Neurocritical care monitoring would greatly benefit from the establishment of a unified information architecture. The architecture would consist of a bedside environment where a seamless data pathway exists from the patient, to processing methods, to displays, to the patient record, to workflow software and decision support methods, and to a data warehouse for later use. The architecture would support medical devices that would plug-in and be automatically recognized as well as knowledge modules that would "plug-in" to provide new calculations, new displays, practice guidelines, decision support methods, and other bedside-usable information. The system would understand the context of the patient and customize the monitoring to specific problems. A unified information architecture would accelerate innovation since it would provide a simplified commercialization pathway for the various components of the system.

Such a system is futuristic, but it is useful to set our sights on the goal and consolidate efforts towards achieving it. Work is currently underway to design components of the architecture as well as to minimize some significant barriers involving the adoption of standards and mitigating safety and regulatory concerns. The Integrated Clinical Environment (ICE) is a proposed standard for information flow at the bedside which includes medical device "plug-and-play", the creation of a patient-centric database, and the use of the data by knowledge tools such as workflows. The concept is being developed by a consortium of industry members, academic organizations, professional societies, and regulatory agencies and is currently moving through the standards development process.

A key benefit of the information architecture is uncoupling the knowledge components from basic data collection devices together with the ability to distribute them as separate "knowledge plug-ins". At present, at least in the U.S., a medical device manufacturer must "re-clear" a device if a new metric is added to it that provides an additional use (e.g. a measure of autoregulation, a seizure detection method, etc.) and the costs and time involved oftentimes makes it not worth the effort. This results in many innovative techniques for data analysis and decision support sitting on the shelves of academia. Uncoupling these knowledge components from the device manufacturer would necessarily shift the verification and validation requirements to the developers of the components. Thus, a process for certifying their effectiveness and their conformance to a "plug-in" standard is needed. A valuable spin-off of the information architecture is a database of neurocritical care signals which, if annotated, can be used to test these plug-in components. Such a database would also be a valuable teaching resource.