Abstract
Low-dose ionizing radiation exposure remains a major challenge for long-term health risk assessment, particularly in retrospective cohorts with heterogeneous exposure scenarios and limited biological material. Although next-generation sequencing (NGS) technologies dominate contemporary molecular research, DNA microarrays remain relevant in radiation biology due to their standardization, reproducibility, cost-effectiveness, and compatibility with archived biospecimens. This narrative review examines the contribution of microarray-based transcriptomic and epigenomic profiling to the study of low-dose radiation effects (≤100 mSv, millisievert), with emphasis on human observational studies, radiation epidemiology, and biodosimetric applications. The literature was identified through targeted searches in PubMed and Web of Science (2000-2025). Evidence from experimental models and exposed populations is synthesized to identify recurrent molecular pathways, major sources of variability, and challenges affecting reproducibility and cross-cohort comparability. Based on this evidence, a conceptual framework is proposed to define conditions under which microarray-based analyses remain interpretable and translationally informative. Machine learning approaches are discussed in a supportive role, with emphasis on interpretability and biological plausibility. Overall, DNA microarrays are positioned as a mature, niche technology that complements next-generation sequencing platforms and remains particularly suited for retrospective cohort studies and long-term molecular monitoring in radiation research.