Sage Journals: Discover world-class research

Abstract

Background:

Artificial intelligence (AI) has emerged as a transformative tool for advancing gestational diabetes mellitus (GDM) care, offering dynamic, data-driven methods for early detection, management, and personalized intervention.

Objective:

This systematic review aims to comprehensively explore and synthesize the use of AI models in GDM care, including screening, diagnosis, management, and prediction of maternal and neonatal outcomes. Specifically, we examine (1) study designs and population characteristics; (2) the use of AI across different aspects of GDM care; (3) types of input data used for AI modeling; and (4) AI model types, validation strategies, and performance metrics.

Methods:

A systematic search was conducted across six electronic databases, identifying 126 eligible studies published up to February 2025. Data extraction and quality appraisal were independently conducted by six reviewers, using a modified version of the Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool for risk of bias assessment.

Results:

Among 126 studies, 75% employed retrospective designs, with sample sizes ranging from 17 to over 100 000 participants. Most AI applications (85%) focused on early GDM prediction, while fewer addressed management, outcomes, or monitoring. Classical machine learning dominated (84%), with logistic regression and random forest frequently used. Internal validation was common (68%), but external validation was rare (6%). Our risk of bias appraisal indicated an overall moderate-to-good methodological quality, with notable deficiencies in analysis reporting.

Conclusions:

AI demonstrates strong potential to improve GDM prediction, screening, and management. Nonetheless, broader validation, enhanced model interpretability, and prospective studies in diverse populations are needed to translate these technologies into clinical practice.

Keywords

artificial intelligence diabetes gestational diabetes pregnancy women’s health

Introduction

Gestational diabetes mellitus (GDM) is one of the most common metabolic disorders during pregnancy, affecting approximately 14% to 18% of pregnancies globally, with prevalence rates exhibiting significant regional and population-specific variability due to differences in genetic predisposition, lifestyle factors, and diagnostic criteria.^1-4 During pregnancy, women with GDM are at increased risk of antenatal complications such as preeclampsia⁵ and higher rates of cesarean delivery, while their offspring are at risk of macrosomia,⁶ neonatal hypoglycemia,⁷ preterm delivery, and admission to neonatal intensive care units. In the long term, mothers with a history of GDM are at elevated risk of developing type 2 diabetes mellitus (T2DM)⁸ and cardiovascular diseases,⁹ and their offspring face an increased risk of childhood obesity.¹⁰ Given its rising prevalence, particularly in populations with high obesity rates and sedentary lifestyles, there is an urgent need for improved screening, early detection, and personalized management strategies for GDM.

Despite the availability of standardized screening guidelines,^11,12 traditional approaches for GDM diagnosis and prediction remain suboptimal, time-consuming, and often reactive rather than proactive.¹³ The oral glucose tolerance test (OGTT), which is the gold standard for GDM diagnosis, is invasive, inconvenient, and prone to variability due to factors such as dietary habits, stress, and physiological changes during pregnancy. In addition, risk-based screening methods relying on maternal age, body mass index (BMI), family history, and ethnicity often lack precision, leading to both underdiagnosis and overtreatment. Moreover, disparities in health care access and inconsistent adherence to screening protocols further limit the effectiveness of traditional methods, emphasizing the need for more dynamic, data-driven approaches to improve GDM prediction, early intervention, and personalized management.^12,13

Artificial intelligence (AI) has emerged as a transformative tool in modern medicine, offering new possibilities for predictive analytics, personalized interventions, and automated decision-making.¹⁴ In the field of GDM, AI-driven approaches, particularly machine learning (ML) and deep learning (DL) algorithms, have demonstrated promising capabilities in the early risk prediction, continuous glucose monitoring, insulin-dosing recommendations, and outcome prediction. These techniques leverage vast amounts of clinical, biochemical, and lifestyle data to enhance the precision and efficiency of GDM management. The AI applications in GDM range from traditional supervised learning models using structured electronic health records (EHRs) to advanced DL frameworks incorporating real-time sensor data, medical imaging, and genomics. Several recent studies^15-17 have explored the role of AI in GDM, with models achieving high accuracy in predicting GDM onset based on maternal risk factors, optimizing insulin therapy, and improving continuous glucose monitoring via wearable devices and mobile health (mHealth) applications. In addition, AI-driven decision support systems (DSSs) have been developed to assist clinicians in tailoring treatment plans and providing real-time feedback for dietary and physical activity interventions.

Although a few prior reviews have explored AI applications in GDM care, significant limitations remain.^14-17 Some reviews, such as the Sub-Saharan Africa-focused review by Gadhia and Loyal,¹⁶ were geographically restricted and lacked broader applicability beyond low-resource settings. Others, like the review on early GDM prediction using ML,¹⁷ were limited to models developed before 24 to 28 weeks of pregnancy, overlooking AI applications in diagnosis, management, and outcome prediction later in pregnancy. In addition, previous reviews often focused on specific technologies, such as mobile health (mHealth) platforms,¹⁵ or were limited to screening and diagnostic models,¹⁴ without covering the full spectrum of AI approaches or addressing the entire continuum of GDM care. Notably, none of the prior reviews were systematic reviews; they were primarily scoping or narrative reviews, limiting the depth and rigor of their synthesis. In contrast, this systematic review offers a comprehensive synthesis across the entire clinical trajectory of GDM, including screening, diagnosis, management, glycemic control, and prediction of maternal and neonatal outcomes, without restrictions on geographic location, population characteristics, or AI model type. Furthermore, by evaluating methodological quality using a modified Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool, our review critically appraises study rigor, addresses validation gaps, and highlights future research priorities, offering a more holistic and clinically actionable overview than prior work.

This systematic review aims to provide a comprehensive analysis of the current clinical applications of AI in gestational diabetes care, highlighting its potential benefits, limitations, and future directions. The primary research question guiding this review is: How has artificial intelligence been applied clinically to improve the screening, diagnosis, management, and outcome prediction of gestational diabetes mellitus and its associated maternal and neonatal outcomes during pregnancy and the neonatal period? By addressing this research question, we aim to provide insights into the current state of AI applications in GDM care. In particular, we synthesize findings across four key aspects: (1) study designs and population characteristics; (2) use of AI in different areas of gestational diabetes care; (3) types of input data used for AI modeling; and (4) AI model types, validation strategies, and performance metrics. This review also assesses methodological strengths and gaps and proposes priorities for future research and clinical implementation.

Methods

To achieve the objectives of this study, we conducted a systematic review following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.¹⁸ The corresponding PRISMA checklist is provided in Supplementary Material 1. The following sections present a detailed description of the methods employed in this review.

Search Strategy

A comprehensive search was conducted on October 4, 2024, across the following electronic databases: MEDLINE (via Ovid), EMBASE (via Ovid), ACM Digital Library, Scopus, IEEE Xplore, and Google Scholar. To ensure ongoing capture of newly published studies, an automated alert was set to run the search query biweekly over a four-month period, concluding on February 4, 2025. Given the extensive volume of results retrieved from Google Scholar, only the first 100 results (equivalent to the first 10 pages) were screened for relevance, in line with common systematic review practices. To identify additional eligible studies, we also performed backward reference list checking (reviewing references cited by included studies) and forward reference list checking (reviewing studies that cited the included studies). The search strategy incorporated two primary categories of keywords: (1) terms related to AI and (2) terms related to Gestational Diabetes. Detailed search queries for each database are provided in Supplementary Material 2.

Study Eligibility Criteria

The eligibility criteria were designed to ensure the inclusion of studies that specifically evaluated the application of AI for clinical purposes related to GDM during pregnancy, birth, or the neonatal period.

Inclusion criteria

Studies involving pregnant women who were screened or tested for GDM.

Studies that employed AI algorithms, including ML, DL, or other AI models, for the detection or prediction of GDM or its associated fetal, maternal, and neonatal outcomes.

Studies addressing outcomes occurring during pregnancy, at birth, or during the neonatal period.

Original research articles, including retrospective cohort studies, prospective cohort studies, randomized controlled trials, case-control studies, and cross-sectional studies.

Studies published in English.

No restrictions on the publication date.

Exclusion criteria

Studies focusing exclusively on outcomes beyond the neonatal period (eg, long-term maternal or child outcomes).

Review articles, commentaries, editorials, conference abstracts, and preprints.

Studies not involving the use of AI algorithms for GDM detection, prediction, or management.

