Prerpints.org logo
Preprint
Review

This version is not peer-reviewed.

A Comprehensive Review of Artificial Intelligence-Driven Enhancements in Non-Invasive Prenatal Testing: Advancing Genomic Precision Through Deep Learning and Computational Genomics

Submitted:

24 June 2025

Posted:

26 June 2025

You are already at the latest version

Abstract
Non-Invasive Prenatal Testing (NIPT) has emerged as a pivotal tool in prenatal care by enabling early, risk-free detection of fetal chromosomal anomalies. However, despite its clinical significance, challenges such as limited sensitivity in low fetal fraction samples and a narrow detection spectrum persist.Current methodologies largely depend on statistical techniques and conventional sequencing interpretation, which may not fully exploit the complex and high-dimensional nature of cffDNA data. There is a lack of comprehensive analysis on how Artificial Intelligence (AI) can systematically address these limitations and enhance the accuracy and scope of NIPT.This review synthesizes existing research to evaluate the role of AI particularly machine learning and deep learning models in improving NIPT accuracy, reducing false positives, and expanding the spectrum of detectable conditions.An extensive literature review was conducted on AI applications in NIPT, including studies using Convolutional Neural Networks (CNNs), Support Vector Machines (SVMs), Random Forests, and ensemble methods. The analysis includes performance comparisons, case studies, and evaluations of emerging trends such as multi-omics integration and explainable AI.AI has demonstrated superior accuracy and predictive power over traditional models in detecting common aneuploidies like trisomy 21, 18, and 13. AI also shows potential in detecting rarer conditions, improving diagnostic clarity, and minimizing "no-call" results.This review highlights AI’s transformative potential in prenatal diagnostics. Continued interdisciplinary research and ethical oversight are essential to guide the integration of AI into clinical practice.
Keywords: 
artificial intelligence
;  
non-invasive prenatal testing
;  
machine learning
;  
deep learning
;  
genomic diagnostics
Subject: 
Computer Science and Mathematics  -   Mathematical and Computational Biology

Introduction

Prenatal screening has long been a cornerstone of modern obstetrical care, aiming to identify pregnancies at an increased risk for fetal anomalies. Traditional methods, such as maternal serum screening and ultrasound, have played a crucial role in this endeavor. However, these approaches often suffer from limitations, including relatively high false-positive rates, which can lead to unnecessary anxiety for expectant parents and the need for invasive diagnostic procedures like amniocentesis or chorionic villus sampling, both carrying a small risk of miscarriage [1]. The development of Non-Invasive Prenatal Testing (NIPT) marked a significant advancement in prenatal screening. By analyzing cell-free fetal DNA (cfDNA) circulating in the mother's blood, NIPT offers a safer and more accurate method for assessing the risk of certain fetal chromosomal abnormalities[1]. The clinical utility and prevalence of NIPT have steadily increased since its introduction, becoming an integral part of prenatal care pathways for many women[4]. As the adoption of NIPT continues to grow, researchers and clinicians are exploring innovative ways to further enhance its accuracy and expand its capabilities. One promising avenue of research lies in the integration of Artificial Intelligence (AI) to analyze the complex datasets generated by NIPT, potentially leading to even more reliable and comprehensive prenatal assessments[11]. The transition from traditional screening to NIPT effectively addressed the issues of invasiveness and improved accuracy for common aneuploidies. Now, the application of AI is being investigated to overcome the inherent limitations of current NIPT methodologies and to potentially broaden the spectrum of detectable fetal conditions. This review aims to provide a comprehensive analysis of the role of AI in enhancing the accuracy of NIPT for predicting chromosomal abnormalities also exploring the evolution of AI in NIPT, the fundamental concepts involved, current trends in AI applications, numerous use cases and case studies, the detailed working mechanisms of AI-enhanced NIPT, a country-wise analysis of adoption and research, insights from reputable reviews, and future research trends in this rapidly evolving field.

The Evolutionary Landscape of AI in NIPT

The foundation for NIPT was laid with the groundbreaking discovery of cell-free DNA (cfDNA) in maternal plasma, including a fraction originating from the fetus[1]. This pivotal finding in the late 20th century opened up the possibility of non-invasively accessing the fetal genome. The advent of high-throughput sequencing (NGS) technologies proved to be the key enabler for NIPT, allowing for the rapid and cost-effective analysis of millions of cfDNA fragments[1]. Initial approaches to NIPT involved massive parallel shotgun sequencing combined with chromosome counting to identify differences in the quantity of DNA from specific chromosomes, indicative of aneuploidy[1]. Subsequent advancements led to the development of targeted sequencing methods and the analysis of single nucleotide polymorphisms (SNPs) to differentiate between maternal and fetal DNA, improving the efficiency and sensitivity of the testing.[1] Over time, the increasing affordability and accessibility of NGS have been significant factors in the wider adoption of NIPT in clinical practice[1]. Parallel to these developments in NIPT, the field of prenatal diagnostics also witnessed the gradual integration of computational methods to analyze the growing volumes of data. This included the early use of statistical approaches and, increasingly, machine learning techniques to interpret the complex datasets generated by NGS in NIPT[12]. Before the direct application of AI to NIPT, prenatal diagnostics saw the emergence of AI in other areas, most notably in the analysis of ultrasound images to detect fetal anomalies[11]. The specific evolution of AI applications to cfDNA analysis in NIPT began with the application of traditional statistical methods to identify deviations in chromosome representation. As the complexity of genomic data increased and the potential for detecting more subtle abnormalities became apparent, the field transitioned towards incorporating more sophisticated machine learning algorithms and, eventually, deep learning techniques[12]. This progression reflects the broader advancements in both genomics and artificial intelligence, with increasingly powerful computational tools being applied to extract meaningful insights from the vast amounts of data generated by NIPT.

