P.118 Near infrared spectroscopy based indices of cerebrovascular reactivity cluster with intracranial pressure based indices in moderate to severe TBI patients

Background: Cerebrovascular reactivity has been identified as an important contributor to secondary injury following moderate severe traumatic brain (TBI). “Gold-standard” intracranial pressure (ICP) based indies of cerebrovascular are limited by their invasive nature poor spatial resolution. Near infrared spectroscopy (NIRS) indices minimally and have improved In this study, classical machine-learning algorithms leveraged better characterize the relationship between these indices. Methods: High-resolution physiologic data was collected in a cohort adult TBI patients. From both ICP NIRS were derived. Utilizing Agglomerative Hierarchical Clustering (AHC) Principal Component Analysis, higher dimensional space examined. Results: A total 83 patients with 314,395 minutes unique complete obtained. Through AHC PCA there order clustering indices, separate from other parameters. Conclusions: relate one another space. may be viable alternative