Studies not involving human subjects.

Study Selection

The study selection process was conducted in three phases. First, duplicate records were removed using EndNote X9 reference management software. In the second phase, the titles and abstracts of the remaining articles were screened for relevance according to the predefined eligibility criteria. In the final phase, the full texts of the shortlisted studies were retrieved and thoroughly assessed for inclusion. The selection process was independently performed by four reviewers. Any disagreements between reviewers were resolved through discussion and consensus; when necessary, a senior reviewer was consulted to reach a final decision.

Data Extraction

Four reviewers independently extracted data using a standardized Microsoft Excel form. The extracted information was structured to align with the main focus areas of this review. Any disagreements between reviewers were resolved through discussion. The data extraction form was piloted on a sample of ten studies to ensure clarity and consistency and is available in Supplementary Material 3.

Risk of Bias and Applicability Appraisal

To assess the quality of the studies included in this review, we adapted the QUADAS-2 tool¹⁹ to better align with the specific objectives of our review. This adaptation involved substituting certain original QUADAS-2 criteria, which were not applicable to our context, with more relevant criteria drawn from the Prediction Model Risk of Bias Assessment Tool (PROBAST).²⁰ The modified tool encompassed four key domains tailored to this review: “Participants,” “Index Test” (focused on AI algorithms), “Reference Standard” (representing the ground truth), and “Analysis.”

For each domain, we developed four targeted signaling questions specifically aligned with the aims of this review. In addition to assessing the risk of bias, we also evaluated the applicability of results derived from the first three domains (“Participants,” “Index Test,” and “Reference Standard”). To optimize and calibrate the adapted tool, it was piloted on a subset of ten studies. Subsequently, all included studies were independently evaluated by two reviewers using the finalized modified QUADAS-2 tool (Supplementary Material 4). Any discrepancies between reviewers were discussed and resolved through consensus.

Data Synthesis

We synthesized the extracted data from the included studies using a narrative approach, providing a comprehensive summary through text, tables, and figures. The synthesis was organized across five key domains: (1) an overview of included studies, summarizing the number of studies, publication years, countries of origin, and publication types; (2) study designs and population details, outlining the types of study designs employed, sample sizes, and participant characteristics; (3) applications of AI in gestational diabetes, detailing the specific clinical goals targeted by AI models, such as screening, diagnosis, risk prediction, and management of GDM; (4) input data used for AI modeling, describing the types and sources of data utilized, including clinical, demographic, laboratory, and imaging data; and (5) AI model types, validation strategies, and performance metrics, summarizing the AI techniques applied, methods used for model validation, and key performance metrics reported.

Results

Search Results

As illustrated in Figure 1, a total of 1347 records were identified through database searches. After removing 431 duplicate records, 916 records remained for title and abstract screening. Of these, 642 records were excluded for not meeting the inclusion criteria. The full texts of 274 articles were sought for retrieval; however, eight articles could not be retrieved. A total of 266 full-text articles were assessed for eligibility, and 145 were excluded for various reasons, including irrelevant publication type (eg, posters, preprints, and reviews; n = 91), gestational diabetes not being the main focus (n = 26), the study not involving AI (n = 24), not involving human subjects (n = 2), or not being in English (n = 2). In addition, five relevant articles were identified through backward and forward reference list checking. In total, 126 studies were included in the final review.^21-145

Figure 1.

Flowchart of the study selection process.

Overview of Included Studies

As shown in Figure 2 and detailed in Supplemental Table 1, the volume of research on AI applications in GDM has increased substantially in recent years, with over 60% (76/126) of the included studies published between 2021 and 2024. China and India are the leading countries in terms of publication volume, collectively contributing to over 50% (66/126) of all included studies. Geographic representation was broad, with contributions from Asia, Europe, North America, and Africa.

Figure 2.

Overview of included studies. (a) Distribution of included studies by year of publication. (b) Country of origin of the included studies. (c) Type of publication. (d) Presence of ethical statements reported in the included studies. (e) Availability of funding sources among the included studies.

Most studies were journal articles (84%, 106/126), and the remainder were conference papers. Reporting of ethical approval and funding was inconsistent. While 52% (65/126) of the studies reported ethical approval and 53% (67/126) disclosed funding, a considerable proportion did not mention neither, limiting transparency.

Study Designs and Population Details

Most studies were retrospective in design (75%, 95/126), while only a minority were prospective (21%, 27/126). The majority of studies were retrospective (75%, 95/126), with prospective designs accounting for only 21% (27/126) of the studies. Nearly half of the studies were conducted at a single site (44%, 55/126), 30% (38/126) were multicenter, while 26% (33/126) did not report the number of study sites. Sample sizes varied widely, with a mean of 5468 participants and a range from 17 to over 100 000, indicating a broad spectrum of data set sizes. Where reported, the mean participant age was 32 years, consistent with the typical reproductive age range (Supplemental Table 2).

Applications of Artificial Intelligence in Gestational Diabetes

Applications of AI in gestational diabetes across the included studies are presented in Supplemental Table 3. As shown in Figure 3, the most common application, observed in 85% (107/126) of studies, was the prediction, screening, and diagnosis of GDM. These studies primarily focused on the early identification of at-risk individuals using clinical, demographic, or biochemical features. A smaller subset, 8% (10/126) of studies, addressed treatment and intervention success, evaluating how AI could predict or optimize responses to therapeutic strategies. Applications targeting pregnancy and maternal health outcomes were reported in 6% (8/126) of studies and focused on predicting pregnancy-related complications in women with GDM, such as preeclampsia, preterm birth, cesarean delivery, and excessive maternal weight gain. Similarly, applications related to glycemic control and monitoring were described in 6% (7/126) of studies, aiming to predict blood glucose fluctuations and support ongoing management of glucose levels throughout pregnancy. Only 4% (5/126) of studies explored AI applications for neonatal health outcomes in pregnancies complicated by GDM, such as predicting birth weight abnormalities or neonatal intensive care unit admission.

Figure 3.

Main applications of artificial intelligence in gestational diabetes care.

At the specific application level, early screening and risk prediction dominated, accounting for 70% (88/126) of studies, underscoring the emphasis on proactive management of GDM. Biomarker-based prediction was reported in 15% (19/126) of studies and referred to the use of specific biological markers, such as blood-based metabolites or genetic data, to improve prediction accuracy. Applications addressing maternal and neonatal health outcomes were identified in 8% (10/126) of studies, while delivery and birth outcome prediction appeared in 6% (8/126) of studies, both focusing on broader pregnancy-related complications associated with GDM. Other niche applications included glucose monitoring and prediction, insulin and pharmacotherapy management, and diet and lifestyle interventions, each comprising 1% (1/126) of studies.

Regarding outcome types, 87% (109/126) of studies focused exclusively on maternal outcomes, 13% (16/126) addressed combined maternal and fetal outcomes, and only 2% (2/126) exclusively targeted fetal outcomes.

Input Data Used for AI Modeling

As depicted in Figure 4 and comprehensively outlined in Supplemental Table 4, over half of the studies used private data sets (59%, 74/126), while 32% (39/126) used public data sets. Among the public data sets, the Pima Indian Diabetes data set was the most frequently used (14%, 18/126), followed by GEO (4%, 5/126) and the Kaggle GDM data set (4%, 5/126).

Figure 4.

Input data used for AI modeling. (a) Distribution of data sources used across included studies. (b) Distribution of specific data sets utilized for developing AI models for GDM. (c) Types of input features employed in AI models for GDM prediction or diagnosis.

The most commonly used input features were maternal demographics (87%, 110/126), anthropometric measures (87%, 109/126), obstetric history (82%, 103/126), and biochemical or lab data (79%, 99/126). Less frequently used features included vital signs (45%, 57/126), lifestyle factors (29%, 37/126), and genetic data (9%, 11/126). Only a small number of studies incorporated imaging data (13%, 16/126) or continuous monitoring devices (12%, 15/126).

Artificial Intelligence Model Types, Validation Strategies, and Performance Metrics

As illustrated in Figures 5 and 6 and comprehensively detailed in Supplemental Table 5, the majority of studies used classical ML models (84%, 106/126), with logistic regression (38%, 48/126), random forest (36%, 45/126), and gradient boosting (29%, 37/126) being the most commonly applied algorithms. Deep learning approaches were used in 28% (36/126) of studies, primarily in more recent publications.