Fundamental Concepts of AI-Enhanced Cell-Free Fetal DNA Analysis

At the core of AI-enhanced NIPT lies the analysis of cell-free DNA (cfDNA), which refers to small fragments of DNA circulating in the bloodstream.[2] During pregnancy, a portion of this cfDNA originates from the placenta and is known as cell-free fetal DNA (cffDNA).[2] CffDNA exhibits unique characteristics, including specific fragmentation patterns and a slightly smaller average size compared to maternal cfDNA.[2] The proportion of cffDNA in the mother's blood, known as the fetal fraction, typically ranges from 10% to 20% during the first and second trimesters.[2] Non-invasive prenatal testing relies on high-throughput sequencing (NGS) to generate massive datasets of these cffDNA fragments[1]. These sequencing data provide information about the abundance of DNA from different chromosomes, which can be used to detect chromosomal abnormalities such as trisomy 21 (Down syndrome), trisomy 18 (Edwards syndrome), and trisomy 13 (Patau syndrome). Artificial Intelligence (AI) and Machine Learning (ML) play a crucial role in analyzing this complex sequencing data.[12] Machine learning, a subset of AI, involves the development of algorithms that can learn from data without being explicitly programmed. These algorithms can be broadly categorized into supervised learning, where the model learns from labeled data (e.g., samples with known chromosomal status); unsupervised learning, where the model identifies patterns in unlabeled data; and deep learning, a subfield of machine learning that uses artificial neural networks with multiple layers to analyze complex data. Key algorithms commonly employed in AI-enhanced NIPT include Linear Regression and Logistic Regression for modeling relationships and predicting outcomes; Support Vector Machines (SVMs) for classification tasks, such as distinguishing between euploid and aneuploid samples; Random Forests, an ensemble learning method that constructs multiple decision trees; and Neural Networks, including Convolutional Neural Networks (CNNs) particularly effective for analyzing patterns in sequential or grid-like data (like DNA sequences or binned genomic data), and Recurrent Neural Networks (RNNs) suitable for analyzing sequential data.[27] AI/ML algorithms analyze cffDNA data by extracting relevant features from the sequencing reads. These features can include the number of reads mapping to each chromosome or specific genomic regions, the distribution of fragment lengths, the presence of epigenetic markers like methylation patterns, and variations in the DNA sequence such as single nucleotide polymorphisms (SNPs).[2] The algorithms then learn to recognize complex patterns in this high-dimensional genomic data that are indicative of chromosomal abnormalities. For instance, an AI model might learn that a subtle deviation in the ratio of reads mapping to chromosome 21 compared to other chromosomes, in combination with specific fragment length distributions, is highly predictive of trisomy 21. These are primarily classification tasks, where the AI aims to classify a sample as either euploid (normal chromosome number) or aneuploid (abnormal chromosome number). Regression tasks are also employed, for example, to accurately estimate the fetal fraction in a maternal blood sample, which is crucial for the reliability of NIPT. The development of robust AI models for NIPT requires high-quality data, accurate estimation of fetal fraction, and the use of appropriate control samples to account for biological and technical variations.[39]

Current Trends in AI Applications for NIPT Accuracy

The field of AI applications in NIPT is rapidly evolving, with several key trends emerging in the pursuit of enhanced accuracy. Convolutional Neural Networks (CNNs) are increasingly being utilized for their ability to analyze patterns in data, including fragment patterns derived from cffDNA sequencing and potentially even raw sequencing data itself.[11] Support Vector Machines (SVMs) remain a popular choice for classification problems in NIPT, effectively distinguishing between samples with and without chromosomal abnormalities based on various extracted features.[13] Beyond CNNs, other Deep Learning (DL) algorithms, such as Deep Neural Networks (DNNs) and Recurrent Neural Networks (RNNs), are gaining traction for their capacity to model more complex relationships within genomic sequences and other relevant data.[11] Random Forests are also frequently employed for both classification and prediction tasks in NIPT, leveraging their robustness and ability to handle high-dimensional data.[13] Logistic Regression continues to be a valuable tool for predicting the likelihood of chromosomal abnormalities based on identified biomarkers and features derived from sequencing data.[13] The reported efficiency of these AI algorithms in NIPT studies is often high. Accuracy rates of up to 99% for the detection of common trisomies (21, 18, 13) have been reported in various academic research papers.[7] Studies also consistently report high sensitivity and specificity for different algorithms and specific chromosomal abnormalities.[7] The Positive Predictive Value (PPV) is increasingly recognized as a critical metric for assessing the clinical utility of NIPT, and AI-enhanced methods often demonstrate improved PPVs compared to traditional approaches.[14] Direct comparisons between AI-enhanced NIPT and traditional NIPT methods frequently highlight the superior performance of AI in terms of accuracy, sensitivity, specificity, and PPV.[13] Several potential alternative algorithms and emerging trends are shaping the future of AI in NIPT. Fragment distance-based methods, often coupled with CNNs, have shown promising results by analyzing the distribution of cfDNA fragment lengths rather than just the counts.[15] Deep learning is being explored for advanced applications like fetal genotyping, potentially paving the way for comprehensive genome-wide NIPT.[51] The integration of multi-omics data, combining genomic, transcriptomic, and proteomic information, with AI algorithms is another active area of research aimed at enhancing biomarker discovery and diagnostic accuracy.[14] Ensemble methods, which combine the predictions of multiple AI algorithms, are being investigated to improve the overall robustness and performance of NIPT.[15] Self-learning neural networks are also being explored for specialized tasks, such as the detection of fetal single nucleotide variants (SNVs).[16] These trends collectively suggest a continued movement towards leveraging more sophisticated AI algorithms, particularly within the realm of deep learning, and the incorporation of a broader range of data beyond simple read counts to achieve the highest possible accuracy and expand the clinical utility of NIPT.
Table 1. Comparison of AI Algorithms and Their Reported Efficiency in NIPT.
Table 1. Comparison of AI Algorithms and Their Reported Efficiency in NIPT.
Algorithm Name Type Targeted Abnormality Reported Sensitivity Reported Specificity Reported PPV Snippet IDs
aiD-Ensemble CNN-based T21, T18, T13 99.07% >99.40% accuracy 88.43% [15], B2, B18
aiD-Mean CNN-based T21, T18, T13 99.07% >99.40% accuracy 87.70% [38], B2
aiD-IQR CNN-based T21, T18, T13 98.15% >99.40% accuracy 80.92% [38], B2
aiD-Median CNN-based T21, T18, T13 97.22% >99.40% accuracy 67.74% [38], B2
DeepHoobari DNN Fetal Genotyping High (surpasses Hoobari) Not Specified Not Specified [51], B6, B10
Random Forest ML Down Syndrome 85.2% (validation) 95% Not Specified [14]
Convolutional Neural Network DL Down Syndrome 96.72% Sensitivity, 98.45% Specificity Not Specified Not Specified [14]
Artificial Neural Network DL Aneuploidy (T21, T18, T13) 100% for T21, >80% for others Minimal False Positives High Detection Rates [14]
WisecondorX Statistical T13, T18, T21 (non-mosaic) 100% 98.5% Not Specified [39]
VINIPT Statistical T13, T18, T21 (non-mosaic) 100% 99.9% Not Specified [39]
NIPT-PG Statistical + Pan-Genome Chromosomal Aneuploidies Outperforms standard Z-score Not Specified Not Specified [41]

Use Cases and Case Studies of AI in NIPT

Numerous real-world examples and case studies highlight the improved accuracy achieved through the application of AI in detecting common chromosomal abnormalities such as trisomy 21, 18, and 13. Studies have demonstrated the superior performance of AI algorithms, including CNN-based methods like aiD-NIPT, compared to traditional Z-score based NIPT analysis.[38] For instance, the aiD-NIPT algorithm, utilizing fragment distance and convolutional neural networks, has shown higher sensitivity and positive predictive value for these trisomies in large clinical cohorts.[15] In some instances, AI models have correctly identified aneuploidies that were missed by conventional NIPT methods, underscoring the potential of AI to enhance diagnostic accuracy.[38] Furthermore, AI has played a significant role in reducing both false positive and false negative results in NIPT. Studies have reported lower false positive rates with AI algorithms compared to traditional methods, which is crucial in minimizing the need for invasive confirmatory testing and the associated anxiety for expectant parents.[14]AI models have also demonstrated improved sensitivity, leading to a reduction in false negatives and ensuring more accurate detection of true positive cases.[15]Beyond the common trisomies, AI is being increasingly applied to expand the scope of NIPT to detect rarer genetic conditions and microdeletions. Case studies have explored the use of AI-enhanced NIPT for the detection of less common chromosomal abnormalities and microdeletion syndromes, such as 22q11.2 deletion syndrome.[53] Panorama AI serves as a specific case study, where artificial intelligence is leveraged to improve the accuracy of result calling and significantly reduce the rate of inconclusive results, known as "no-calls".[53] This AI-powered approach has also demonstrated increased accuracy in screening for the 22q11.2 deletion syndrome, even detecting smaller deletions that many other NIPTs cannot identify.[53] Research is also actively exploring the application of AI-enhanced NIPT for the detection of monogenic disorders, further broadening the clinical utility of this non-invasive screening method.[3] These real-world examples and case studies collectively illustrate the tangible benefits of integrating AI into NIPT, leading to more accurate detection of common aneuploidies, expansion of the test's scope to include rarer conditions, and a reduction in inconclusive results, ultimately enhancing the clinical value of NIPT in prenatal care.

