Trisomy 21 (Down syndrome) remains the most prevalent autosomal aneuploidy, necessitating accurate prenatal diagnosis. While cell-free fetal DNA (cffDNA)-based non-invasive prenatal testing (NIPT) has transformed screening, challenges persist in low fetal fraction cases and confined placental mosaicism. Integrating artificial intelligence (AI), multi-omics, and fetal cell-based approaches represents a paradigm shift toward comprehensive prenatal diagnostics. This AI-assisted meta-analysis evaluates NIPT platforms for Trisomy 21 detection, comparing cffDNA-based, SNP-based, digital PCR, and emerging fetal cell-based approaches. Following PRISMA guidelines, we systematically searched PubMed, Web of Science, and Scopus (2010–2023). Ten high-quality studies were analyzed for diagnostic accuracy metrics across multiple NIPT platforms. MPS-based cffDNA NIPT demonstrated pooled sensitivity of 99.3% (95% CI: 98.7–99.7%) and specificity of 99.9% (95% CI: 99.8–100%). Alternative platforms, including fetal cell-based NIPT, showed promising but moderately lower accuracy. AI-driven risk stratification models demonstrated enhanced diagnostic precision in complex cases. MPS-based NIPT remains the gold standard for Trisomy 21 screening. Future integration of AI-enhanced fetal cell isolation and multi-omics profiling may expand diagnostic capabilities beyond aneuploidy detection.
Elmetwalli et al. (Thu,) studied this question.