<![CDATA[The increased physical requirements of top-level volleyball players, such as the numerous high-intensity jumps and the quick eccentric landings, have contributed to the rise of overuse injuries and made athlete-monitoring technologies more important. Digital biomarkers extracted from wearable biosensors combined with artificial intelligence (AI) provide a scalable method to measure internal and external load, describe fatigue, and predict injury even before the appearance of symptoms. This systematic literature review provides an overview of the use of wearable biosensing, machine learning, and injury prediction in volleyball, including the common grounds between sports-technology and sports-medicine research fields. Following PRISMA 2020 guidelines, a search of the Scopus database returned 386 records, which were further reduced to 10 after a series of eligibility assessments and exclusions. The results were grouped into three main categories: performance and load monitoring, injury prediction, and biosensing and digital biomarkers. Research shows that inertial measurement units (IMUs) are the most widely used instruments in volleyball. They allow for automated jump detection and jump-load quantification using deep learning techniques such as temporal convolutional networks. Besides, personalized machine-learning models give better results than group-level models for monitoring overuse injuries. Newly developed textile and biochemical biosensors can also detect physiological biomarkers like lactate, extending monitoring beyond mere kinematics. The methods used are highly varied and mainly consisting of supervised learning on small, sport-specific cohort. There is very little external validation and football-specific injury endpoints are scarce. A brief theoretical detour in the review interprets digital biomarkers as a unified concept that combines both biomechanical and physiological monitoring. In a more practical sense, the review provides coaches and clinicians with a well-structured description of sensor placement, modeling strategies, and levels of validation maturity. Future studies should encourage synergistic sensor use, prospective volleyball-specific injury cohorts, model interpretability and standardization of reporting in order to make predictive analytics the basis of trustworthy injury-prevention decisions.]]>