Detailed Working Mechanisms of AI-Enhanced NIPT

The process of AI-enhanced NIPT involves a series of intricate steps, starting with the acquisition of biological samples and culminating in the generation of a risk assessment. The initial stage involves the collection of a maternal blood sample, from which plasma is isolated.[2] Cell-free DNA (cfDNA), including the crucial cell-free fetal DNA (cffDNA), is then extracted from this plasma.[2] Following DNA extraction, the cfDNA undergoes library preparation, a process that prepares the DNA fragments for next-generation sequencing (NGS).[1,41,42,43,44,45,46,47,48,49] NGS is then performed, generating millions of short DNA sequence reads.[1] These reads are subsequently aligned to a reference human genome.[38] To ensure the quality of the data, several preprocessing steps are typically performed, including filtering out low-quality reads, removing duplicate reads that may arise during PCR amplification, and applying corrections for biases such as those related to GC content.[39] Once the data is preprocessed, the crucial stage of data analysis begins. This involves extracting relevant features from the aligned sequencing reads.[2] These features can include the number of reads mapping to each chromosome or specific genomic bins, the statistical distribution of the lengths of the sequenced fragments, the presence of single nucleotide polymorphisms (SNPs), and even patterns of DNA methylation.[2,49,50,51,52,53,54,55,56] Feature selection and engineering play a vital role in identifying and transforming the most informative features to train effective AI models.[24] The processed data is then typically divided into training, validation, and testing sets. The training set is used to teach the AI model to recognize patterns associated with different chromosomal conditions. The validation set helps in tuning the model's parameters and preventing overfitting, while the testing set provides an independent evaluation of the model's performance.[13,57,58,59,60,61,62,63,64,65,66,67,68,69] Various types of AI models are employed in this analysis. Convolutional Neural Networks (CNNs) are often used to analyze patterns in binned genomic data or even visual representations of fragment distributions, leveraging their ability to learn hierarchical features through convolutional layers, pooling layers, and fully connected layers that perform the final classification.[13] Support Vector Machines (SVMs) are used to find optimal hyperplanes that separate samples into different classes based on the extracted features.[13] Deep learning models, including DNNs and RNNs, are capable of learning intricate patterns from large datasets, often outperforming traditional machine learning methods in complex tasks such as sequence analysis and the integration of diverse data types.[11] The AI models are trained using labeled data, where each sample is associated with a known chromosomal status (e.g., euploid or having a specific aneuploidy). This allows the model to learn the statistical relationships between the input features and the output class.[13] The performance of the trained AI models is then validated using independent datasets to ensure their generalizability to new, unseen samples and to prevent overfitting, where the model learns the training data too well and performs poorly on new data.[13] In many AI-enhanced NIPT workflows, traditional statistical measures like Z-scores, which quantify the deviation of chromosome representation from expected values, are often used in conjunction with AI models to provide a more robust and interpretable result.[15]

Country-Wise Analysis of AI-Enhanced NIPT

The adoption and research landscape of AI-enhanced NIPT exhibits significant variations across different countries, reflecting diverse healthcare policies, research priorities, and ethical considerations. In the USA, NIPT has been widely implemented in clinical practice; however, there is no national consensus policy guiding its use.[8] Europe presents a more varied picture, with countries like Belgium and the Netherlands offering NIPT to all pregnant women as part of their national programs, while most other European nations have implemented it as an option primarily for women identified as being at higher risk following first-trimester screening.[6,69,70,71,72,73,74,75,76,77,78,79] The Asia-Pacific region is witnessing increasing adoption of NIPT, particularly in China and Japan. In China, while NIPT has gained wide attention and clinical guidelines have been published, complete government sponsorship is not yet established, with coverage varying by region.[8] Japan introduced NIPT in 2013, recommending it primarily for pregnant women with known high risks of fetal abnormalities, and it is mostly paid out-of-pocket.[9] Croatia has seen increasing efficiency and accessibility of NIPT within its private healthcare system.[6] Italy is actively considering the incorporation of NIPT into its publicly funded national healthcare system.[61] Several countries are at the forefront of research initiatives in AI-enhanced NIPT. South Korea has been a key contributor, with the development and evaluation of the novel aiD-NIPT algorithm that utilizes cfDNA fragment distance and convolutional neural networks.[42] China has a robust research landscape in NIPT, with numerous studies focusing on the application of machine learning and deep learning algorithms to improve accuracy and expand the scope of testing.[9] In the Netherlands, alongside high adoption rates, research is being conducted to understand the impact of specific genetic variants in pregnant women on the accuracy of NIPT results.[61] Croatia has undertaken its first national monocentric study to analyze the usage patterns and outcomes of NIPT over a decade, aiming to inform the country's future prenatal care strategy.[6] In Europe, a collaborative research project involving England, France, and Germany is specifically exploring the ethical issues arising from the implementation of NIPT in these different socio-cultural contexts.[71] The adoption and research trends in AI-enhanced NIPT are significantly influenced by the healthcare infrastructure and regulatory environments of different countries. The presence of supportive healthcare policies, the extent of insurance coverage for prenatal screening, and government-driven initiatives play a crucial role in the widespread adoption of NIPT.[1] Furthermore, ethical and societal values regarding prenatal testing, disability, and reproductive choices also shape the implementation and public perception of AI-enhanced NIPT in various regions.[62]
Table 2. Country-Wise Analysis of AI-Enhanced NIPT Adoption and Research.
Table 2. Country-Wise Analysis of AI-Enhanced NIPT Adoption and Research.
Country Adoption Level/Policy Key Research Initiatives/Studies Healthcare/Regulatory Factors
USA Wide implementation, no national policy Research focused on improving accuracy and expanding scope Primarily private market, increasing insurance coverage
Netherlands Universal offer Research on genetic variants affecting NIPT accuracy Strong public healthcare system, high NIPT uptake
Belgium Universal offer Not specified in snippets Strong public healthcare system, very high NIPT uptake
China Increasing adoption, partial government coverage Development of aiD-NIPT, numerous studies on AI in NIPT Growing healthcare market, varying insurance coverage by region
Japan Recommended for high-risk, mostly out-of-pocket Nationwide clinical study to evaluate NIPT Primarily private payment, specific guidelines for NIPT use
South Korea Not specified in snippets Development of aiD-NIPT algorithm Advanced healthcare system, GC Genome research center
Croatia Private healthcare setting Monocentric study on NIPT usage patterns NIPT not included in national prenatal care strategy
Italy Considering nationwide public funding Focus on contingent implementation and cost-effectiveness Regional variations in public healthcare coverage
England Second-tier test (after risk assessment) Research project on ethical implications of NIPT National Health Service, ethical guidelines
France Second-tier test (after risk assessment) Research project on ethical implications of NIPT Public healthcare system, ethical guidelines
Germany Second-tier test (after risk assessment) Research project on ethical implications of NIPT Public healthcare system, varying insurance coverage

