NeuroIS knowledge discovery approach to prediction of traumatic brain injury survival rates: A semantic data analysis regression feasibility study

Abstract

The study of Neuro-IS often contains huge amounts of data. While the outcomes of this process are well documented, little has been written about the collection and dissemination of this data. In order to fill this gap, we looked at hospital ships which provide a medical asset in support of military operations. We collected data on three ship variables and four physiological body region injuries (head, torso, extremities and abrasions). We ran an exploratory regression analysis and found a significant relationship may exist (p < 0.000) for the overall model. In medical diagnosis, it is important to not only maximize correct classifications, but also to minimize Type I and Type II errors. We contend that predicting a patient that does not have TBI, will survive, when in fact the patient does have TBI, is a worse error than when a patient that has been diagnosed with TBI and in reality does not.