Figure 5.

Distribution of AI models, evaluation metrics, and training data percentages among the included studies. (a) Frequency of specific AI models used. (b) Frequency of reported evaluation metrics. (c) Distribution of the proportion of training data used in model development.

Figure 6.

Characteristics of the AI modeling approaches across the included studies. (a) Proportion of machine learning vs deep learning models. (b) Type of validation approach adopted (internal vs external validation). (c) Techniques used for model validation (eg, k-fold cross-validation, hold-out). (d) Implementation of interpretability or explainability mechanisms.

Internal validation was conducted in 68% (86/126) of studies, most commonly through k-fold cross-validation (52%, 66/126). External validation was rare, reported in only 6% (7/126) of studies. Notably, 29% (36/126) of studies did not report the validation method used.

In terms of performance evaluation, the most frequently reported metrics were accuracy (77%, 97/126), sensitivity (63%, 79/126), and area under the receiver operating characteristic curve (AUC-ROC) (55%, 69/126). However, explainability mechanisms were incorporated in only 33% (42/126) of models.

Regarding the proportion of data used for model training, 6% (8/126) of studies used less than 60%, 22% (28/126) used moderate proportions (60%-74%), while 44% (55/126) did not report this information.

Results of Risk of Bias Appraisal

The risk of bias across the included studies was evaluated using a modified version of the QUADAS-2 tool, with details of the tool and the associated signaling questions provided in Supplementary Material 4. This tool assessed risk of bias across four domains: Participant selection, Index test, Reference standard, and Analysis. A summary of the risk of bias results across studies is presented in Figure 7, and the results of the applicability concerns assessment are shown in Figure 8. A detailed breakdown of the “risk of bias” and “applicability concerns” for each domain in every study is available in Supplementary Material 5. Overall, the majority of studies demonstrated good methodological quality, although certain areas of concern were noted.

Figure 7.

Results of assessment of risk of bias in the studies.

Figure 8.

Results of assessment of applicability concerns in the studies.

Participant domain

In the “Participant” domain, 71% (89/126) of studies were assessed as having a low risk of bias (Figure 7). Specifically, 100% (126/126) of studies enrolled a consecutive or random sample of participants (Signaling Question 1.1) and appropriately avoided inappropriate exclusions (1.2). However, only 86% (108/126) of studies included a sufficient sample size (≥100 participants; 1.3), and only 37% (46/126) had a balanced distribution across key subgroups (1.4). Consequently, 27% (34/126) of studies were categorized as high risk, primarily due to concerns regarding sample size and subgroup representation.

Index test domain

Nearly all studies described their AI models in detail (Figure 7), with 98% (124/126) providing a comprehensive model description (2.1). Similarly, 99% (125/126) of studies clearly identified all features used in model development (2.2), and 99% (125/126) ensured standardized feature assessment across participants (2.3). In addition, 99% (125/126) of studies collected features without knowledge of the outcome, minimizing the potential for bias (2.4).

Reference standard domain

In the “Reference Standard” domain, 97% (122/126) of studies were assessed as low risk (Figure 7). The majority used validated clinical criteria for diagnosing gestational diabetes (3.1) and consistently applied outcome assessments across participants (3.2). Blinding between predictor assessment and outcome determination (3.3) was reported in 99% (125/126) of studies. In addition, nearly all studies (99%, 125/126) minimized the risk of bias related to the timing of predictor and outcome assessments (3.4).

Analysis domain

The “Analysis” domain showed slightly more variability (Figure 7). A total of 71% (89/126) of studies were categorized as low risk, whereas 21% (26/126) were assessed as high risk, and 9% (11/126) were unclear. While 77% (97/126) of studies included all participants in the final analysis (4.1), only 50% (63/126) appropriately reported the breakdown of training, validation, and testing sets (4.3), and 67% (84/126) reported model performance appropriately (4.4). Data preprocessing was appropriately described in 98% (124/126) of studies (4.2).

Applicability concerns

Regarding applicability, most studies had low concerns and demonstrated strong methodological quality, as shown in Figure 8. Specifically, 71% (89/126) of studies were rated as low concern in the “Participant” domain, 98% (123/126) in the “Index Test” domain, and 97% (122/126) in the “Reference Standard” domain. Only a small proportion of studies were rated as unclear or high concern, primarily due to insufficient reporting of participant characteristics or the use of non-standard reference criteria.

Discussion

Principal Findings

This systematic review provides a comprehensive synthesis of the current clinical applications of AI in gestational diabetes care, drawing on 126 original studies. The findings reveal a rapidly evolving landscape, characterized by a surge in research output in recent years: more than 60% of the included studies were published between 2021 and 2024, underscoring the growing recognition of AI’s potential to transform GDM care in response to increasing global disease burden and the pressing need for precision medicine approaches during pregnancy.

The most common application of AI was in the early prediction, screening, and diagnosis of GDM, reported in 85% of the included studies. These models predominantly utilized structured data, such as demographic characteristics, clinical histories, and laboratory results, to enhance the precision and timeliness of GDM detection. This strong focus on early identification reflects a paradigm shift in obstetric care, moving away from reactive diagnosis toward proactive risk stratification and prevention. Early prediction is clinically critical, as it allows for the implementation of lifestyle modifications, glucose monitoring, and pharmacologic interventions that can significantly improve maternal-fetal outcomes. However, the relative underrepresentation of studies targeting later stages of GDM care—such as treatment personalization, glycemic control, and neonatal outcome prediction—highlights a gap in current AI research, particularly considering the complex, dynamic nature of glycemic regulation throughout pregnancy.

In terms of methodological approaches, classical ML algorithms dominated the field, with logistic regression, random forest, and gradient boosting being the most frequently applied. The popularity of these models is understandable given their relatively low computational cost, interpretability, and robust performance on tabular health care data. However, their dominance may also indicate a methodological conservatism, potentially limiting the exploration of more advanced architectures that could exploit richer, multimodal data sources. Deep learning techniques, although applied in only 28% of studies, are gradually gaining traction, particularly in projects leveraging imaging, continuous glucose monitoring, or omics data. This gradual adoption suggests an encouraging trend toward more sophisticated modeling strategies, aligned with the increasing availability of high-dimensional and longitudinal data sets in maternal-fetal medicine.

Several important methodological challenges persist. Internal validation, predominantly via k-fold cross-validation, was reported in the majority of studies (68%), reflecting efforts to reduce overfitting. However, external validation—crucial for assessing model generalizability across populations and settings—was exceedingly rare (6%). The absence of external validation remains a critical barrier to clinical translation, as AI models trained and validated on homogeneous or geographically restricted data sets may fail to perform adequately in more diverse real-world populations. Furthermore, nearly one-third of studies did not adequately describe their validation strategies, undermining reproducibility and raising concerns about potential reporting biases that could inflate model performance estimates.

The overreliance on accuracy as a primary performance metric, particularly in imbalanced data sets where GDM prevalence may be low, can lead to misleading assessments of clinical utility. Incorporating complementary metrics such as sensitivity, specificity, and F1-score would provide a more nuanced and clinically relevant understanding of model strengths and weaknesses. Moreover, the explainability of AI models remains underdeveloped; only one-third of studies incorporated any form of interpretability mechanisms. In a high-stakes context like pregnancy care, where clinical decisions directly impact maternal and neonatal health, black-box models are unlikely to gain clinician trust or regulatory approval.

Our risk of bias appraisal indicated overall moderate-to-good methodological quality, with notable strengths observed in the index test and reference standard used. Nevertheless, deficiencies in analysis reporting, especially regarding the documentation of training and testing set partitions, were notable. Poor reporting not only compromises study reproducibility but also hampers the ability to critically appraise the robustness of model development pipelines.

Research and Practical Implications

The findings of this review have several important implications for the future development and clinical translation of AI applications in gestational diabetes care. First, while the focus on early prediction and diagnosis is justified by the potential for preventive interventions, there is a clear need to broaden the scope of AI research to encompass the full continuum of gestational diabetes care. Future studies should explore the use of AI for treatment optimization, dynamic glucose monitoring, individualized insulin dosing, and prediction of maternal and neonatal outcomes beyond the diagnosis of GDM itself. Such expansion is critical for realizing AI’s potential not only as a diagnostic aid but also as a tool for comprehensive pregnancy management and postnatal care.