A Synthesis of Existing Literature on AI in NIPT

A comprehensive review of the scientific and medical literature reveals a strong consensus on the transformative potential of Artificial Intelligence in enhancing the accuracy and capabilities of Non-Invasive Prenatal Testing.[11] These reviews consistently highlight the effectiveness of various AI and machine learning algorithms, including Convolutional Neural Networks (CNNs), Support Vector Machines (SVMs), deep learning models, and Random Forests, in analyzing the complex data generated by cffDNA sequencing.[11] A key finding across multiple reviews is the ability of AI to not only improve the detection rates for common aneuploidies like trisomy 21, 18, and 13, but also to expand the scope of NIPT to potentially identify rarer genetic conditions and microdeletions.[3] Many reviews emphasize the importance of integrating diverse data sources beyond just genomic information, such as ultrasound imaging and maternal health records, with AI algorithms to achieve a more holistic and accurate prenatal risk assessment.[11]While the potential benefits of AI in NIPT are substantial, the existing literature also acknowledges several limitations and challenges. These include the need for larger and more diverse datasets to train robust AI models, the importance of addressing potential biases within the algorithms to ensure equitable performance across different populations, and the crucial ethical considerations surrounding the use of AI in such a sensitive area of healthcare.[5] The reviews also highlight key challenges for future research, such as the need for improved interpretability of AI models to facilitate clinical trust and adoption, the development of more sophisticated algorithms capable of handling complex genomic data, and the establishment of clear regulatory frameworks for the development and deployment of AI-enhanced NIPT.5 Overall, the existing body of literature robustly supports the significant role of AI in revolutionizing NIPT, offering the potential for substantial improvements in prenatal screening and diagnosis. However, it also underscores the importance of ongoing research and a careful, ethical approach to the integration of these powerful technologies into clinical practice.

Future Directions and the Scope of Innovation in AI for NIPT

The future of AI in NIPT is poised for significant advancements, driven by ongoing research and technological innovation. Potential advancements in algorithms and techniques include the development of even more sophisticated deep learning architectures designed for enhanced feature extraction and classification from complex genomic data.[13] Researchers are also focusing on improving the performance of AI models in analyzing samples with low fetal fraction, which can be challenging for traditional NIPT methods.[16,17,18,19,20,21,22,23] The integration of explainable AI (XAI) is gaining importance to enhance the interpretability of AI models, making them more transparent and trustworthy for clinical use.[24,25,26,27,28,29,30] Furthermore, the application of federated learning, a technique that allows for collaborative AI model development using data from multiple institutions without sharing sensitive patient information, holds promise for creating more robust and generalizable models.[79,80,81,82,83,84,85,86] Several emerging technologies are also expected to impact the future of prenatal diagnostics and the role of AI within it. Advancements in sequencing technologies may provide richer and more detailed genomic data, which can be further leveraged by AI algorithms.[1,2,3,4,7,8,9,10] The potential of AI in analyzing epigenetic markers, such as DNA methylation patterns, could lead to the detection of a broader range of fetal conditions beyond chromosomal aneuploidies.[12] The integration of AI with other prenatal screening modalities, including ultrasound imaging and maternal serum markers, is likely to lead to more comprehensive and accurate risk assessment strategies.[11] AI-powered tools are also being developed to provide personalized risk prediction and to assist in genetic counseling, helping expectant parents understand complex NIPT results and make informed decisions.[14] The use of AI in remote prenatal monitoring and care is an emerging trend, with the potential to improve access to quality prenatal care, especially in underserved areas.[31,32,33,34,35] As AI continues to be integrated into NIPT, it is crucial to address the ethical considerations and challenges associated with its implementation. Data privacy and security are paramount concerns when dealing with large genomic datasets used for AI training.[19] Ensuring equitable performance of AI models across diverse populations and mitigating potential algorithmic biases are essential for responsible implementation.[19,36,37,38,39,40] Transparency and interpretability of AI models are also critical for building trust among clinicians and facilitating their adoption in clinical practice.19 The development of clear regulatory frameworks and guidelines is necessary to govern the development and deployment of AI-enhanced NIPT.[22] Finally, the role of genetic counseling will become even more important in the context of increasingly sophisticated AI-driven NIPT results, ensuring that expectant parents receive adequate support and guidance.[5] Numerous international collaborations and research projects are currently underway, focused on advancing the application of AI in prenatal testing, highlighting the global interest and the potential for future breakthroughs.[79,80,81,82,83]

Conclusion

In conclusion, Artificial Intelligence has emerged as a powerful tool with the potential to significantly enhance the accuracy and expand the capabilities of Non-Invasive Prenatal Testing for the detection of chromosomal abnormalities. By leveraging sophisticated machine learning and deep learning algorithms, AI can analyze the complex genomic data generated by NIPT with remarkable precision, often surpassing the performance of traditional statistical methods. The key trends in this field include the increasing use of convolutional neural networks, support vector machines, and other deep learning architectures, as well as the integration of diverse data types beyond simple read counts. Numerous use cases and case studies have demonstrated the tangible benefits of AI in improving the detection of common aneuploidies, reducing false positive and false negative rates, and even expanding the scope of NIPT to include rarer genetic conditions and microdeletions. The working mechanisms of AI-enhanced NIPT involve a complex pipeline, from biological sample processing and sequencing to sophisticated computational analysis where AI models learn intricate patterns indicative of chromosomal abnormalities. The adoption and research landscape of AI in NIPT vary across countries, influenced by healthcare policies, ethical considerations, and research priorities. Reputable scientific and medical reviews consistently highlight the transformative role of AI in prenatal diagnostics, underscoring its potential to revolutionize prenatal care. The future of AI in NIPT promises even greater advancements, with ongoing research focused on developing more sophisticated algorithms, improving performance in challenging scenarios like low fetal fraction, and integrating AI with other prenatal screening modalities. However, realizing the full potential of AI in NIPT necessitates careful attention to ethical considerations, including data privacy, algorithmic bias, and the need for transparency and interpretability. Continued research, coupled with responsible implementation and robust regulatory frameworks, will be crucial in ensuring that AI-enhanced NIPT translates into improved outcomes and more personalized prenatal care for expectant parents worldwide.

