Summary
Preeclampsia, a complex and potentially life-threatening hypertensive disorder of pregnancy, affects 5-10% of pregnancies worldwide. It remains a leading cause of maternal and perinatal mortality, largely because early and effective prediction remains a significant challenge. While traditional predictive models have limitations, the rise of Artificial Intelligence (AI) offers new promise. AI’s ability to process vast, diverse datasets (clinical, genetic, imaging, biomarker data) holds potential for improving early detection and risk stratification.
The Problem & Our Objective
While many individual AI studies exist, a comprehensive, high-level understanding of the evidence is lacking.
- How accurate are these AI algorithms compared to traditional methods?
- How effective are they at identifying high-risk women?
- What are the current clinical applications, and what are their limitations and ethical challenges?
To answer these questions, our research group is conducting an umbrella review, a systematic review of existing systematic reviews. This approach will allow us to synthesise the highest level of evidence available.
Our Method
This protocol outlines our rigorous approach, adhering to JBI and PRISMA guidelines. Our search strategy will cover 8 electronic databases (including PubMed, Embase, Cochrane Library, and Web of Science). All identified systematic reviews and meta-analyses will be assessed for methodological quality using the AMSTAR 2 checklist.
We will use a standardised data extraction form to capture key variables, including:
Types of AI algorithms used
Study populations and risk profiles
Reported accuracy metrics (Sensitivity, Specificity, AUC)
Reported ethical considerations
Authors’ conclusions on benefits and limitations
Why This Matters
This umbrella review will synthesise the current, broad evidence on the applications, benefits, and limitations of using AI in preeclampsia care. By evaluating the accuracy of AI algorithms and identifying knowledge gaps, this review seeks to inform future research and guide the safe, ethical, and effective integration of AI into maternal healthcare.
