SiMMoN, Simplifying bedside multimodal neuromonitoring using decision support systems

Despite the breadth of knowledge acquired from multimodal neuromonitoring (MMM) data in traumatic brain injury (TBI), clinical trials using MMM are scarce and the clinical use of MMM is hindered by the complexity of complex data interpretation. To this effect, we lack simplified visualization tools that summarize complex trend data analysis and enable continuous data interpretation, which would allow MMM data to be seamlessly integrated into management strategies with the goal of improving patient outcomes. Here, we aim to develop and evaluate a decision support tool integrating visualization of MMM data (using spider plots) backed up by a patient and time-specific Neuro Deterioration Index (NDI) to identify patients at risk for secondary deterioration (e.g. high intracranial pressure (ICP), brain tissue hypoxia, etc.), otherwise known as secondary brain injury. The NDI will use monitoring phenotypes (unsupervised clustering) and machine learning algorithms (raw data, trend analysis, waveform analysis) to predict the burden of prolonged episodes of elevated ICP and brain tissue hypoxia. The spider plot and NDI will then be assessed for feasibility in a prospective observational pilot study and assessed for its predictive value. Translating MMM data into clinical use has a high potential to influence management strategies by early identification of patients at risk for secondary deterioration.