs and improving clinical outcomes It is necessary that these systems can help IVF applicants to adopt the best treatment process for them.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eBuilding trust in AI within the medical field requires transparency and interpretability in the models used, enabling specialists to review and understand their functioning. In areas like IVF, where AI adoption is increasing, establishing a solid regulatory framework hinges on open discussions about the potential risks and benefits of AI to ensure its safe and effective use. This transparency is essential for fostering trust among physicians and patients, making them more confident in the reliability of AI-driven outcomes. A critical factor for AI success in clinical settings is data accessibility. Data sharing must be done in a way that safeguards confidentiality and fairness, necessitating the development of appropriate methods to streamline data processing. With over three million women worldwide undergoing IVF treatment annually, access to diverse and comprehensive datasets significantly enhances the accuracy and efficiency of AI models, ultimately leading to better treatment outcomes.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003e\u003cstrong\u003e\u0026nbsp;IVF - In Vitro Fertilization\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;TLI - Time-Lapse Imaging\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;ART - Assisted Reproductive Technologies\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSPICE\u003c/strong\u003e - Setting, Perspective, Intervention, Comparison, and Evaluation\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eJBI\u003c/strong\u003e - Joanna Briggs Institute\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePRISMA\u003c/strong\u003e - Preferred Reporting Items for Systematic Reviews and Meta-Analyses\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRCT\u003c/strong\u003e -\u0026nbsp;Randomized Controlled Trial\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRF\u003c/strong\u003e - Random Forest\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSVM\u003c/strong\u003e -\u0026nbsp;\u003cstrong\u003eSupport Vector Machine\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCNN\u003c/strong\u003e -\u0026nbsp;\u003cstrong\u003eConvolutional Neural Network\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAUC - Area under the Curve\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eNB\u003c/strong\u003e - Naive Bayes\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRL\u003c/strong\u003e -\u0026nbsp;Reinforcement Learning\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAI\u003c/strong\u003e: Artificial Intelligence\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eML\u003c/strong\u003e: Machine Learning\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDL\u003c/strong\u003e: Deep Learning\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEnsemble ML\u003c/strong\u003e: Ensemble Machine Learning\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCNN\u003c/strong\u003e: Convolutional Neural Network\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSVM\u003c/strong\u003e: Support Vector Machine\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRF\u003c/strong\u003e: Random Forest\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eXGBoost\u003c/strong\u003e: Extreme Gradient Boosting\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eLSTM\u003c/strong\u003e: Long Short-Term Memory\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eANN\u003c/strong\u003e: Artificial Neural Network\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFNN\u003c/strong\u003e: Feedforward Neural Network\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eK-NN\u003c/strong\u003e: K-Nearest Neighbors\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eLR\u003c/strong\u003e: Logistic Regression\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRNN\u003c/strong\u003e: Recurrent Neural Network\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcc:\u003c/strong\u003e Accuracy\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;Sen:\u003c/strong\u003e Sensitivity or Recall\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSpe:\u003c/strong\u003e Specificity\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAUC:\u003c/strong\u003e Area under the ROC Curve\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eROC:\u003c/strong\u003e Receiver Operating Characteristic\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRCT\u003c/strong\u003e - Randomized Controlled Trial\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePCOS\u003c/strong\u003e - Polycystic Ovary Syndrome\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEHIC\u003c/strong\u003e - Electronic Health Information Center (in context with embryo and sperm detection)\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eFunding Statement\u003c/strong\u003e: We don\u0026rsquo;t have an any funding\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of interest statement for all authors:\u0026nbsp;\u003c/strong\u003eThe authors declare that there is no conflict of interest. Any institution has not financially supported this article or the authors have provided organization and all its financial resources.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCRediT Authorship Contribution Statement:\u0026nbsp;\u003c/strong\u003eS.S. designed the study and wrote the main manuscript text, Z. Kh collected data and evaluated quality articles. Authors reviewed the manuscript\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAttestation statements (see below):\u0026nbsp;\u003c/strong\u003e\u003cem\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/em\u003e I declare that the above statements are true and accurate to the best of my knowledge, information and belief.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData sharing statement (see below):\u003c/strong\u003e The datasets used and/or analyses during the current study available from the corresponding author on reasonable request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eHassan MR, Al-Insaif S, Hossain MI, Kamruzzaman J. A machine learning approach for prediction of pregnancy outcome following IVF treatment. Neural computing and applications. 2020 Apr;32(7):2283-97.\u003c/li\u003e\n\u003cli\u003eYiğit P, Bener A, Karabulut S. Comparison of machine learning classification techniques to predict implantation success in an IVF treatment cycle. Reproductive BioMedicine Online. 2022 Nov 1;45(5):923-34.\u003c/li\u003e\n\u003cli\u003eBlank C, Wildeboer RR, DeCroo I, Tilleman K, Weyers B, De Sutter P, Mischi M, Schoot BC. Prediction of implantation after blastocyst transfer in in vitro fertilization: a machine-learning perspective. Fertility and sterility. 2019 Feb 1;111(2):318-26.\u003c/li\u003e\n\u003cli\u003eLiu L, Jiao Y, Li X, Ouyang Y, Shi D. Machine learning algorithms to predict early pregnancy loss after in vitro fertilization-embryo transfer with fetal heart rate as a strong predictor. Computer Methods and Programs in Biomedicine. 2020 Nov 1;196:105624.\u003c/li\u003e\n\u003cli\u003eOseguera-L\u0026oacute;pez I, Ruiz-D\u0026iacute;az S, Ramos-Ibeas P, P\u0026eacute;rez-Cerezales S. Novel techniques of sperm selection for improving IVF and ICSI outcomes. Frontiers in cell and developmental biology. 2019 Nov 29;7:298.\u003c/li\u003e\n\u003cli\u003eG\u0026uuml;venir HA, Misirli G, Dilbaz S, Ozdegirmenci O, Demir B, Dilbaz B. Estimating the chance of success in IVF treatment using a ranking algorithm. Medical \u0026amp; biological engineering \u0026amp; computing. 2015 Sep;53:911-20.\u003c/li\u003e\n\u003cli\u003eG\u0026uuml;venir HA, Misirli G, Dilbaz S, Ozdegirmenci O, Demir B, Dilbaz B. Estimating the chance of success in IVF treatment using a ranking algorithm. Medical \u0026amp; biological engineering \u0026amp; computing. 2015 Sep;53:911-20.\u003c/li\u003e\n\u003cli\u003eHariton E, Pavlovic Z, Fanton M, Jiang VS. Applications of artificial intelligence in ovarian stimulation: a tool for improving efficiency and outcomes. Fertility and sterility. 2023 Jul 1;120(1):8-16.\u003c/li\u003e\n\u003cli\u003eVinson DR, Rauchwerger AS, Karadi CA, Shan J, Warton EM, Zhang JY, Ballard DW, Mark DG, Hofmann ER, Cotton DM, Durant EJ. Clinical decision support to O ptimize C are of patients with A trial F ibrillation or flutter in the E mergency department: protocol of a stepped-wedge cluster randomized pragmatic trial (O\u0026rsquo;CAF\u0026Eacute; trial). Trials. 2023 Mar 31;24(1):246.\u003c/li\u003e\n\u003cli\u003eHanassab S, Abbara A, Yeung AC, Voliotis M, Tsaneva-Atanasova K, Kelsey TW, Trew GH, Nelson SM, Heinis T, Dhillo WS. The prospect of artificial intelligence to personalize assisted reproductive technology. npj Digital Medicine. 2024 Mar 1;7(1):55.\u003c/li\u003e\n\u003cli\u003eTugwell P, Tovey D. PRISMA 2020. Journal of Clinical Epidemiology. 2021 Jun 1;134:A5-6.\u003c/li\u003e\n\u003cli\u003eJordan Z, Lockwood C, Munn Z, Aromataris E. The updated Joanna Briggs Institute model of evidence-based healthcare. JBI Evidence Implementation. 