Perspective on AI-Driven High-Risk Pregnancy Prediction

A perspective article examining the benefits and risks of AI-based obstetric risk prediction over the past five years. The author, Qingqing Ji, integrates evidence on AI prediction, prenatal screening, and principles of trustworthy digital health. The document is a 18.0 KB DOCX file.

Use Cases

• Analyze the discussion of 'probabilistic outputs' and 'predictive anxiety' for research on AI risk communication.
• Extract arguments on 'biased data and labels' and 'socially disadvantaged groups' for fairness audits in clinical AI.
• Synthesize the proposed 'integrated safeguards' like 'transparent model documentation' and 'continuous fairness monitoring' for policy guideline development.
• Study the 'lessons from prenatal screening and high-risk labeling' for historical context in digital health ethics reviews.

Strengths

• Focuses on a critical and timely topic in digital health, integrating evidence from the past five years.
• Proposes concrete safeguards like 'structured and patient-centered risk communication' and 'tiered psychological support'.
• Released under a permissive CC BY 4.0 license for broad reuse.

Limitations

• The 18.0 KB size indicates it is a short perspective article, not a primary research dataset.
• Content is a narrative synthesis and argument, not structured quantitative data for direct analysis.
• No sample data or defined columns are provided for computational use cases.

Provenance

Source

figshare, author Qingqing Ji.

Collection Method

Targeted, non-systematic narrative synthesis.

Time Range

Evidence reviewed spans the past five years.

Freshness

Last updated March 26, 2026.

Geography

null