References

  1. CarusoPhD, A. Evolution of non-invasive prenatal testing (NIPT),” Diagnostic Precision, Dec. 20, 2022. https://blog.seracare.com/ngs/evolution-of-non-invasive-prenatal-testing-nipt-testing.
  2. PhD, R.R. The Scientist, Nov. 08, 2023. https://www.the-scientist.com/the-basics-and-applications-of-cell-free-dna-71479.
  3. McCreary, D.; McCreary, D. , “The Past, Present and Future of Non-Invasive Prenatal Testing (NIPT),” Sapio Sciences, Mar. 13, 2024. https://www.sapiosciences.com/blog/non-invasive-prenatal-testing-nipt-sapio-sciences/.
  4. Berenger, M. , “Continued Advances in NIPT Drives Updated Recommendations,” Clinical Diagnostics, Sep. 14, 2023. https://blog.lgcclinicaldiagnostics.com/nipt-evolutions.
  5. Bedei, I.; Wolter, A.; Weber, A.; Signore, F.; Axt-Fliedner, R. Chances and Challenges of New Genetic Screening Technologies (NIPT) in Prenatal Medicine from a Clinical Perspective: A Narrative Review. Genes 2021, 12, 501. [Google Scholar] [CrossRef] [PubMed]
  6. Podobnik, P.; Meštrović, T.; Đorđević, A.; Kurdija, K.; Jelčić, D.; Ogrin, N.; Bertović-Žunec, I.; Gebauer-Vuković, B.; Hočevar, G.; Lončar, I.; et al. A Decade of Non-Invasive Prenatal Testing (NIPT) for Chromosomal Abnormalities in Croatia: First National Monocentric Study to Inform Country’s Future Prenatal Care Strategy. Genes 2024, 15, 1590. [Google Scholar] [CrossRef] [PubMed]
  7. Sadvakassova, L.; Kurmangali, Z.; Beloussov, V.; Rakisheva, Z.; Terzic, M. The effectiveness of non-invasive prenatal test technology and the prenatal screening algorithm based on various methods for determining foetal aneuploidy. J. Turk. Gynecol. Assoc. 2023, 24, 152–158. [Google Scholar] [CrossRef]
  8. Ltd, M.D.F. , “Global Prenatal Testing Market Size, Share, Trends & Growth Forecast Report, Segmented By Technology (Non-Invasive (Non-Invasive Prenatal Testing (NIPT), Maternal Serum Screening and Ultrasound-Based Testing) And Invasive (Amniocentesis, Chorionic Villus Sampling (CVS) and Others), Type, Application, End-Users and Region (North America, Europe, Asia-Pacific, Latin America, Middle East and Africa), Industry Analysis (2024 to 2032),” Market Data Forecast. https://www.marketdataforecast.com/market-reports/prenatal-testing-market-report.
  9. Jayashankar, S.S.; Nasaruddin, M.L.; Hassan, M.F.; Dasrilsyah, R.A.; Shafiee, M.N.; Ismail, N.A.S.; Alias, E. Non-Invasive Prenatal Testing (NIPT): Reliability, Challenges, and Future Directions. Diagnostics 2023, 13, 2570. [Google Scholar] [CrossRef]
  10. Demko, Z.; Prigmore, B.; Benn, P. A Critical Evaluation of Validation and Clinical Experience Studies in Non-Invasive Prenatal Testing for Trisomies 21, 18, and 13 and Monosomy X. J. Clin. Med. 2022, 11, 4760. [Google Scholar] [CrossRef]
  11. Miskeen, E.; Alfaifi, J.; Alhuian, D.M.; Alghamdi, M.; Alharthi, M.H.; Alshahrani, N.A.; Alosaimi, G.; Alshomrani, R.A.; Hajlaa, A.M.; Khair, N.M.; et al. Prospective Applications of Artificial Intelligence In Fetal Medicine: A Scoping Review of Recent Updates. Int. J. Gen. Med. 2025, ume 18, 237–245. [Google Scholar] [CrossRef]
  12. Tsui, W.A.; Ding, S.C.; Jiang, P.; Lo, Y.D. Artificial intelligence and machine learning in cell-free-DNA-based diagnostics. Genome Res. 2025, 35, 1–19. [Google Scholar] [CrossRef]
  13. Yeganegi, M.; Danaei, M.; Azizi, S.; Jayervand, F.; Bahrami, R.; Dastgheib, S.A.; Rashnavadi, H.; Masoudi, A.; Shiri, A.; Aghili, K.; et al. Research advancements in the Use of artificial intelligence for prenatal diagnosis of neural tube defects. Front. Pediatr. 2025, 13, 1514447. [Google Scholar] [CrossRef]
  14. Danaei, M.; Rashnavadi, H.; Yeganegi, M.; Dastgheib, S.A.; Bahrami, R.; Azizi, S.; Jayervand, F.; Masoudi, A.; Shahbazi, A.; Shiri, A.; et al. Advancements in machine learning and biomarker integration for prenatal Down syndrome screening. J. Turk. Soc. Obstet. Gynecol. 2025, 22, 75–82. [Google Scholar] [CrossRef]
  15. Yang, J.; Ding, X.; Zhu, W.; Shomron, N. Improving the calling of non-invasive prenatal testing on 13-/18-/21-trisomy by support vector machine discrimination. PLOS ONE 2018, 13, e0207840. [Google Scholar] [CrossRef]
  16. Qi, Y.; Hu, C.; Yang, J.; Gao, Y.; Yin, A. Self-learning neural network as a prediction model in non-invasive prenatal testing to detect fetal SNVs. J. Transl. Med. 2024, 22, 1–4. [Google Scholar] [CrossRef] [PubMed]
  17. News-Medical, “Non-invasive prenatal testing (NIPT): What it is, advancements and future prospects,” News-Medical, Aug. 05, 2024. https://www.news-medical.net/whitepaper/20240805/Non-invasive-prenatal-testing-(NIPT)-What-it-is-advancements-and-future-prospects.
  18. Liscovitch-Brauer, N.; Mesika, R.; Rabinowitz, T.; Volkov, H.; Grad, M.; Matar, R.T.; Basel-Salmon, L.; Tadmor, O.; Beker, A.; Shomron, N. Machine learning-enhanced noninvasive prenatal testing of monogenic disorders. Prenat. Diagn. 2024, 44, 1024–1032. [Google Scholar] [CrossRef] [PubMed]
  19. Shetty, H.P.; Patil, M.S.; Yadav, S.; Reddy. I, V. The Role of AI and Machine Learning in Revolutionizing Prenatal Screening and Genetic Analysis. Int. J. Innov. Sci. Res. Technol. (IJISRT) 2024, 2065–2067. [Google Scholar] [CrossRef]
  20. Miskeen, E.; Alfaifi, J.; Alhuian, D.M.; Alghamdi, M.; Alharthi, M.H.; Alshahrani, N.A.; Alosaimi, G.; Alshomrani, R.A.; Hajlaa, A.M.; Khair, N.M.; et al. Prospective Applications of Artificial Intelligence In Fetal Medicine: A Scoping Review of Recent Updates. Int. J. Gen. Med. 2025, ume 18, 237–245. [Google Scholar] [CrossRef]
  21. Yeganegi, M.; Danaei, M.; Azizi, S.; Jayervand, F.; Bahrami, R.; Dastgheib, S.A.; Rashnavadi, H.; Masoudi, A.; Shiri, A.; Aghili, K.; et al. Research advancements in the Use of artificial intelligence for prenatal diagnosis of neural tube defects. Front. Pediatr. 2025, 13, 1514447. [Google Scholar] [CrossRef]