2019 Mar 1;17(1):58-71.\u003c/li\u003e\n\u003cli\u003eBadiola AC. \u003cem\u003eArtificial Intelligence (AI) for Embryo Ranking and its Use in Human Assisted Reproduction\u003c/em\u003e (Doctoral dissertation, University of Kent (United Kingdom)).\u003c/li\u003e\n\u003cli\u003eGoss DM, Vasilescu SA, Vasilescu PA, Cooke S, Kim SH, Sacks GP, Gardner DK, Warkiani ME. AI facilitated sperm detection in azoospermic samples for use in ICSI. medRxiv. 2023 Oct 25:2023-10.\u003c/li\u003e\n\u003cli\u003eBarnett-Itzhaki Z, Elbaz M, Butterman R, Amar D, Amitay M, Racowsky C, Orvieto R, Hauser R, Baccarelli AA, Machtinger R. Machine learning vs. classic statistics for the prediction of IVF outcomes. Journal of assisted reproduction and genetics. 2020 Oct;37:2405-12.\u003c/li\u003e\n\u003cli\u003eGhasemian F, Bahadori MH, Hosseini Kolkooh SZ, Esmaeili M. Using deep learning algorithm: the study of sperm head vacuoles and its correlation with protamine mRNA ratio. Cell Journal (Yakhteh). 2022 Jan 1;24(1):7-14.\u003c/li\u003e\n\u003cli\u003eKothandaraman R, Andavar S, Raj RS. A Hybrid Feature Ranking Algorithm for Assisted Reproductive Technology Outcome Prediction. Brazilian Archives of Biology and Technology. 2022 Jun 27;65:e22210605.\u003c/li\u003e\n\u003cli\u003eG\u0026uuml;venir HA, Misirli G, Dilbaz S, Ozdegirmenci O, Demir B, Dilbaz B. Estimating the chance of success in IVF treatment using a ranking algorithm. Medical \u0026amp; biological engineering \u0026amp; computing. 2015 Sep;53:911-20. \u003c/li\u003e\n\u003cli\u003eChen F, Chen Y, Mai Q. Multi-Omics Analysis and Machine Learning Prediction Model for Pregnancy Outcomes After Intracytoplasmic Sperm Injection\u0026ndash;in vitro Fertilization. Frontiers in Public Health. 2022 Jun 30;10:924539.\u003c/li\u003e\n\u003cli\u003eHandayani N, Louis CM, Erwin A, Aprilliana T, Polim AA, Sirait B, Boediono A, Sini I. Machine Learning Approach to Predict Clinical Pregnancy Potential in Women Undergoing IVF Program. Fertility \u0026amp; Reproduction. 2022 Jun 10;4(02):77-87. S\u003c/li\u003e\n\u003cli\u003eWang CW, Kuo CY, Chen CH, Hsieh YH, Su EC. Predicting clinical pregnancy using clinical features and machine learning algorithms in in vitro fertilization. Plos one. 2022 Jun 8;17(6):e0267554.\u003c/li\u003e\n\u003cli\u003eYou JB, McCallum C, Wang Y, Riordon J, Nosrati R, Sinton D. Machine learning for sperm selection. Nature Reviews Urology. 2021 Jul;18(7):387-403.\u003c/li\u003e\n\u003cli\u003eNguyen Thanh T, Nguyen DM, Dinh Le T, Ngoc Do L, Tien Nguyen S, Nguyen Minh P, Nguyen Van P, Minh Bui T, Thi Bui TT, Nguyen Dao H, Trung Nguyen K. The Relationship Between Smooth Endoplasmic Reticulum Clusters in Metaphase II Oocytes and Embryological and Birth Outcomes in Infertile Couples. International Journal of General Medicine. 2024 Dec 31:3269-77.\u003c/li\u003e\n\u003cli\u003eBarnes J, Brendel M, Gao VR, Rajendran S, Kim J, Li Q, Malmsten JE, Sierra JT, Zisimopoulos P, Sigaras A, Khosravi P. A non-invasive artificial intelligence approach for the prediction of human blastocyst ploidy: a retrospective model development and validation study. The Lancet Digital Health. 2023 Jan 1;5(1):e28-40.S\u003c/li\u003e\n\u003cli\u003eWang G, Wang K, Gao Y, Chen L, Gao T, Ma Y, Jiang Z, Yang G, Feng F, Zhang S, Gu Y. A generalized AI system for human embryo selection covering the entire IVF cycle via multi-modal contrastive learning. Patterns. 2024 May 2.\u003c/li\u003e\n\u003cli\u003eRaef B, Maleki M, Ferdousi R. Computational prediction of implantation outcome after embryo transfer. Health informatics journal. 2020 Sep;26(3):1810-26.\u003c/li\u003e\n\u003cli\u003eSato T, Kishi H, Murakata S, Hayashi Y, Hattori T, Nakazawa S, Mori Y, Hidaka M, Kasahara Y, Kusuhara A, Hosoya K. A new deep‐learning model using YOLOv3 to support sperm selection during intracytoplasmic sperm injection procedure. Reproductive Medicine and Biology. 2022 Jan;21(1):e12454.\u003c/li\u003e\n\u003cli\u003eHassan MR, Al-Insaif S, Hossain MI, Kamruzzaman J. A machine learning approach for prediction of pregnancy outcome following IVF treatment. Neural computing and applications. 2020 Apr;32(7):2283-97.\u003c/li\u003e\n\u003cli\u003eLiu X, Chen Z, Ji Y. Construction of the machine learning-based live birth prediction models for the first in vitro fertilization pregnant women. BMC Pregnancy and Childbirth. 2023 Jun 27;23(1):476.\u003c/li\u003e\n\u003cli\u003eCao SS, Liu XM, Song BT, Hu YY. Interpretable machine learning models for predicting clinical pregnancies associated with surgical sperm retrieval from testes of different etiologies: a retrospective study. BMC urology. 