<p>As an integral part of the world's transportation network, bridges are subject to a number of variables that might accelerate their deterioration, such as constant exposure to high levels of traffic, extreme weather, and seismic activity. Structural health monitoring (SHM) systems must be strong and dependable to protect them from this constant exposure, prolong their useful life, and allow for preventative maintenance. There are major problems with the scalability, cost-effectiveness, and real-time damage detection capabilities of traditional SHM techniques, which often depend on periodic eye inspections or conventional sensor networks. Focusing on the development and progress of vibration-based SHM for bridges, this research offers a thorough systematic evaluation of 39 peer-reviewed studies published between 2009 and 2025. As this review shows, there has been a big change in thinking about modal analysis methods, with more modern approaches using data-driven methodology that uses AI, IoT, and digital twin technologies. Important results indicate that low-cost sensing methods, such as MEMS accelerometers and smartphone-based systems, are gaining popularity, and that frequency monitoring in Operational Modal Analysis (OMA) is becoming increasingly accepted as a reliable method. But there are still problems with environmental and operational variability (EOC) hiding damage warnings and with finding scalable, reliable solutions for different kinds of bridges. The use of several AI/ML algorithms for damage classification, and the results of these integrated systems in real-life case studies, are all summarised in this paper. In order to encourage the subsequent generation of self-sufficient and robust bridge SHM systems, it finishes by pointing out important knowledge gaps and suggesting future paths, such as creating physics-based machine learning models, standardising protocols for wireless sensor networks, and incorporating crowdsourced monitoring projects.</p>