Artificial Intelligence and Machine Learning Applications in Genomic Analysis of Cervical Cancer: A Systematic Review (Preprint)

BACKGROUND Artificial intelligence (AI) and machine learning (ML) technologies have emerged as powerful tools for genomic data analysis in cancer research. Their application in cervical cancer genomics holds promise for improved diagnostic accuracy, prognostic prediction, and therapeutic target identification. OBJECTIVE To systematically review the current evidence on AI and ML applications in genomic analysis of cervical cancer, evaluating their performance, clinical utility, and limitations. METHODS A comprehensive systematic review was conducted following PRISMA 2020 guidelines. Multiple databases were searched from January 2015 to December 2024. Studies were included if they applied AI/ML methods to genomic data analysis in cervical cancer. Data extraction focused on AI/ML algorithms, genomic data types, performance metrics, and clinical applications. RESULTS The systematic search identified 3,247 records across multiple databases. After removing duplicates (n=1,091), 2,156 records underwent title and abstract screening. Following full-text review of 287 articles, 52 studies met inclusion criteria and were included in the qualitative synthesis, with 31 studies suitable for quantitative analysis. Inter-rater agreement was substantial (κ=0.84 for abstract screening, κ=0.91 for full-text review). CONCLUSIONS AI and ML applications in cervical cancer genomics show promising results for biomarker discovery, diagnostic enhancement, and prognostic prediction, though standardization and validation challenges remain.