Abstract
Bartonella endocarditis is notoriously challenging to diagnose due to its insidious onset, nonspecific symptoms, and the limitations of conventional culture-based techniques. Artificial intelligence (AI)-driven metagenomic next-generation sequencing (mNGS) represents a transformative diagnostic approach by integrating machine learning algorithms with culture-independent sequencing data to improve accuracy and sensitivity. This technique overcomes barriers such as culture bottlenecks and post-surgical sample limitations, achieving high diagnostic specificity while enabling the detection of rare or novel pathogens. Despite challenges including limited reference databases, contamination risks, and cost-related barriers in low-resource settings, AI-enhanced metagenomics offers a promising path toward faster and more precise diagnosis of Bartonella endocarditis. Its integration into clinical workflows, supported by continuous algorithm development, cost optimization, and standardized protocols, has the potential to improve patient outcomes in rare infectious diseases.