Abstract
Traumatic brain injury (TBI) remains a significant global health burden and demands rapid, objective diagnostics that work outside imaging suites and intensive lab workflows. Here, we developed a copper sulfide (CuS) nanocube-assisted laser desorption/ionization mass spectrometry (LDI-MS) platform to acquire serum metabolic profiles (SMPs) with high sensitivity, low background in the low-m/z region, and excellent reproducibility for TBI diagnosis and severity grading. Using this platform, we profiled serum from individuals with TBI and healthy controls and achieved high-accuracy discrimination of TBI from Healthy Controls (HCs) (AUC = 0.999 in both training and independent test sets). The model further enabled reliable grading between mild and severe TBI. Feature selection yielded a panel of discriminative m/z signals that reflect systemic metabolic reprogramming after brain injury. Pathway analysis indicated perturbations in central carbon and amino-acid metabolism, consistent with altered energy production and nitrogen handling in TBI. In conclusion, CuS-assisted LDI-MS and SMP-based modeling provide a fast, reproducible, and mechanistically grounded route toward deployable serum tests for TBI diagnosis and staging.