Second, the predominance of studies relying on retrospective data and single-center cohorts highlights the urgent need for prospective, multicenter studies with diverse populations. The AI models trained and validated on narrow, homogeneous data sets risk poor generalizability and may exacerbate existing disparities in GDM care if deployed without external validation. Future research should prioritize the development of models using multiethnic, geographically diverse cohorts and should report performance across clinically meaningful subgroups, such as by ethnicity, BMI category, or age. This will be essential for ensuring equitable AI adoption in obstetric practice.

Third, enhancing model transparency and interpretability must become a central research priority. The low proportion of studies incorporating explainability techniques is concerning, given that clinical decision-making in pregnancy demands not only accuracy but also clinician understanding and justification of model outputs. Researchers should routinely integrate explainable AI (XAI) methods, such as SHapley Additive exPlanations (SHAP) values, Local Interpretable Model-agnostic Explanations (LIME), attention maps, or counterfactual explanations, and report how these mechanisms can inform clinical actionability. The development of interpretable-by-design models, rather than post hoc explanations, may be particularly valuable for sensitive clinical domains such as maternal-fetal medicine.

Fourth, to facilitate reproducibility, critical appraisal, and eventual clinical translation, future studies must adhere to standardized reporting guidelines such as transparent reporting of a multivariable prediction model for individual prognosis Or diagnosis–artificial intelligence (TRIPOD-AI) and Consolidated Standards of Reporting Trials–Artificial Intelligence (CONSORT-AI).¹⁴⁶ Transparent documentation of data sources, preprocessing steps, model development pipelines, training-validation-testing splits, and full performance metrics should be considered mandatory.

Finally, collaboration between AI developers, obstetricians, endocrinologists, data scientists, and patients will be vital. Co-designing AI tools with end-users in mind can improve clinical relevance, usability, and trust. Moreover, pilot implementation studies and randomized trials evaluating the clinical impact of AI-guided interventions in GDM care are urgently needed. Without such evidence, the clinical deployment of AI will remain aspirational rather than transformational.

Limitations

This review has several limitations that should be considered when interpreting the findings. First, it included only studies published in English, which may have resulted in the exclusion of relevant research conducted in other languages. Second, although this review systematically synthesized a large number of studies, it did not conduct a meta-analytical synthesis of model performance. The decision against meta-analysis was primarily due to the substantial heterogeneity across studies in terms of study designs, AI model types, data sources, feature sets, validation strategies, and performance reporting. Studies varied widely in their input features (eg, clinical vs imaging data), prediction targets (eg, GDM onset vs maternal or neonatal complications), and evaluation metrics (eg, accuracy, AUC-ROC, sensitivity, specificity), often without standardized thresholds or consistent methodologies. In addition, many studies lacked sufficient statistical reporting (eg, confidence intervals, calibration plots), making it inappropriate and potentially misleading to pool performance estimates quantitatively. Therefore, while a meta-analysis would have been valuable for producing pooled estimates of AI model performance, the methodological diversity and incomplete reporting across studies precluded a meaningful aggregation of results in this review. Future systematic reviews with meta-analysis are recommended, particularly when sufficient homogeneous studies become available, to quantitatively compare the predictive accuracy of AI models across different application domains, data modalities, and model complexities. Finally, while this review comprehensively evaluated the clinical applications of AI during pregnancy and the neonatal period, it excluded studies that focused solely on long-term maternal or offspring outcomes beyond the immediate postpartum period. Given the increasing interest in the long-term metabolic and cardiovascular risks associated with GDM exposure, future research should extend the scope to assess how AI can support prediction and prevention strategies beyond birth.

Conclusions

This systematic review provides the most comprehensive synthesis to date of clinical applications of AI in GDM. Analyzing 126 studies, we observed a rapid growth in research efforts, reflecting increasing recognition of AI’s potential to transform GDM care. The majority of studies focused on early prediction, screening, and diagnosis using structured clinical data, with fewer addressing treatment personalization, glycemic control, or neonatal outcomes. Methodologically, classical ML models dominated, while DL approaches are emerging as the field matures. Despite encouraging model performances, critical challenges remain, including limited external validation, underreporting of methodological details, and insufficient integration of interpretability mechanisms.

Addressing these gaps will be crucial to advancing AI from theoretical promise to clinical utility. Future research must prioritize prospective, multicenter studies, model transparency, standardized reporting, and collaboration with clinical end-users. Moreover, expanding AI applications to support the full continuum of GDM care—from early risk prediction to postpartum follow-up—will be essential for maximizing patient benefit. Overall, while AI holds strong promise for enhancing the management of GDM, realizing its clinical impact will require a more rigorous, transparent, and patient-centered research agenda.

Supplemental Material

sj-docx-1-dst-10.1177_19322968251355967 – Supplemental material for Artificial Intelligence in Gestational Diabetes Care: A Systematic Review

Supplemental material, sj-docx-1-dst-10.1177_19322968251355967 for Artificial Intelligence in Gestational Diabetes Care: A Systematic Review by Rawan AlSaad, Ali Elhenidy, Aliya Tabassum, Nour Odeh, Eman AboArqoub, Aya Odeh, Maya AlTamimi, Alaa Abd-alrazaq, Rajat Thomas, Mohammed Bashir and Javaid Sheikh in Journal of Diabetes Science and Technology

Supplemental Material

sj-docx-2-dst-10.1177_19322968251355967 – Supplemental material for Artificial Intelligence in Gestational Diabetes Care: A Systematic Review

Supplemental material, sj-docx-2-dst-10.1177_19322968251355967 for Artificial Intelligence in Gestational Diabetes Care: A Systematic Review by Rawan AlSaad, Ali Elhenidy, Aliya Tabassum, Nour Odeh, Eman AboArqoub, Aya Odeh, Maya AlTamimi, Alaa Abd-alrazaq, Rajat Thomas, Mohammed Bashir and Javaid Sheikh in Journal of Diabetes Science and Technology

Supplemental Material

sj-docx-3-dst-10.1177_19322968251355967 – Supplemental material for Artificial Intelligence in Gestational Diabetes Care: A Systematic Review

Supplemental material, sj-docx-3-dst-10.1177_19322968251355967 for Artificial Intelligence in Gestational Diabetes Care: A Systematic Review by Rawan AlSaad, Ali Elhenidy, Aliya Tabassum, Nour Odeh, Eman AboArqoub, Aya Odeh, Maya AlTamimi, Alaa Abd-alrazaq, Rajat Thomas, Mohammed Bashir and Javaid Sheikh in Journal of Diabetes Science and Technology

Supplemental Material

sj-docx-4-dst-10.1177_19322968251355967 – Supplemental material for Artificial Intelligence in Gestational Diabetes Care: A Systematic Review

Supplemental material, sj-docx-4-dst-10.1177_19322968251355967 for Artificial Intelligence in Gestational Diabetes Care: A Systematic Review by Rawan AlSaad, Ali Elhenidy, Aliya Tabassum, Nour Odeh, Eman AboArqoub, Aya Odeh, Maya AlTamimi, Alaa Abd-alrazaq, Rajat Thomas, Mohammed Bashir and Javaid Sheikh in Journal of Diabetes Science and Technology

Supplemental Material

sj-docx-5-dst-10.1177_19322968251355967 – Supplemental material for Artificial Intelligence in Gestational Diabetes Care: A Systematic Review

Supplemental material, sj-docx-5-dst-10.1177_19322968251355967 for Artificial Intelligence in Gestational Diabetes Care: A Systematic Review by Rawan AlSaad, Ali Elhenidy, Aliya Tabassum, Nour Odeh, Eman AboArqoub, Aya Odeh, Maya AlTamimi, Alaa Abd-alrazaq, Rajat Thomas, Mohammed Bashir and Javaid Sheikh in Journal of Diabetes Science and Technology

Footnotes

Acknowledgements

None.

