CHART-ADAPT is a two year, Innovate UK funded, research project which brings together a multi-disciplinary team consisting of healthcare technology providers, clinicians and a major NHS provider. They are working together to develop a high performance computing platform which will enable important physiological models and analyses to be implemented at the bedside for the first time within an NHS environment.
Enabling development and deployment of clinical models and algorithms
The novel platform will empower specialist care providers and enables the analysis of patient data (both high and low frequency) to create new and novel closed loop diagnostic or therapeutic models/algorithms relevant to patient treatment. Additionally, the platform enables the assessment of these newly developed models/algorithms or existing models directly in the clinical environment. This will help provide the basis for better treatment and more cost-effective and sustainable healthcare by closing the loop between clinical research and practice.
Fast results at the patient bedside
A key component of the platform is the use of high performance computing to provide the required processing power to enable complex clinical algorithms to process large volumes of patient data in clinically meaningful timescales; results can be delivered within seconds if required.
Automated de-identification tools
As part of the platform, a highly configurable de-identification service is provided to allow the removal of sensitive information from live streaming of patient data. The service also enables the automated integration of analysis results with the relevant sensitive data.
Live implementation in critical care
The platform will be demonstrated in the Neurointensive Care domain. Head Injury is a devastating injury not only to the victim but also to their carers and to the society that supports their recovery, which is often long term. Unlike other forms of pathology including cancer, stroke or cardiovascular disease, there have been few recently proven effective therapies for brain injury. What is needed is a step-change in approach, one that brings recent advances in big data modelling directly into clinical practice, allows agile development, testing of new interpretations of high-frequency data for improved detection and prediction of clinically relevant and treatable events that occur during their early management in intensive care.
Enabling the analysis of high-frequency data and the implementation of clinically important physiological models will immediately deliver previously unavailable evidence-driven patient care. Hospital treatment of head injury is expensive and the loss of employment to the victim and the stress and increased burden of care to family members has significant social and economic effects. The smallest improvement in the treatment of head injured patients has the potential to generate a huge impact on the social and economic effects of head injury.