  22. Shetty, H.P.; Patil, M.S.; Yadav, S.; Reddy. I, V. The Role of AI and Machine Learning in Revolutionizing Prenatal Screening and Genetic Analysis. Int. J. Innov. Sci. Res. Technol. (IJISRT) 2024, 2065–2067. [Google Scholar] [CrossRef]
  23. Xu, X.; Wang, L.; Cheng, X.; Ke, W.; Jie, S.; Lin, S.; Lai, M.; Zhang, L.; Li, Z. Machine learning-based evaluation of application value of the USM combined with NIPT in the diagnosis of fetal chromosomal abnormalities. Math. Biosci. Eng. 2022, 19, 4260–4276. [Google Scholar] [CrossRef]
  24. Boddupally, K.; Thuraka, E.R. Artificial intelligence for prenatal chromosome analysis. Clin. Chim. Acta 2023, 552, 117669. [Google Scholar] [CrossRef]
  25. Shaikh, M.A.; Al-Rawashdeh, H.S.; Sait, A.R.W. A Review of Artificial Intelligence-Based Down Syndrome Detection Techniques. Life 2025, 15, 390. [Google Scholar] [CrossRef]
  26. Ramakrishnan, R.; Rao, S.; He, J.-R. Perinatal health predictors using artificial intelligence: A review. Women's Heal. 2021, 17. [Google Scholar] [CrossRef]
  27. Tsui, W.A.; Ding, S.C.; Jiang, P.; Lo, Y.D. Artificial intelligence and machine learning in cell-free-DNA-based diagnostics. Genome Res. 2025, 35, 1–19. [Google Scholar] [CrossRef] [PubMed]
  28. Yeganegi, M.; Danaei, M.; Azizi, S.; Jayervand, F.; Bahrami, R.; Dastgheib, S.A.; Rashnavadi, H.; Masoudi, A.; Shiri, A.; Aghili, K.; et al. Research advancements in the Use of artificial intelligence for prenatal diagnosis of neural tube defects. Front. Pediatr. 2025, 13, 1514447. [Google Scholar] [CrossRef] [PubMed]
  29. Miskeen, E.; Alfaifi, J.; Alhuian, D.M.; Alghamdi, M.; Alharthi, M.H.; Alshahrani, N.A.; Alosaimi, G.; Alshomrani, R.A.; Hajlaa, A.M.; Khair, N.M.; et al. Prospective Applications of Artificial Intelligence In Fetal Medicine: A Scoping Review of Recent Updates. Int. J. Gen. Med. 2025, ume 18, 237–245. [Google Scholar] [CrossRef]
  30. New study highlights AI’s potential to help doctors detect congenital heart defects - Society for Maternal-Fetal Medicine.” https://www.smfm.org/news/new-study-highlights-ais-potential-to-help-doctors-detect-congenital-heart-defects.
  31. Patel, L. , “AI in prenatal care: Transforming fetal medicine,” Sonio, Feb. 05, 2025. https://sonio.ai/en/blog/ai-prenatal-care/.
  32. Miskeen, E.; Alfaifi, J.; Alhuian, D.M.; Alghamdi, M.; Alharthi, M.H.; Alshahrani, N.A.; Alosaimi, G.; Alshomrani, R.A.; Hajlaa, A.M.; Khair, N.M.; et al. Prospective Applications of Artificial Intelligence In Fetal Medicine: A Scoping Review of Recent Updates. Int. J. Gen. Med. 2025, ume 18, 237–245. [Google Scholar] [CrossRef]
  33. Shivarkar, A. , “AI in Ultrasound Imaging Market to Uplift >.57 Bn by 2034,” Mar. 10, 2025. https://www.towardshealthcare.com/insights/ai-in-ultrasound-imaging-market-size.
  34. Alius, “AI in Maternal Healthcare: Transforming Prenatal Care & Outcomes,” SentiSight.ai, Feb. 11, 2025. https://www.sentisight.ai/role-of-ai-in-maternal-healthcare/.
  35. Sonio at SMFM 2025: Advancing AI in Maternal-Fetal Medicine,” Sonio, Feb. 27, 2025. https://sonio.ai/news/smfm-2025-ai-ultrasound-innovation/. 2025.
  36. Ali, M.O.A.M.; Elhabeeb, S.M.A.; Elsheikh, N.E.A.; Mohammed, F.S.A.; Ali, S.H.M.; Abdelhalim, A.A.I.; Altom, D.S.; Omar, M.; Sr. , F.S.A.M. Advancing Obstetric Care Through Artificial Intelligence-Enhanced Clinical Decision Support Systems: A Systematic Review. Cureus 2025, 17. [Google Scholar] [CrossRef]
  37. Yeganegi, M.; Danaei, M.; Azizi, S.; Jayervand, F.; Bahrami, R.; Dastgheib, S.A.; Rashnavadi, H.; Masoudi, A.; Shiri, A.; Aghili, K.; et al. Research advancements in the Use of artificial intelligence for prenatal diagnosis of neural tube defects. Front. Pediatr. 2025, 13, 1514447. [Google Scholar] [CrossRef]
  38. Lee, J.; Lee, S.-M.; Ahn, J.M.; Lee, T.-R.; Kim, W.; Cho, E.-H.; Ki, C.-S. Development and performance evaluation of an artificial intelligence algorithm using cell-free DNA fragment distance for non-invasive prenatal testing (aiD-NIPT). Front. Genet. 2022, 13, 999587. [Google Scholar] [CrossRef]
  39. Nguyen, T.H.; Nguyen, H.V.; Pham, M.D.; Le, V.S. An Efficient Computational Method to Create Positive NIPT Samples with Autosomal Trisomy. 2024 12th International Conference on Bioinformatics and Computational Biology (ICBCB). LOCATION OF CONFERENCE, JapanDATE OF CONFERENCE; pp. 126–131.
  40. Gc_Admin, “NIPT,” GC Genome. https://gc-genome.com/our-science/focused-areas/nipt/.
  41. Xue, Z.; Zhou, A.; Zhu, X.; Li, L.; Zhu, H.; Jin, X.; Wang, J. NIPT-PG: empowering non-invasive prenatal testing to learn from population genomics through an incremental pan-genomic approach. Briefings Bioinform. 2024, 25. [Google Scholar] [CrossRef]
  42. Lee, J.; Lee, S.-M.; Ahn, J.M.; Lee, T.-R.; Kim, W.; Cho, E.-H.; Ki, C.-S. Development and performance evaluation of an artificial intelligence algorithm using cell-free DNA fragment distance for non-invasive prenatal testing (aiD-NIPT). Front. Genet. 2022, 13, 999587. [Google Scholar] [CrossRef]
  43. Lee, J.; Lee, S.-M.; Ahn, J.M.; Lee, T.-R.; Kim, W.; Cho, E.-H.; Ki, C.-S. Development and performance evaluation of an artificial intelligence algorithm using cell-free DNA fragment distance for non-invasive prenatal testing (aiD-NIPT). Front. Genet. 2022, 13, 999587. [Google Scholar] [CrossRef]
  44. Duboc, V.; Pratella, D.; Milanesio, M.; Boudjarane, J.; Descombes, S.; Paquis-Flucklinger, V.; Bottini, S. NiPTUNE: an automated pipeline for noninvasive prenatal testing in an accurate, integrative and flexible framework. Briefings Bioinform. 2021, 23. [Google Scholar] [CrossRef] [PubMed]
