CogStack: assisting clinical decision-making with advanced text analytics

CogStack is an information retrieval and extraction platform developed by researchers at the NIHR Maudsley Biomedical Research Centre (BRC). It offers near real-time natural language processing of electronic health records, analytics and visualisation technologies to unlock the health record and assist in clinical decision making and research. CogStack is currently implementing these new data mining techniques with South London and Maudsley, King’s College Hospital, and University College London Hospitals NHS Foundation Trusts. 

Using CogStack to improve the management of diabetes in people with severe mental illness 

CogStack is being used in a variety of ways to improve and inform care. For example, the applied informatics research team are currently assessing whether CogStack can be used as an electronic clinical decision support system to optimise clinician-led management of diabetes and dysglycaemia in people with severe mental illness.

Contact details for Cogstack:

Compass: managing the emotional and physical side of living with a long-term condition

Compass is an evidence-based online programme to support people in managing both the emotional and physical side of living with a long-term condition. Compass is funded by King’s Health Partners, and was created in collaboration SPIKA Ltd. This project is led by Professor Rona Moss-Morris.

CONSULT: a decision-support system to help patients suffering from chronic conditions self-manage their treatments

CONSULT combines input from commercial wellness sensors and a patient's electronic health record and clinical guidelines. It interacts with these inputs using data science tools, and a branch of AI known as argumentation to draw inferences and generate insights. These insights are shared with the user directly, using a dashboard and interactive chatbot, to help patients make decisions on how to manage their care. CONSULT is currently being used by researchers to support stroke patients. 

Data provenance server

Data provenance is a form of contextual metadata that describes entities and processes involved in producing and delivering or otherwise influencing that resource. Provenance provides a critical foundation for assessing authenticity, enabling trust, and allowing reproducibility. Applied to healthcare, this typically involves attributes about the origin of health information at the time it is first created and tracks the uses and permutations of the health information over its lifecycle.

Provenance templates are a now established methodology for describing reusable, abstract patterns of data provenance, each representing a meaningful domain-specific action. The ability to manage and construct data provenance records using this template model is offered as a service-based interface to domain-specific software tools in order that they may routinely capture the provenance of their data and activities.

Medichec: minimising the risks from anticholinergic drugs

A free web-based app designed to review patient’s medication and minimise the risks from anticholinergic drugs. It is supported by the Centre for Translational Informatics (CTI), which encourages engagement with multiple key stakeholders including service users within their CTI Programme Board. Interested patients and members of the public are encouraged to get in touch.

Phenoflow: portable, workflow-based phenotype definitions

Phenoflow generates software that can be used to automatically extract patients that exhibit complex phenotypical characteristics from a population, for the purpose of research or conducting clinical trials.

Specifically, phenotype definitions authored using Phenoflow, or stored in the Phenoflow portal, can be downloaded and run against local data sources as executable workflows, which are coupled with collaboratively sourced implementation units. Different workflow permutations of the same phenotype definition support interoperability with a number of data source standards.

TRANSFoRm: clinical trial management toolkit

The clinical trial toolkit originated from the TRANSFoRm project, which demonstrated integration of research tasks in the routine workflows in primary care. The toolkit provides a common technology for semantic mediation between clinical and research data at the point of care while interacting, in real time, with the workflow of a consultation, via the electronic health record. It can be used to recruit patients into available, machine-readable, clinical trials and manage their follow-up.

TRANSFoRm has recently completed the management of The Runny Ear STudy, a clinical trial testing whether giving an antibiotic ear drop or a delayed antibiotic by mouth is as good as immediately giving antibiotics by mouth for children who have developed acute otitis media with discharge.