Congenital heart defects (CHDs) represent a significant burden of neonatal morbidity and mortality worldwide. Maternal biomarkers related to homocysteine and methionine metabolism have emerged as potential indicators of CHD risk during pregnancy. This review examines the current literature surrounding maternal biomarkers such as homocysteine, methionine, and related metabolites in the context of CHD development. Various studies investigating the association between abnormal levels of these biomarkers and the risk of CHDs are critically evaluated. Additionally, the underlying mechanisms linking aberrant homocysteine and methionine metabolism to CHD pathogenesis are discussed. Understanding the role of maternal biomarkers in CHD risk assessment holds promise for early detection, intervention, and prevention strategies to mitigate the impact of CHDs on neonatal health.

The transition from monolithic architectures to decentralized cloud-native systems represents a fundamental shift in software engineering, characterized by the pursuit of extreme scalability and high availability. This research presents a comprehensive investigation into the mechanisms governing microservice building, serverless adoption, and the modularization of legacy systems. Drawing upon evidence-based software engineering principles and grounded theory, this article synthesizes the current state of high-availability frameworks, structural coupling metrics, and the emerging influence of machine learning in service boundary detection. We explore the evolutionary path from the "big ball of mud" monolith to a dataflow-driven microservices ecosystem, emphasizing the critical role of Conway’s Law in organizational alignment. Furthermore, the study provides a deep conceptualization of technical debt in serverless computing (Function-as-a-Service) and the architectural motivations behind Micro-Frontends. Central to this discourse is the reconciliation of structural coupling with quality attributes defined by ISO/IEC 25010, ensuring that the decomposition process enhances testability and maintainability. By evaluating circuit breaker patterns and the nuances distinguishing Service-Oriented Architecture (SOA) from microservices, this work establishes a publication-ready taxonomy for modern software architects. The results highlight that while machine learning significantly optimizes boundary detection, the human-centric aspects of refactoring and qualitative requirements remain indispensable.