Rutgers University, in partnership with an international research consortium, has unveiled an artificial intelligence system capable of continuously monitoring and assessing the health of laboratory mice. The breakthrough, detailed in a 2026 publication in Lab Animal, addresses longstanding limitations in traditional veterinary protocols, which typically rely on daytime visual inspections that often miss subtle, early-stage health deterioration. Led by Dr. Jeetendra Eswaraka, Vice President of Universitywide Core Services at the Rutgers Office for Research, the collaborative team engineered a noninvasive tracking system that records locomotor activity as a digital biomarker around the clock. The collected data streams are processed through a large language model algorithm that identifies deviations indicative of impending illness. In preclinical trials, the system successfully flagged potential health complications three to five days before overt clinical symptoms emerged, offering a critical window for veterinary intervention. According to Dr. Eswaraka, the technology not only enhances animal welfare but also streamlines facility operations. Early detection capabilities have improved veterinary response efficiency by more than fifty percent, reducing manual labor while maximizing data-driven decision-making. Dr. Michael E. Zwick, Senior Vice President for Research at Rutgers, emphasized that the project exemplifies the institution's commitment to integrating advanced computational tools into biomedical research infrastructure. The system's deployment aligns with the recent reaccreditation of the Rutgers University Animal Care Program by AAALAC International, which validates the institution's adherence to rigorous ethical, regulatory, and technological standards in laboratory animal science. By transitioning from reactive, human-dependent checks to proactive, algorithm-driven monitoring, the Rutgers-led initiative establishes a new benchmark for preclinical animal monitoring. The technology is poised for broader adoption across research facilities seeking to optimize animal welfare, comply with evolving regulatory expectations, and enhance the reproducibility of biomedical studies. The international team continues to refine the algorithm's predictive accuracy while expanding its application to additional species, signaling a transformative shift toward autonomous, data-centric laboratory management.