Abbreviations

ACC, accuracy; AI, artificial intelligence; AUC-ROC, area under the receiver operating characteristic curve; BMI, body mass index; CM, confusion matrix; CNN, convolutional neural network; CV, cross-validation; DL, deep learning; DNN, deep neural network; DSS, decision support system; EHR, electronic health record; ET, extra trees; GDM, gestational diabetes mellitus; GEO, gene expression omnibus; KNN, K-nearest neighbors; LASSO, least absolute shrinkage and selection operator; LDA, linear discriminant analysis; LGB, light gradient boosting; LIME, local interpretable model-agnostic explanations; LOOCV, leave-one-out cross-validation; LSTM, long short-term memory; ML, machine learning; MLP, multilayer perceptron; mHealth, mobile health; NPV, negative predictive value; OGTT, oral glucose tolerance test; PCA, principal component analysis; PLS, partial least squares; PLS-LDA, partial least squares discriminant analysis; PPV, positive predictive value; PROBAST, prediction model risk of bias assessment tool; QUADAS-2, quality assessment of diagnostic accuracy studies-2; RNN, recurrent neural network; SENS, sensitivity; SGD, stochastic gradient descent; SHAP, shapley additive explanations; SPEC, specificity; SVM, support vector machine; TL, transfer learning; T2DM, type 2 diabetes mellitus; TRIPOD-AI, transparent reporting of a multivariable prediction model for individual prognosis Or diagnosis–artificial intelligence; XAI, explainable artificial intelligence; XGBoost, extreme gradient boosting.

Author Contributions

Study design: R.A. Databases search: A.A. Study screening: N.O., E.A.A., A.O., M.A., and R.A. Data extraction: A.T., N.O., E.A.A., A.O., M.A., A.E., and R.A. Risk of bias assessment: A.E. and R.A. Manuscript writing: R.A. Manuscript review: M.B., R.T., A.A., and J.S. All authors have read and approved the final version of the manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Open Access funding provided by the Qatar National Library.

Ethics Approval and Consent to Participate

Not applicable.

Consent for Publication

Not applicable.

ORCID iD

Rawan AlSaad

Availability of Data and Materials

All data generated during this study are provided in the Supplementary Materials. In addition, a curated collection of all research articles included in this review is publicly available on GitHub at . The repository will be continuously updated, as new articles on AI applications in gestational diabetes are published.

Code Availability

This study did not involve the utilization of any custom code or mathematical algorithm.

Supplemental Material

Supplemental material for this article is available online.

References

Iman

AEH

Huniadi

Sandor

Zaha

Rotar

Iuhas

. Prevalence and risk factors of gestational diabetes mellitus in Romania: maternal and fetal outcomes. Medicina (Kaunas). 2025;61(2):194.

Reitzle

Heidemann

Krause

Hoebel

Scheidt-Nave

. Prevalence of gestational diabetes mellitus in Germany: temporal trend and differences by regional socioeconomic deprivation. J Health Monit. 2024;9(2):e12086.

Dewi

Martini

Isfandiari

. Prevalence and risk factors for gestational diabetes in Jambi, Indonesia. Afr J Reprod Health. 2024;28(10s):118-124.

Bashir

Ibrahim

Beer

, et al. Integrated care of diabetes during pregnancy: a Qatari nationwide cohort. EClinicalMedicine. 2024;72:102605.

Zhang

Xie

Cai

Zhang

. The impact of gestational diabetes mellitus on the development of preeclampsia in twin pregnancies: a retrospective cohort study conducted at a tertiary hospital. Int J Womens Health. 2025;17:75-86.

Bernea

Uyy

Mihai

, et al. New born macrosomia in gestational diabetes mellitus. Exp Ther Med. 2022;24(6):710.

Cao

Yang

Liu

. Analysis of risk factors of neonatal hypoglycemia and its correlation with blood glucose control of gestational diabetes mellitus: a retrospective study. Medicine (Baltimore). 2023;102(35):e34619.

Zhao

Yao

Zhang

, et al. Predictive value of machine learning for the progression of gestational diabetes mellitus to type 2 diabetes: a systematic review and meta-analysis. BMC Med Inform Decis Mak. 2025;25(1):18.

Sharma

Singh

Kamal

Kumar

. From pregnancy to postpartum: the cardiovascular risks associated with gestational diabetes[published online ahead of print April 17, 2025]. Curr Cardiol Rev. doi:10.2174/011573403X354645250329171327.

10.

Mantzorou

Papandreou

Pavlidou

, et al. Maternal gestational diabetes is associated with high risk of childhood overweight and obesity: a cross-sectional study in pre-school children aged 2-5 years. Medicina (Kaunas). 2023;59(3):455.

11.

Champion

Jauk

Biggio

Sychowski

Tita

Harper

. Early gestational diabetes screening based on ACOG guidelines. Am J Perinatol. 2024;41(S 01):e641-e647

12.

Bakshi

Kumar

Gupta

Radhika

Misra

Bhardwaj

. Review of the screening guidelines for gestational diabetes mellitus: how to choose wisely. Indian J Community Med. 2023;48(6):828-834.

13.

Li-Zhen

Yun

Xiao-Dong

, et al. Evaluation of guidelines on the screening and diagnosis of gestational diabetes mellitus: systematic review. BMJ Open. 2019;9(5):e023014.

14.

Kokori

Olatunji

Aderinto

, et al. The role of machine learning algorithms in detection of gestational diabetes; a narrative review of current evidence. Clin Diabetes Endocrinol. 2024;10(1):18.

15.

Daley

Ni’Man

Neves

, et al. mHealth apps for gestational diabetes mellitus that provide clinical decision support or artificial intelligence: a scoping review. Diabet Med. 2022;39(1):e14735.

16.

Gadhia

Loyal

. Review of genetic and artificial intelligence approaches to improving gestational diabetes mellitus screening and diagnosis in sub-Saharan Africa. Yale J Biol Med. 2024;97(1):67-72.

17.

Mennickent

Rodríguez

Farías-Jofré

Araya

Guzmán-Gutiérrez

. Machine learning-based models for gestational diabetes mellitus prediction before 24-28 weeks of pregnancy: a review. Artif Intell Med. 2022;132:102378.

18.

McInnes

MDF

Moher

Thombs

McGrath

Bossuyt

. Preferred reporting items for a systematic review and meta-analysis of diagnostic test accuracy studies: the PRISMA-DTA statement. JAMA. 2018;319(4):388-396.

19.

Whiting

Rutjes

AWS

Westwood

, et al. QUADAS-2: a revised tool for the quality assessment of diagnostic accuracy studies. Ann Intern Med. 2011;155(8):529-536.

20.

Wolff

Moons

KGM

Riley

, et al. PROBAST: a tool to assess the risk of bias and applicability of prediction model studies. Ann Intern Med. 2019;170(1):51-58.

21.

Abbineni

Rani

Jambavathi

Bhavitha

. Machine learning revolutionizing in gestational diabetes care. In: Venu Gopal Rao

Krishna Prasad

Vijaya Bhaskar

, eds. International Conference on Advances in Computational Intelligence. Cham: Springer; 2023:58-68.

22.

Abe

Obe

Boyinbode

Olagbuji

. Early gestational diabetes mellitus diagnosis using classification algorithms: an ensemble approach. Paper presented at 2023 IEEE AFRICON; September 2023; Nairobi.

23.

Al Jahwari

Al Owisi

Al Mamari

Subramanian

Hajamohideen

. Gestational diabetes mellitus detection system-an advanced preventive measure using DNN. Paper presented at 2023 International Conference on Emerging Techniques in Computational Intelligence (ICETCI); September 2023; Nairobi.

24.

Al-Azzawi

Kurnaz

. Evaluating machine learning algorithms for early detection of GDM. Paper presented at 2024 International Congress on Human-computer Interaction, Optimization and Robotic Applications (HORA); June 2024; Istanbul.

25.

Ali

Khan

Ahmad

Masud

Adam

Ahmed

. Predictive modeling for the diagnosis of gestational diabetes mellitus using epidemiological data in the United Arab Emirates. Information. 2022;13(10):485.

26.

Al-kateeb

Abdullah

. A smart architecture leveraging fog computing fusion and ensemble learning for prediction of gestational diabetes. Fusion Pract Appl. 2023;12(2):70-87.

27.

Alotaibi

Aljehane

. Early prediction of gestational diabetes using machine learning techniques. J Theor Appl Inf Technol. 2023;101:21.

28.

Amarnath

Selvamani

Varadarajan

. Prognosis model for gestational diabetes using machine learning techniques. Sens Mater. 2021;33:3011-3025.

29.

Assegie

Suresh

Purushothaman

Ganesan

Kumar

. Early prediction of gestational diabetes with parameter-tuned K-nearest neighbor classifier. J Robot Control. 2023;4(4):452-457.

30.

Balani

Hyer

Shehata

Mohareb

. Visceral fat mass as a novel risk factor for predicting gestational diabetes in obese pregnant women. Obstet Med. 2018;11(3):121-125.