2024 Jul 29;24(1):156.\u003c/li\u003e\n\u003cli\u003eXi Q, Yang Q, Wang M, Huang B, Zhang B, Li Z, Liu S, Yang L, Zhu L, Jin L. Individualized embryo selection strategy developed by stacking machine learning model for better in vitro fertilization outcomes: an application study. Reproductive biology and endocrinology. 2021 Dec;19:1-0.\u003c/li\u003e\n\u003cli\u003eMa BX, Zhao GN, Yi ZF, Yang YL, Jin L, Huang B. Enhancing clinical utility: deep learning-based embryo scoring model for non-invasive aneuploidy prediction. Reproductive Biology and Endocrinology. 2024 May 22;22(1):58.\u003c/li\u003e\n\u003cli\u003eWang J, Jin Y, Jiang A, Chen W, Shan G, Gu Y, Ming Y, Li J, Yue C, Huang Z, Librach C. Testing the generalizability and effectiveness of deep learning models among clinics: sperm detection as a pilot study. Reproductive Biology and Endocrinology. 2024 May 22;22(1):59.\u003c/li\u003e\n\u003cli\u003eXiao YH, Hu YL, Lv XY, Huang LJ, Geng LH, Liao P, Ding YB, Niu CC. The construction of machine learning-based predictive models for high-quality embryo formation in poor ovarian response patients with progestin-primed ovarian stimulation. Reproductive Biology and Endocrinology. 2024 Jul 10;22(1):78.\u003c/li\u003e\n\u003cli\u003eGoyal A, Kuchana M, Ayyagari KP. Machine learning predicts live-birth occurrence before in-vitro fertilization treatment. Scientific reports. 2020 Dec 1;10(1):20925. \u003c/li\u003e\n\u003cli\u003eMehrjerd A, Dehghani T, Jajroudi M, Eslami S, Rezaei H, Ghaebi NK. Ensemble machine learning models for sperm quality evaluation concerning success rate of clinical pregnancy in assisted reproductive techniques. Scientific Reports. 2024 Oct 16;14(1):24283. \u003c/li\u003e\n\u003cli\u003eSene AA, Zandieh Z, Soflaei M, Torshizi HM, Sheibani K. Using artificial intelligence to predict the intrauterine insemination success rate among infertile couples. Middle East Fertility Society Journal. 2021 Dec 15;26(1):46. \u003c/li\u003e\n\u003cli\u003eSujata PN, Madiwalar SM, Aparanji VM. Machine learning techniques to improve the success rate in in-vitro fertilization (IVF) procedure. InIOP Conference Series: Materials Science and Engineering 2020 Sep 1 (Vol. 925, No. 1, p. 012039). IOP Publishing. \u003c/li\u003e\n\u003cli\u003eCanosa S, Licheri N, Bergandi L, Gennarelli G, Paschero C, Beccuti M, Cimadomo D, Coticchio G, Rienzi L, Benedetto C, Cordero F. A novel machine-learning framework based on early embryo morphokinetics identifies a feature signature associated with blastocyst development. Journal of Ovarian Research. 2024 Mar 15;17(1):63.\u003c/li\u003e\n\u003cli\u003eKarami N, Iravani F, Bavarsad SB, Asadollahi S, Hoseini SM, Montazeri F, Kalantar SM. Comparing the advantages, disadvantages and diagnostic power of different non-invasive pre-implantation genetic testing: A literature review. International Journal of Reproductive BioMedicine (IJRM). 2024 May 12:177-90.\u003c/li\u003e\n\u003cli\u003eVon Wolff M. The role of natural cycle IVF in assisted reproduction. Best practice \u0026amp; research Clinical endocrinology \u0026amp; metabolism. 2019 Feb 1;33(1):35-45.\u003c/li\u003e\n\u003cli\u003eGiacobbe M, Conatti M, Gomes A, Bonetti TC, Monteleone PA. Effectivity of conventional in vitro fertilization (IVF) and intracytoplasmic sperm injection (ICSI) when male factor is absent: a perspective point of view. JBRA Assisted Reproduction. 2022 Jan;26(1):123.\u003c/li\u003e\n\u003cli\u003evan den Hoven L, Hendriks JC, Verbeet JG, Westphal JR, Wetzels AM. Status of sperm morphology assessment: an evaluation of methodology and clinical value. Fertility and sterility. 2015 Jan 1;103(1):53-8.\u003c/li\u003e\n\u003cli\u003eKragh MF, Karstoft H. Embryo selection with artificial intelligence: how to evaluate and compare methods?. Journal of assisted reproduction and genetics. 2021 Jul;38(7):1675-89.\u003c/li\u003e\n\u003cli\u003eVerMilyea M, Hall JM, Diakiw SM, Johnston A, Nguyen T, Perugini D, Miller A, Picou A, Murphy AP, Perugini M. Development of an artificial intelligence-based assessment model for prediction of embryo viability using static images captured by optical light microscopy during IVF. Human Reproduction. 2020 Apr 28;35(4):770-84.\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Table 4","content":"\u003cp\u003eTable 4 is available in the Supplementary Files section.