A new study published in the journal npj digital medicine suggests that routine hospitalblood testscould be used to accurately predict injury severity and evenmortalityoutcomes in patients withspinal cord injuries. The study used machine learning to analyze millions of data points from standard bloodwork, offering a potentially affordable and accessible tool for improving early diagnosis and critical care decision-making. Spinal cord injuries affect more than 20 million people globally, with nearly a million new cases occurring each year. These injuries are difficult to assess in emergency settings, as early neurological examinations often rely on patient responsiveness, which can be compromised. This variability in presentation complicates both prognosis and treatment decisions during the early stages of care. To tackle this challenge, researchers led by Dr. Abel Torres Espín from Waterloo’s School of Public Health Sciences examined medical data from more than 2,600 patients in the United States. Over the first three weeks following injury, the team analyzed routine blood markers—such as electrolytes and immune cell levels—using machine learning models to identify hidden patterns that correlate with injury severity and patient outcomes. Significantly, the model’s predictions did not rely on traditional neurological assessments, making them useful even in cases where patient cooperation is limited. The findings reveal that the predictive accuracy of these models increases as more blood data is collected over time. Compared to traditional imaging or advanced biomarker analysis—which may not be readily available—routine bloodwork is low-cost, widely available, and can offer real-time clinical insights. “This foundational work can open new possibilities in clinical practice,” said Torres Espín, “allowing for better-informed decisions about treatment priorities and resource allocation in critical care settings for many physical injuries.” Reference:Mussavi Rizi, M., Fernández, D., Kramer, J.L.K. et al. Modeling trajectories of routine blood tests as dynamic biomarkers for outcome in spinal cord injury. npj Digit. Med. 8, 470 (2025). https://doi.org/10.1038/s41746-025-01782-0