31.

Belsti

Moran

, et al. Comparison of machine learning and conventional logistic regression-based prediction models for gestational diabetes in an ethnically diverse population; the Monash GDM machine learning model. Int J Med Inform. 2023;179:105228.

32.

Bigdeli

Ghazisaedi

Ayyoubzadeh

Hantoushzadeh

Ahmadi

. Predicting gestational diabetes mellitus in the first trimester using machine learning algorithms: a cross-sectional study at a hospital fertility health center in Iran. BMC. 2025;25(1):3.

33.

Caballero-Ruiz

García-Sáez

Rigla

, et al. Automatic blood glucose classification for gestational diabetes with feature selection: decision trees vs. neural networks. Paper presented at XIII Mediterranean Conference on Medical and Biological Engineering and Computing 2013: MEDICON 2013; 25-28 September, 2013; Seville.

34.

Caballero-Ruiz

García-Sáez

Rigla

Villaplana

Pons

Hernando

. A web-based clinical decision support system for gestational diabetes: Automatic diet prescription and detection of insulin needs. Int J Med Inform. 2017;102:35-49.

35.

Catic

Heinzl

Göbl

Yerlikaya-Schatten

Reischer

. Risk stratification in twin pregnancies complicated by GDM. J Diabetes Res. 2024;2024:5203116.

36.

Chan

Y-N

Wang

Chun

K-H

Lum

JT-S

Wang

Zhang

, et al. A machine learning approach for early prediction of gestational diabetes mellitus using elemental contents in fingernails. Sci Rep. 2023;13(1):4184.

37.

Chen

Guo

Yan

. Predicting recurrent gestational diabetes mellitus using artificial intelligence models: a retrospective cohort study. Arch Gynecol Obstet. 2024;310(3):1621-1630.

38.

Chen

Wan

Gong

Zhang

Liang

Wang

. Assessing the relationship between pregravid body mass index and risk of adverse maternal pregnancy and neonatal outcomes: prospective data in Southwest China. Sci Rep. 2021;11(1):7591.

39.

Comlan

Alokpo

, eds. Classification of gestational diabetes using machine learning techniques. In: Applied Mathematics, Modeling and Computer Simulation. Amsterdam: IOS Press; 2023:1224-1234.

40.

Cubillos

Monckeberg

Plaza

, et al. Development of machine learning models to predict gestational diabetes risk in the first half of pregnancy. BMC Pregnancy Childbirth. 2023;23(1):469.

41.

Cui

Yang

, et al. Elevated liver enzymes in the first trimester are associated with gestational diabetes mellitus: a prospective cohort study. Diabetes Metab Res Rev. 2024;40(4):e3799.

42.

Debata

Mohapatra

. Diagnosis of diabetes in pregnant woman using a Chaotic-Jaya hybridized extreme learning machine model. J Integr Bioinform. 2021;18(1):81-99.

43.

Zhong

, et al. Prediction of pregnancy diabetes based on machine learning. Paper presented at BIBE 2019; the Third International Conference on Biological Information and Biomedical Engineering; June 2019; Hangzhou, China.

44.

Rafferty

McAuliffe

Wei

Mooney

. An explainable machine learning-based clinical decision support system for prediction of gestational diabetes mellitus. Sci Rep. 2022;12(1):1170.

45.

Eleftheriades

Chatzakis

Papachatzopoulou

, et al. Prediction of insulin treatment in women with gestational diabetes mellitus. Nutr Diabetes. 2021;11(1):30.

46.

Gallardo-Rincón

Ríos-Blancas

Ortega-Montiel

, et al. MIDO GDM: an innovative artificial intelligence-based prediction model for the development of gestational diabetes in Mexican women. Sci Rep. 2023;13(1):6992.

47.

Hotta

Kytö

Koivusalo

Heinonen

Marttinen

. Optimizing postprandial glucose prediction through integration of diet and exercise: leveraging transfer learning with imbalanced patient data. PLoS ONE. 2024;19(8):e0298506.

48.

Hou

Cheng

Kang

Zheng

. Prediction of gestational diabetes based on lightgbm. Paper presented at Proceedings of the 2020 Conference on Artificial Intelligence and Healthcare; October 2020; Taiyuan, China.

49.

Hou

Gao

Poon

, et al. Maternal plasma diacylglycerols and triacylglycerols in the prediction of gestational diabetes mellitus. BJOG. 2023;130(3):247-256.

50.

Wang

. Prediction model for gestational diabetes mellitus using the XG Boost machine learning algorithm. Front Endocrinol (Lausanne). 2023;14:1105062.

51.

Zhang

. Establishment of clinical diagnostic models using glucose, lipid, and urinary polypeptides in gestational diabetes mellitus. J Clin Lab Anal. 2021;35(7):e23833.

52.

Huynh

Hoang

. A novel approach for determining meal plan for gestational diabetes mellitus using artificial intelligence. Comput J. 2022;65(5):1088-1097.

53.

Jader

Aminifar

. Fast and accurate artificial neural network model for diabetes recognition. Neuroquantology. 2022;20(10):2187-2196.

54.

Jader

Aminifar

Abd

MHM

. Diabetes detection system by mixing supervised and unsupervised algorithms. J Stud Sci Eng. 2022;2(3):52-65.

55.

Kang

Lee

Hong

, et al. Prediction of gestational diabetes mellitus in Asian women using machine learning algorithms. Sci Rep. 2023;13(1):13356.

56.

Kang

Liang

Wang

, et al. Prediction model comparison for gestational diabetes mellitus with macrosomia based on risk factor investigation. J Matern Fetal Neonatal Med. 2021;34(15):2481-2490.

57.

Kaya

Bütün

Çelik Salik

ÖEA

Tahta

Yavuz

. The early prediction of gestational diabetes mellitus by machine learning models. BMC Pregnancy Childbirth. 2024;24(1):574.

58.

Kennedy

Selvam

Palanisamy

. Cloud-centric IoT based decision support system for gestational diabetes mellitus using optimal support vector machine. IJRTE. 2019:8;2277-3878.

59.

Kolozali

White

Norris

Fasli

van Heerden

. Explainable early prediction of gestational diabetes biomarkers by combining medical background and wearable devices: a pilot study with a cohort group in South Africa. IEEE J Biomed Health Inform. 2024;28(4):1860-1871.

60.

Kumar

Rajalakshmi

Shashikala

Maithli

Aman

Rajneesh

. Gestational diabetes detection using machine learning algorithm: research challenges of big data and data mining. Int J Intell Syst Appl Eng. 2022;10(2 s):260-263.

61.

Kumar

Ang

Png

, et al. Automated machine learning (AutoML)-derived preconception predictive risk model to guide early intervention for gestational diabetes mellitus. Int J Environ Res Public Health. 2022;19(11):6792.

62.

Kumar

Chen

Tan

, et al. Population-centric risk prediction modeling for gestational diabetes mellitus: a machine learning approach. Diabetes Res Clin Pract. 2022;185:109237.

63.

Kurt

Gürlek

Keskin

, et al. A prospective study for gestational diabetes mellitus: analysis of risk factors in Turkish women for early prediction. Int J Gynaecol Obstet. 2023;161(2):525-535.

64.

Kurt

Gürlek

Keskin

, et al. Prediction of gestational diabetes using deep learning and Bayesian optimization and traditional machine learning techniques. Med Biol Eng Comput. 2023;61(7):1649-1660.

65.

Lee

Hwangbo

Norwitz

, et al. Nonalcoholic fatty liver disease and early prediction of gestational diabetes mellitus using machine learning methods. Clin Mol Hepatol. 2022;28(1):105-116.

66.

Leszczełowska

Mazur-Milecka

Kowalczyk

Sobotka

. Maternal health risk assessment using digital twin application. Paper presented at 2024 16th International Conference on Human System Interaction (HSI); July 2024; Paris.

67.

Luo

Wang

. Identification of diagnostic biomarkers in patients with gestational diabetes mellitus based on transcriptome gene expression and methylation correlation analysis. Reprod Biol Endocrinol. 2019;17(1):112.

68.

Zheng

, et al. The oral microbiome of pregnant women facilitates gestational diabetes discrimination. J Genet Genomics. 2021;48(1):32-39.

69.