  45. Yang, S.; He, Y.; Lv, J.; Li, R.; Fu, X. The evolution of cell-free fetal DNA testing: expanded non-invasive prenatal testing and its effect on target populations. Front. Med. 2025, 12, 1522680. [Google Scholar] [CrossRef] [PubMed]
  46. Paluoja, P.; Teder, H.; Ardeshirdavani, A.; Bayindir, B.; Vermeesch, J.; Salumets, A.; Krjutškov, K.; Palta, P.; Li, J. Systematic evaluation of NIPT aneuploidy detection software tools with clinically validated NIPT samples. PLOS Comput. Biol. 2021, 17, e1009684. [Google Scholar] [CrossRef] [PubMed]
  47. Sotoudeh, H.; Shafaat, O.; Bernstock, J.D.; Brooks, M.D.; Elsayed, G.A.; Chen, J.A.; Szerip, P.; Chagoya, G.; Gessler, F.; Sotoudeh, E.; et al. Artificial Intelligence in the Management of Glioma: Era of Personalized Medicine. Front. Oncol. 2019, 9, 768. [Google Scholar] [CrossRef]
  48. Okon, E.S.; Michael, K.O.; Friday, R.E.; Efiong, A.J.; Obi, O.-A.M.; Timothy, A.J.; Edet, O.P.; Dominic, E.R.; Jimmy, U.A. Application of AI Algorithms for the Prediction of the Likelihood of Sickle Cell Crises. Sch. J. Eng. Technol. 2024, 12, 394–403. [Google Scholar] [CrossRef]
  49. Davidson, L.; Boland, M.R. Towards deep phenotyping pregnancy: a systematic review on artificial intelligence and machine learning methods to improve pregnancy outcomes. Briefings Bioinform. 2021, 22. [Google Scholar] [CrossRef]
  50. Ocak, H. A Medical Decision Support System Based on Support Vector Machines and the Genetic Algorithm for the Evaluation of Fetal Well-Being. J. Med Syst. 2013, 37, 1–9. [Google Scholar] [CrossRef]
  51. Schwammenthal, Y.; Rabinowitz, T.; Basel-Salmon, L.; Tomashov-Matar, R.; Shomron, N. Noninvasive fetal genotyping using deep neural networks. Briefings Bioinform. 2024, 26. [Google Scholar] [CrossRef]
  52. Danaei, M.; Rashnavadi, H.; Yeganegi, M.; Dastgheib, S.A.; Bahrami, R.; Azizi, S.; Jayervand, F.; Masoudi, A.; Shahbazi, A.; Shiri, A.; et al. Advancements in machine learning and biomarker integration for prenatal Down syndrome screening. J. Turk. Soc. Obstet. Gynecol. 2025, 22, 75–82. [Google Scholar] [CrossRef]
  53. Schmid, M.; Wang, E.; Bogard, P.E.; Bevilacqua, E.; Hacker, C.; Wang, S.; Doshi, J.; White, K.; Kaplan, J.; Sparks, A.; et al. Prenatal Screening for 22q11.2 Deletion Using a Targeted Microarray-Based Cell-Free DNA Test. Fetal Diagn. Ther. 2017, 44, 299–304. [Google Scholar] [CrossRef]
  54. News-Medical, “Genomics advances in non-invasive prenatal testing (NIPT) research,” News-Medical, Feb. 13, 2024. https://www.news-medical.net/whitepaper/20230727/Genomics-Advances-in-non-invasive-prenatal-testing-(NIPT)-Research.
  55. Natera, “Panorama – Non-Invasive Prenatal Testing (NIPT) | NaTERA,” Natera, Nov. 27, 2024. https://www.natera.com/womens-health/panorama-nipt-prenatal-screening/.
  56. Zhang, R.; Zhang, H.; Zhang, L.; Kong, X.; Wang, W.; Jia, Y.; Ren, M.; Liu, Y.; Zhu, L.; Chen, H.; et al. Clinical indications and Z-score-assisted NIPT testing: a new perspective in prenatal screening. Gynecol. Obstet. Clin. Med. 2025, 5, e000187. [Google Scholar] [CrossRef]
  57. Natera, “Panorama® SMART Study Confirms Market-Leading Performance in largest prospective NIPT Trial to date | Natera,” Natera, , 2023. https://www.natera.com/company/news/panorama-smart-study-confirms-market-leading-performance-in-largest-prospective-nipt-trial-to-date/. 10 May.
  58. Natera, “NaTera implements first wave of Panorama® AI improvements with immediate impact on patient experience and COGS | NATERA,” Natera, , 2023. https://www.natera.com/company/news/natera-implements-first-wave-of-panorama-ai-improvements-with-immediate-impact-on-patient-experience-and-cogs-2/. 09 May.
  59. triSure - Non-invasive prenatal testing (NIPT),” Gene Solutions. https://genesolutions.com/our-test/trisure-nipt.
  60. Noninvasive Prenatal Testing (NIPT) | NGS-based technology.” https://www.illumina.com/clinical/reproductive-genetic-health/nipt.html.
  61. Gadsbøll, K.; Petersen, O.B.; Gatinois, V.; Strange, H.; Jacobsson, B.; Wapner, R.; Vermeesch, J.R.; Vogel, I. Current use of noninvasive prenatal testing in Europe, Australia and the USA: A graphical presentation. Acta Obstet. Gynecol. Scand. 2020, 99, 722–730. [Google Scholar] [CrossRef] [PubMed]
  62. Connor, C.; Sato, T.; Bianchi, D.W.; Fenton, K.; Somani, E.; Turriff, A.; Berkman, B.; Hendriks, S. Comparing the Introduction and Implementation of Noninvasive Prenatal Testing (NIPT) in Japan, the Netherlands, and the United States: An Integrative Review. Prenat. Diagn. 2025. [Google Scholar] [CrossRef] [PubMed]
  63. Jayashankar, S.S.; Nasaruddin, M.L.; Hassan, M.F.; Dasrilsyah, R.A.; Shafiee, M.N.; Ismail, N.A.S.; Alias, E. Non-Invasive Prenatal Testing (NIPT): Reliability, Challenges, and Future Directions. Diagnostics 2023, 13, 2570. [Google Scholar] [CrossRef]
  64. Takahashi, M.; Linh, L.K.; Sayed, A.M.; Imoto, A.; Sato, M.; Dila, K.A.S.; Huy, N.T.; Moji, K. Non-Invasive Prenatal Testing (NIPT) Implementation in Japan: A Comparison with the United Kingdom, Germany, Italy, Sweden, and Taiwan. Int. J. Environ. Res. Public Heal. 2022, 19, 16404. [Google Scholar] [CrossRef]
  65. Sesa, G.; Czabanowska, K.; Kok, K.; Vrijenhoek, T. Mainstreaming genetics into Italian prenatal care: exploring the future implementation of non-invasive prenatal testing (NIPT) in the Italian National Healthcare System using stakeholder interviews. BMJ Public Heal. 2024, 2, e001074. [Google Scholar] [CrossRef]