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"discover-artificial-intelligence","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"diai","sideBox":"Learn more about [Discover Artificial Intelligence](https://www.springer.com/44163)","snPcode":"","submissionUrl":"","title":"Discover Artificial Intelligence","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Discover Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-5504223/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5504223/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eImportance:\u003c/strong\u003e Assisted Reproductive Technologies (ART) have been developed to address infertility by improving embryo selection. Artificial intelligence (AI), using Time-Lapse Imaging (TLI), enhances predictions from fertilization to the blastocyst stage.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eObjective:\u003c/strong\u003e Studies show AI can identify suitable embryos more effectively than specialists, improving in-vitro fertilization (IVF) success rates by enhancing transfer success and reducing miscarriage risks. With IVF success rates below 40%, it is essential to explore AI methods to boost outcomes\u003cstrong\u003e.\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFindings:\u003c/strong\u003e A systematic review in October 2024 searched databases like PubMed and Scopus using terms related to IVF and AI, excluding non-English and qualitative studies. Twenty-seven studies were reviewed; 17 predicted treatment responses with deep learning. Two studies used neural networks for successful treatment prediction, and eight employed ML methods such as NB, SVM, and RF, with an average AUC of 0.91. Models showed 90-96% accuracy, sensitivity, and precision.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion:\u003c/strong\u003e AI technologies, particularly NB and Reinforcement Learning, show promise in improving IVF outcomes by enhancing classification and diagnosis while saving time. Interdisciplinary approaches using micro and Nano-biotechnology can help overcome clinical challenges\u003cstrong\u003e.\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRelevance:\u003c/strong\u003e Examining the quality of sperm and egg separately using AI could further improve fertility testing and success in ART, optimizing clinical results.\u003c/p\u003e","manuscriptTitle":"Investigating the Artificial intelligence in prediction and evaluation sperm and embryo quality in In vitro fertilization (IVF): A systematic review","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-12-25 15:41:22","doi":"10.21203/rs.3.rs-5504223/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-02-19T11:11:33+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-01-20T09:24:03+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"166447101056801396677544457061656534296","date":"2025-01-18T13:13:06+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-01-02T08:26:57+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"312593273446643391784919411712319063517","date":"2024-12-27T18:13:51+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"51171745430856369013727326846953709202","date":"2024-12-26T14:37:36+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-12-23T12:51:18+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-12-19T08:19:17+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-12-18T15:57:19+00:00","index":"","fulltext":""},{"type":"submitted","content":"Discover Artificial Intelligence","date":"2024-11-22T11:15:02+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"discover-artificial-intelligence","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"diai","sideBox":"Learn more about [Discover Artificial Intelligence](https://www.springer.com/44163)","snPcode":"","submissionUrl":"","title":"Discover Artificial Intelligence","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Discover Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"e18712da-c94b-4171-829d-ca0ada1930a9","owner":[],"postedDate":"December 25th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-08-04T16:44:40+00:00","versionOfRecord":{"articleIdentity":"rs-5504223","link":"https://doi.org/10.1007/s44163-025-00420-8","journal":{"identity":"discover-artificial-intelligence","isVorOnly":false,"title":"Discover Artificial Intelligence"},"publishedOn":"2025-07-29 16:13:31","publishedOnDateReadable":"July 29th, 2025"},"versionCreatedAt":"2024-12-25 15:41:22","video":"","vorDoi":"10.1007/s44163-025-00420-8","vorDoiUrl":"https://doi.org/10.1007/s44163-025-00420-8","workflowStages":[]},"version":"v1","identity":"rs-5504223","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5504223","identity":"rs-5504223","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}