Liu

Wang

Huang

. Prediction of gestational diabetes mellitus at the first trimester: machine-learning algorithms. Arch Gynecol Obstet. 2024;309(6):2557-2566.

70.

Liao

Ferrara

Greenberg

, et al. Development and validation of prediction models for gestational diabetes treatment modality using supervised machine learning: a population-based cohort study. BMC Med. 2022;20(1):307.

71.

Lin

Han

Yan

Chen

Zhang

. Gestational weight management and pregnancy outcomes among women of advanced maternal age. Exp Ther Med. 2019;18(3):1723-1728.

72.

Lin

Fang

. Establishment and evaluation of a risk prediction model for gestational diabetes mellitus. World J Diabetes. 2023;14(10):1541-1550.

73.

Liu

Leng

, et al. Machine learning risk score for prediction of gestational diabetes in early pregnancy in Tianjin, China. Diabetes Metab Res Rev. 2021;37(5):e3397.

74.

Liu

Zhan

Liu

, et al. Stacking ensemble method for gestational diabetes mellitus prediction in Chinese pregnant women: a prospective cohort study. J Healthc Eng. 2022;2022:8948082.

75.

Liu

Geng

Duan

Yang

Ding

. Identification of diagnostic CpG signatures in patients with gestational diabetes mellitus. Biomed Res Int. 2021;2021:1984690.

76.

Liu

Wang

Zhang

Cheng

. Detection of gestational diabetes mellitus and influence on perinatal outcomes from B-mode ultrasound images using deep neural network. Sci Program. 2021;2021(1):2283768.

77.

Liu

Wang

Zhao

. Identification of diagnostic cytosine-phosphate-guanine biomarkers in patients with gestational diabetes mellitus. Gynecol Endocrinol. 2021;37(9):857-862.

78.

Hirst

Yang

Mackillop

Clifton

. Standardising the assessment of caesarean birth using an oxford caesarean prediction score for mothers with gestational diabetes. Healthc Technol Lett. 2022;9(1-2):1-8.

79.

Hirst

Mackillop

Clifton

. A stacked long short-term memory approach for predictive blood glucose monitoring in women with gestational diabetes mellitus. Sensors. 2023;23(18):7990.

80.

Luo

Fang

, et al. Discovery of metabolic biomarkers for gestational diabetes mellitus in a Chinese population. Nutr Metab. 2021;18(1):79.

81.

Wang

Cai

Xing

Huang

. Prediction of gestational diabetes and hypertension based on pregnancy examination data. J Mech Med Biol. 2022;22(03):2240001.

82.

Manasrah

Alkayem

Qasaimeh

Nofal

. Assessment of machine learning security: the case of healthcare data. Paper presented at International Conference on Data Science, E-learning and Information Systems 2021; April 2021; Ma’an, Jordan.

83.

Mathew

Cutshaw

Appel

, et al. Diagnosis of pregnancy disorder in the first-trimester patient plasma with Raman spectroscopy and protein analysis. Bioeng Transl Med. 2024;9(6):e10691.

84.

McLaren

Jr Haberman

Moscu

Atallah

Friedmann

. A novel and precise profiling tool to predict gestational diabetes. J Diabetes Sci Technol. 2021;15(4):891-896.

85.

Mennickent

Ortega-Contreras

Gutiérrez-Vega

, et al. Evaluation of first and second trimester maternal thyroid profile on the prediction of gestational diabetes mellitus and post load glycemia. PLoS ONE. 2023;18(1):e0280513.

86.

Mennickent

Romero-Albornoz

Gutiérrez-Vega

, et al. Simple and fast prediction of gestational diabetes mellitus based on machine learning and near-infrared spectra of serum: a proof of concept study at different stages of pregnancy. Biomedicines. 2024;12(6):1142.

87.

Minaei

Dosti

Turk

Golabpour

. Provide a diagnostic model using a combination of two neural network algorithms and a genetic algorithm. Front health Inform. 2021;10(1):91.

88.

Moreira

MWL

Rodrigues

JJPC

Kumar

Al-Muhtadi

Korotaev

. Evolutionary radial basis function network for gestational diabetes data analytics. J Comput Sci. 2018;27:410-417.

89.

Moreira

MWL

Rodrigues

JJPC

Kumar

Niu

Sangaiah

. Multilayer Perceptron Application for Diabetes Mellitus Prediction in Pregnancy Care. Singapore: Springer; 2018.

90.

Muller

Nirmala

. Diagnosis of gestational diabetes mellitus using radial basis function. Paper presented at 2016 Online International Conference on Green Engineering and Technologies (IC-GET); November 2016; Coimbatore, India.

91.

Nagarajan

Chandrasekaran

Ramasubramanian

. Supervised machine learning techniques for predicting the risk levels of gestational diabetes mellitus. 2015;10:38729-38732.

92.

Narwane

Sawarkar

. Is handling unbalanced datasets for machine learning uplifts system performance?: a case of diabetic prediction. Diabetes Metab Syndr. 2022;16(9):102609.

93.

Neves

Daley

Hitman

, et al. Causal dynamic Bayesian networks for the management of glucose control in gestational diabetes. Paper presented at 2021 IEEE 9th International Conference on Healthcare Informatics (ICHI); August 2021; Victoria, BC.

94.

Nisana Siddegowda

Mullur Puttabuddi

. Analysis of association between caesarean delivery and gestational diabetes mellitus using machine learning. Proc Eng Technol Innov. 2020;15:08-15.

95.

Ojugo

Eboka

Yoro

Yerokun

Efozia

. Hybrid model for early diabetes diagnosis. Paper presented at 2015 Second International Conference on Mathematics and Computers in Sciences and in Industry (MCSI); August 2015; Sliema.

96.

Ojugo

Otakore

. Improved early detection of gestational diabetes via intelligent classification models: a case of the Niger Delta region in Nigeria. J Comput Appl. 2018;6(2):82-90.

97.

Pan

Jin

Wang

Dai

. Placenta inflammation is closely associated with gestational diabetes mellitus. Am J Transl Res. 2021;13(5):4068-4079.

98.

Pathiranage

Suraweera

SMDAS

. DiabetCare—a classifier based mobile application for predicting the risk of prediabetes mellitus, gestational diabetes mellitus and type 2 diabetes mellitus. Paper presented at 2024 International Research Conference on Smart Computing and Systems Engineering (SCSE); April 2024; Colombo.

99.

Pokala

VSK

Kumar

. Analysis and comparison for prediction of diabetic among pregnant women using innovative decision tree algorithm over support vector machine algorithm with improved accuracy. J Pharm Negat Results. 2022;13:166-174.

100.

Pokala

VSK

Kumar

. Analysis and comparison for prediction of diabetic among pregnant women using innovative K-nearest neighbor algorithm over logistic regression with improved accuracy. Cardiometry. 2022;(25):949-955.

101.

Pokala

VSK

Kumar

. Analysis and comparison for prediction of diabetic among pregnant women using innovative support vector machine algorithm over random forest algorithm with improved accuracy. Cardiometry. 2022;(25):956-962.

102.

Pokala

VSK

Kumar

. Analysis and comparison for prediction of diabetic pregnant women using innovative principal component analysis algorithm over support vector machine algorithm with improved accuracy. Cardiometry. 2022;(25):942-948.

103.

Polak

Mendyk

. Artificial intelligence technology as a tool for initial GDM screening. Expert Syst Appl. 2004;26(4):455-460.

104.

Posonia

Vigneshwari

Rani

. Machine learning based diabetes prediction using decision tree J48. Paper presented at 2020 3rd International Conference on Intelligent Sustainable Systems (ICISS); December 2020; Thoothukudi, India.

105.

Pustozerov

Tkachuk

Vasukova

, et al. The role of glycemic index and glycemic load in the development of real-time postprandial glycemic response prediction models for patients with gestational diabetes. Nutrients. 2020;12(2):302.

106.

Qiu

Wang

, et al. Electronic health record driven prediction for gestational diabetes mellitus in early pregnancy. Sci Rep. 2017;7(1):16417.

107.

Rigla

Martínez-Sarriegui

García-Sáez

Pons

Hernando

. Gestational diabetes management using smart mobile telemedicine. J Diabetes Sci Technol. 2018;12(2):260-264.

108.

Riyed

Nabi

Dutta

Hasan

Anik

Ornob

. Gestational diabetes prediction in pregnancy: a machine learning and data preprocessing approach. Paper presented at 2023 26th International Conference on Computer and Information Technology (ICCIT); December 2023; Cox’s Bazar, Bangladesh.