  66. Sesa, G.; Czabanowska, K.; Kok, K.; Vrijenhoek, T. Mainstreaming genetics into Italian prenatal care: exploring the future implementation of non-invasive prenatal testing (NIPT) in the Italian National Healthcare System using stakeholder interviews. BMJ Public Heal. 2024, 2, e001074. [Google Scholar] [CrossRef]
  67. AI adoption rate by country.” https://www.byteplus.com/en/topic/516576.
  68. Markets, R.A. , “Non-Invasive Prenatal Testing (NIPT) Market Report 2025: Rising demand for early and accurate Fetal screening spurs growth in Non-Invasive Prenatal Testing Market,” GlobeNewswire News Room, Jan. 07, 2025. [Online]. Available: https://www.globenewswire.com/news-release/2025/01/07/3005691/28124/en/Non-Invasive-Prenatal-Testing-NIPT-Market-Report-2025-Rising-Demand-for-Early-and-Accurate-Fetal-Screening-Spurs-Growth-in-Non-Invasive-Prenatal-Testing-Market.html.
  69. Markets, R.A. , “Asia-Pacific Non-Invasive Prenatal Testing (NIPT) Market Analysis and Forecast Report 2024, with Focus on BGI Genomics and Annoroad Gene Technology,” GlobeNewswire News Room, Dec. 06, 2024. [Online]. Available: https://www.globenewswire.com/news-release/2024/12/06/2992902/28124/en/Asia-Pacific-Non-Invasive-Prenatal-Testing-NIPT-Market-Analysis-and-Forecast-Report-2024-with-Focus-on-BGI-Genomics-and-Annoroad-Gene-Technology.html.
  70. Lundin, A.; Lundin, A. , “How genetic background of pregnant women can influence NIPT results,” Clinical Lab Products, Sep. 26, 2024. https://clpmag.com/disease-states/womens-health/prenatal-testing/how-genetic-background-of-pregnant-women-can-influence-nipt-results/.
  71. Non-invasive prenatal genetics and genomics in England, France and Germany - Exploring practical ethical issues ‘on the ground,’” Haiku. https://www.ethox.ox.ac.uk/Our-research/research-projects/nipt.
  72. Zaami, S.; Orrico, A.; Signore, F.; Cavaliere, A.F.; Mazzi, M.; Marinelli, E. Ethical, Legal and Social Issues (ELSI) Associated with Non-Invasive Prenatal Testing: Reflections on the Evolution of Prenatal Diagnosis and Procreative Choices. Genes 2021, 12, 204. [Google Scholar] [CrossRef]
  73. Zhang, H.; Wang, J.; Qin, Y.; Zhang, C.; Wang, B.; Wang, Y. Anticipatory Governance of Noninvasive Prenatal Testing for “Non-Medical” Traits: Lessons from Regulation of Medically Assisted Reproduction. Am. J. Bioeth. 2023, 23, 45–47. [Google Scholar] [CrossRef]
  74. Deans, Z.; Clarke, A.J.; Newson, A.J. For Your Interest? The Ethical Acceptability of Using Non-Invasive Prenatal Testing to Test ‘Purely for Information’. Bioethics 2014, 29, 19–25. [Google Scholar] [CrossRef]
  75. McDonough, B.M. , “The ethics of prenatal genetic testing,” Harvard Medicine Magazine, Mar. 03, 2025. [Online]. Available: https://magazine.hms.harvard.edu/articles/ethics-prenatal-genetic-testing.
  76. Non-Invasive Prenatal Testing – Ethical & Medico-legal issues - MDA National.” https://www.mdanational.com.au/advice-and-support/library/articles-and-case-studies/2017/11/nipt-ethical-issues. 2017.
  77. Moufarrej, M.N.; Bianchi, D.W.; Shaw, G.M.; Stevenson, D.K.; Quake, S.R. Noninvasive Prenatal Testing Using Circulating DNA and RNA: Advances, Challenges, and Possibilities. Annu. Rev. Biomed. Data Sci. 2023, 6, 397–418. [Google Scholar] [CrossRef] [PubMed]
  78. Shaikh, M.A.; Al-Rawashdeh, H.S.; Sait, A.R.W. A Review of Artificial Intelligence-Based Down Syndrome Detection Techniques. Life 2025, 15, 390. [Google Scholar] [CrossRef] [PubMed]
  79. Honeywood, K. , “International Collaboration Aims to Revolutionise Diagnosis of the Rare Diseases with AI and Federated Learning | Myrovlytis Trust.” https://myrovlytistrust.org/announcing-a-new-collaborative-research-project-funded-by-the-myrovlytis-trust/.
  80. Nuvo.” https://www.nuvocares.com/resources/sheba-medical-center-collaborates-with-nuvo-intl-group-to-develop-advanced-ai-powered-remote-prenatal-care-solutions.
  81. AI-enabled ultrasound.” https://www.gatesfoundation.org/ideas/science-innovation-technology/future-womens-health-technology/ai-ultrasounds.
  82. AI in maternal health – how research and policy intertwine - APRU,” APRU, Aug. 29, 2024. https://apru.org/news/ai-in-maternal-health/.
  83. Liu, X.; Zhang, Y.; Zhu, H.; Jia, B.; Wang, J.; He, Y.; Zhang, H. Applications of artificial intelligence-powered prenatal diagnosis for congenital heart disease. Front. Cardiovasc. Med. 2024, 11, 1345761. [Google Scholar] [CrossRef] [PubMed]
  84. Responsible AI in Global Health: Solutions from the Global South | Center for Global Digital Health Innovation,” Center for Global Digital Health Innovation. https://publichealth.jhu.edu/center-for-global-digital-health-innovation/responsible-ai-in-global-health-solutions-from-the-global-south.
  85. AI in Genetic Testing: How 3billion is Revolutionizing Rare Disease Diagnosis.” https://3billion.io/blog/ai-in-genetic-testing-how-3billion-is-revolutionizing-rare-disease-diagnosis.
  86. Whisperer, L. , “Life Whisperer launches its third global IVF clinic collaboration at ESHRE to apply Artificial Intelligence to Oocytes,” Life Whisperer, Mar. 10, 2023. https://www.lifewhisperer.com/life-whisperer-launches-its-third-global-ivf-clinic-collaboration-at-eshre-to-apply-artificial-intelligence-to-oocytes/.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
Copyright: This open access article is published under a Creative Commons CC BY 4.0 license, which permit the free download, distribution, and reuse, provided that the author and preprint are cited in any reuse.
Prerpints.org logo

Preprints.org is a free preprint server supported by MDPI in Basel, Switzerland.

facebook logotwitter logolinkedin logo
bsky
MDPI Initiatives
Important Links
Subscribe

Disclaimer

Terms of Use

Privacy Policy

Privacy Settings

© 2026 MDPI (Basel, Switzerland) unless otherwise stated