Article Text
PDF
Abstract
Introduction Pre-eclampsia is one of the most serious clinical problems of pregnancy that contribute significantly to maternal mortality worldwide. This systematic review aims to identify and summarise the predictive factors of pre-eclampsia using machine learning models and evaluate the diagnostic accuracy of machine learning models in predicting pre-eclampsia.
Methods and analysis This review adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. This search strategy includes the search for published studies from inception to January 2023. Databases include the Cochrane Central Register, PubMed, EMBASE, ProQuest, Scopus and Google Scholar. Search terms include ‘preeclampsia’ AND ‘artificial intelligence’ OR ‘machine learning’ OR ‘deep learning’. All studies that used machine learning-based analysis for predicting pre-eclampsia in pregnant women will be considered. Non-English articles and those that are unrelated to the topic will be excluded. PROBAST (Prediction model Risk Of Bias ASsessment Tool) will be used to assess the risk of bias and the applicability of each included study.
Ethics and dissemination Ethical approval is not required, as our review will include published and publicly accessible data. Findings from this review will be disseminated via publication in a peer-review journal.
PROSPERO registration number This review is registered with PROSPERO (ID: CRD42023432415).
- obstetrics
- maternal medicine
- prenatal diagnosis
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
STRENGTHS AND LIMITATIONS OF THIS STUDY
A systematic review will provide most of the evidence for developing the predictive model for pre-eclampsia.
This review will be thoroughly, independently double-checked at each stage and follows best practice guidelines.
Our review will adhere to the most rigorous methodological guidelines for scoping review to ensure a high-quality review of the evidence.
The exclusion of non-English language papers may limit results.
Introduction
Pre-eclampsia is a hypertensive disorder that usually manifests itself after 20 weeks of pregnancy.1 It can potentially cause severe morbidity, chronic disability and even the death of mothers and babies. With an estimated incidence of 2%–8%, pre-eclampsia is a significant burden on pregnant women.2 In developing countries, the prevalence of pre-eclampsia ranges from 1.8% to 16.7%.3 Globally, about 12% of mothers die only from pre-eclampsia.4 Because of the poorly understood causes, various risk factors and likely multiple pathogenic phenotypes of pre-eclampsia, early prediction of pre-eclampsia is difficult. Statistical learning methods are well-equipped to deal with many variables, such as clinical and laboratory data from patients, and automatically select the most informative features.5 Artificial intelligence has been increasingly used in health and medicine in recent years. The use of artificial intelligence in obstetrics and gynaecology has piqued the scientific community’s interest.6 7
Artificial intelligence has been propelled forward by recent advances in computer science. Conventional general programming algorithms generate outputs based on the input data and the rules provided, whereas artificial intelligence can generate rules and patterns based on the input and output data.8 Artificial intelligence’s pattern recognition and prediction performance have been demonstrated in a variety of medical fields.9 A systematic review of existing prognostic models was deemed necessary to advance efforts to identify women at risk of pre-eclampsia as early and accurately as possible. This would allow existing models to be evaluated for their suitability for immediate use, or to identify those that perform well internally but require external validation on an independent cohort before being considered for clinical use. This approach has the potential to be more efficient than adding a new model to aid in pre-eclampsia prevention. This systematic review aims to identify pre-eclampsia predictors using machine learning approaches that have been reported in previous studies in this field.
Methods/design
The protocol is reported according to the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols10 (online supplemental file 1). The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines11 will be used to report this study. This review is registered with PROSPERO (ID: CRD42023432415). We intended to begin the study in September 2023 and complete it by the end of December 2023.
Supplemental material
Patient and public involvement
Patients and/or the public were not involved in this research.
Objectives
To identify and summarise the predictive factors of pre-eclampsia using machine learning models and to evaluate the accuracy of machine learning models in predicting pre-eclampsia.
Review questions
According to machine learning analysis, what are the predictive factors for pre-eclampsia?
What is the accuracy of machine learning models for predicting pre-eclampsia?
Eligibility criteria
All studies that used machine learning-based analysis for predicting pre-eclampsia in pregnant women will be considered. Non-English articles and those that are unrelated to the topic were excluded. Letters to the editor and reviews were excluded as well.
Search strategy and selection criteria
This strategy will include the search for published studies from inception to January 2023. Databases include the Cochrane Central Register, PubMed, EMBASE (Via Ovid), ProQuest, Scopus and Google Scholar. Search terms include ‘preeclampsia’ AND ‘artificial intelligence’ OR ‘machine learning’ OR ‘deep learning’. Words and phrases will be selected from a controlled vocabulary (Medical Subject Heading (MeSH) and others) and a free-text search for each database. In addition, the reference lists of the identified articles will also be searched along with hand-searching to ensure that all documents are retrieved, which are combined using Boolean ‘OR’ and ‘AND’ operators (online supplemental file 2). All the databases will be searched by an experienced researcher. Two researchers will independently screen the titles and abstracts and then the full texts of potentially eligible studies against the predefined eligibility criteria after eliminating duplicates. Consensus or an appeal to a senior researcher will be used to resolve disagreements. A PRISMA flow diagram will be used to document the study selection process.
Supplemental material
Data collection and risk of bias assessment
Data will be extracted independently by two investigators. Disagreements will be solved by a third party. The following items will be extracted: (1) demographic information (eg, the country where the data were collected, the setting, the data source, the study design, the prediction time and the outcome definition); (2) the method of data partitioning, the algorithms used to select the features, the features used to train the model, the type of predictive model machine learning and the validation and application of the model; and (3) the prediction results (eg, accuracy, sensitivity, specificity and area under the recurrence curve).
Two researchers will independently assess the quality of all included studies and discussed discrepancies until a consensus is achieved. PROBAST,12 which consists of 20 signalling questions divided into four domains (participants, predictors, outcome and analysis), will be used as a tool to assess the risk of bias and applicability of each included study. PROBAST assists in assessing the outcomes studied by considering how it was determined, how objective it is, whether it incorporates any predictor data, how consistently it was determined across individuals, the timing of determination, whether it was independent of predictor information knowledge and whether it matches the review question.
Data synthesis and analysis
Measures of discriminative ability, calibration and classification accuracy will be used to describe model performance. In tabular form, key findings on study design, data sources, prediction model types, sample size, participant characteristics, model objectives, methods, presentation of the final prediction model and outcome measures will be summarised.
Handling missing data
If studies have missing data, authors will be contacted to avoid inappropriate descriptions of study results and to reduce the risk of bias. We will contact the corresponding author via email at the address listed in the published manuscript. If the corresponding author has not responded within 2 weeks, we will resend the email one more time. If all contact attempts fail, the author will be marked as unable to contact and no further attempts will be made. If the author responds after the final 2 weeks but before the final analysis of this review, the information obtained will be incorporated into the analysis.
Ethics and dissemination
Ethical approval is not required, as our review will include published and publicly accessible data. Findings from this review will be disseminated via publication in a peer-review journal.
Ethics statements
Patient consent for publication
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Twitter @famadarsareh
Contributors AR and FD were in charge of protocol design and manuscript conception. VM is in charge of determining study eligibility and reviewing collected data. The full text of papers and data collection are the responsibility of ET and NR. The authors also read the manuscript, provided significant revisions and approved the final version.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.