109.

Saleh

Yacoub

Salaheldin

. Machine learning_based paradigm for diagnosis of gestational diabetes. Paper presented at 2023 3rd International Conference on Electronic Engineering (ICEEM); October 2023; Menouf.

110.

Salem

Shams

Elzeki

Abd Elfattah

Al-Amri

Elnazer

. Fine-tuning fuzzy KNN classifier based on uncertainty membership for the medical diagnosis of diabetes. Appl Sci. 2022;12(3):950.

111.

Salvatori

Linder

Eppel

Morettini

Göbl

Tura

. Empirical index for easy assessment of pancreatic beta-cell glucose sensitivity during pregnancy: a machine learning approach. In 2022 E-health and Bioengineering Conference (EHB); November 2022; Iasi.

112.

Saradha

Sujatha

. Gestational diabetes diagnosis with MSVM, MJ48 classifier models. Int J Recent Technol Eng. 2019;8(2S11):244-248.

113.

Seo

Park

Kim

Jin

Park

. A personalized blood glucose level prediction model with a fine-tuning strategy: a proof-of-concept study. Comput Methods Programs Biomed. 2021;211:106424.

114.

Setiawan

Zanetta

Kesuma

KAC

Edbert

Suhartono

. Predicting diabetes risk in pregnant women: leveraging AdaBoost as a boosting algorithm. Paper presented at 2023 International Conference on Advanced Mechatronics, Intelligent Manufacture and Industrial Automation (ICAMIMIA); November 2023; Surabaya, Indonesia.

115.

Shen

Chen

Zheng

, et al. An innovative artificial intelligence–based app for the diagnosis of gestational diabetes mellitus (GDM-AI): development study. J Med Internet Res. 2020;22(9):e21573.

116.

Shiva Shankar

Gadiraju

Preethi

Rao

VVRM

. A novel approach for prediction of gestational diabetes based on clinical signs and risk factors. EAI Trans Scalable Inf Syst. 2023;10(3):e8.

117.

Subhash

Ashwin Kumar

. Accuracy of classification algorithms for diabetes prediction. Int J Eng Adv Technol. 2019;8:2249-8958.

118.

Sumathi

Meganathan

. Machine learning based pattern detection technique for diabetes mellitus prediction. Concurr Comput Pract Exp. 2022;34(6):e6751.

119.

Sun

Jiao

Ren

Guo

Wang

. Model application to quantitatively evaluate placental features from ultrasound images with gestational diabetes. J Clin Ultrasound. 2022;50(7):976-983.

120.

Sujatha

Ananthapadmanaban

. Data-driven insights into gestational diabetes mellitus: enhancing models for prediction by SVM imputation for personalized pregnancy care. Int J Electron Commun. 2024;11:140-150.

121.

Tang

Wang

, et al. Longitudinal integrative cell-free DNA analysis in gestational diabetes mellitus. Cell Rep Med. 2024;5(8):101660.

122.

Velardo

Clifton

Hamblin

Khan

Tarassenko

Mackillop

. Toward a multivariate prediction model of pharmacological treatment for women with gestational diabetes mellitus: algorithm development and validation. J Med Internet Res. 2021;23(3):e21435.

123.

Vivek Khanna

Chadaga

Sampathila

, et al. Explainable artificial intelligence-driven gestational diabetes mellitus prediction using clinical and laboratory markers. Cogent Eng. 2024;11(1):2330266.

124.

Wang

Guo

Jing

, et al. Development and validation of risk prediction models for gestational diabetes mellitus using four different methods. Metabolites. 2022;12(11):1040.

125.

Wang

Zhang

Yao

. Analysis and prediction of gestational diabetes mellitus by the ensemble learning method. Int J Comput Intell Syst. 2022;15(1):72.

126.

Wang

Sun

Zhao

, et al. Identify gestational diabetes mellitus by deep learning model from cell-free DNA at the early gestation stage. Brief Bioinform. 2024;25(1):bbad492.

127.

Wang

Zhang

. Identification of diagnostic biomarker in patients with gestational diabetes mellitus based on transcriptome-wide gene expression and pattern recognition. J Cell Biochem. 2019;120(2):1503-1510.

128.

Watanabe

Eguchi

Sakurai

Yamamoto

Mori

. Prediction of gestational diabetes mellitus using machine learning from birth cohort data of the Japan environment and children’s study. Sci Rep. 2023;13:17419.

129.

Wei

L-L

Pan

Y-S

Zhang

Chen

Wang

H-Y

Wang

J-Y

. Application of machine learning algorithm for predicting gestational diabetes mellitus in early pregnancy†. Front Nurs. 2021;8(3):209-221.

130.

Wang

, et al. A risk prediction model of gestational diabetes mellitus before 16 gestational weeks in Chinese pregnant women. Diabetes Res Clin Pract. 2021;179:109001.

131.

Y-T

Zhang

C-J

Mol

, et al. Early prediction of gestational diabetes mellitus in the Chinese population via advanced machine learning. J Clin Endocrinol Metab. 2020;106(3):e1191-e1205.

132.

Xing

Dong

Liu

, et al. Enhancing gestational diabetes mellitus risk assessment and treatment through GDMPredictor: a machine learning approach. J Endocrinol Invest. 2024;47(9):2351-2360.

133.

Xiong

Lin

Chen

, et al. Prediction of gestational diabetes mellitus in the first 19 weeks of pregnancy using machine learning techniques. J Matern Fetal Neonatal Med. 2022;35(13):2457-2463.

134.

Yang

Clifton

Hirst

, et al. Machine learning-based risk stratification for gestational diabetes management. Sensors. 2022;22(13):4805.

135.

Yang

M-N

Zhang

Wang

W-J

, et al. Prediction of gestational diabetes mellitus by multiple biomarkers at early gestation. BMC Pregnancy Childbirth. 2024;24(1):601.

136.

Xiong

Zhou

Xiao

. Comparison of machine learning methods and conventional logistic regressions for predicting gestational diabetes using routine clinical data: a retrospective cohort study. J Diabetes Res. 2020;2020:4168340.

137.

Zermane

Kalla

. Statistical and machine learning-based predictive models for gestational diabetes mellitus prevention. 2024;30:38-55.

138.

Zhang

Wang

. Prediction of gestational diabetes mellitus under cascade and ensemble learning algorithm. Comput Intell Neurosci. 2022;2022:3212738.

139.

Zhang

Yang

, et al. Clinical significance of the genetically variable landscape of the gut microbiome in patients with gestational diabetes mellitus patients. Heliyon. 2024;10(18):e37986.

140.

Zheng

Wang

, et al. A simple model to predict risk of gestational diabetes mellitus from 8 to 20 weeks of gestation in Chinese women. BMC Pregnancy Childbirth. 2019;19(1):252.

141.

Zhou

Ran

Song

, et al. A stepwise prediction and interpretation of gestational diabetes mellitus: foster the practical application of machine learning in clinical decision. Heliyon. 2024;10(12):e32709.

142.

Zhou

Chen

, et al. Predictive value of ultrasonic artificial intelligence in placental characteristics of early pregnancy for gestational diabetes mellitus. Front Endocrinol (Lausanne). 2024;15:1344666.

143.

Zhou

Xie

Chen

. Prediction of birth weight in pregnancy with gestational diabetes mellitus using an artificial neural network. J Zhejiang Univ Sci B. 2022;23(5):432-436.

144.

Zhu

Barupal

Ngo

, et al. Predictive metabolomic markers in early to mid-pregnancy for gestational diabetes mellitus: a prospective test and validation study. Diabetes. 2022;71(8):1807-1817.

145.

Zhu

Y-T

Xiang

L-L

Chen

Y-J

Zhong

T-Y

Wang

J-J

Zeng

. Developing and validating a predictive model of delivering large-for-gestational-age infants among women with gestational diabetes mellitus. World J Diabetes. 2024;15(6):1242.

146.

TRIPOD-AI CONSORT-AI; Collins

Moons

KGM

. Reporting guidelines for artificial intelligence (AI) prediction models: the TRIPOD-AI and PROBAST-AI initiatives. BMJ. 2021;372:n365. doi:10.1136/bmj.n365.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.08 MB

0.03 MB

0.04 MB

0.03 MB

0.36 MB