New spinout CogStack is unlocking the power of healthcare data through AI to transform clinical care

An exciting new spinout company from King’s College London, UCL and a coalition of NHS Foundation Trusts (King’s College Hospital, Guy’s and St Thomas’, South London and Maudsley, and University College London Hospitals) will advance the digital transformation at NHS organisations and international healthcare providers by improving data quality and health interoperability for patient care, patient safety, population health and research.

CogStack’s innovative artificial intelligence (AI) technology generates deep contextualised insights into data from patient health records to support clinical decision making and healthcare research. The platform can search any structured or unstructured clinical data using natural language processing (NLP) and assign diagnostic or treatment codes and medical meaning based on particular words and sentences. Its models are self-supervised, trained on clinical texts from millions of real-world patients and fine-tuned by hundreds of thousands of clinician-supplied labels.

The implementation of CogStack across its partner NHS Foundation Trusts has created lasting change in physical and mental health care through measuring workforce productivity and automating data tasks such as clinical coding, clinical audit and disease registries. Since then, CogStack products have been adopted and evaluated by hospitals across London and in Birmingham, Leeds, Liverpool, Manchester, Southampton, Lancashire, and further afield in hospitals in Asia, Australia and Europe.

Healthcare providers using CogStack recouped 100% of their investment within two years, and the application of CogStack has achieved the following results: 

• Fast-tracking recruitment for clinical trials for the 100k Genome project, which needed patients with suspected rare diseases for sequencing. CogStack was cited in the 2017 annual report, ‘Generation Genome’(1), as a key enabler for accelerated success with recruitment.
• Enhanced data efficiency and savings of £1.2 m per year in 2018 for King’s College Hospital by detecting thousands of missing records of fracture clinic procedures in only 30 minutes.
• Accelerated medication reviews which saved more than two hours of work per pharmacy review at South London and Maudsley NHS Foundation Trust using CogStack AI to read and summarise records.
• Improving patient safety at University College London Hospitals NHS Foundation Trust by providing alerts for patients who are ‘lost-to-follow-up' in their healthcare journey, identifying 100’s of such events to be actioned in the gastroenterology clinic.
• Improving the accuracy of clinical coding for outpatient procedures and delivering £2.5 m in additional income to Guy’s and St Thomas’ NHS Foundation Trust in 2023.  

The award-winning(2) open-source suite of technologies was initially built in 2016, and since then, numerous research programmes have resulted in more than 100 peer-reviewed research publications reporting on methodological, platform, and clinical academic findings from a diverse range of specialties across physical and mental health. The National Institute for Health and Care Research (NIHR) Maudsley Biomedical Research Centre (BRC) and NIHR UCLH BRC supported the technology in its early development. 

Now, as an innovation venture, CogStack Ltd will take the platform beyond the founding organisations, to fulfil its mission of widespread dissemination for the benefit of the NHS and beyond. The CogStack Ltd founding team comprises Dr Tom Searle, Professor Richard Dobson and Professor James Teo.

Dr Tom Searle, CogStack CEO and CogStack Translational Research Director at the Institute of Psychiatry, Psychology & Neuroscience (IoPPN), King’s College London said:

“CogStack empowers organisations to use their disconnected, siloed data at scale, to understand where they can improve care, optimise operational processes and ultimately improve the NHS for patients and staff.”

Professor Richard Dobson, CogStack Chief Research Officer, Deputy Director of NIHR Maudsley BRC and Professor of Medical Bioinformatics at IoPPN, King’s College London said:

“CogStack has the potential to transform healthcare by making sense of NHS data. It will help to inform clinical guidelines across the UK and globally, and we now want to share this to improve the functioning of the NHS and, most importantly, outcomes for patients.”  

Professor James Teo, CogStack Chief Medical Officer and Professor of Neurology at King’s College Hospital and Guy’s and St Thomas’ NHS Foundation Trusts said:

“As a trusted, home-grown innovation from the NHS, CogStack is easy to deploy and empowers healthcare providers with medically-intelligent analytics at low cost. We are proud to demonstrate how our technology can benefit population health in the UK and internationally. We aim to expand Cogstack’s capabilities to deliver our solutions to more NHS trusts and new markets with future investments.”

Professor Graham Lord, Executive Director at King’s Health Partners; Senior Vice President, Health and Life Sciences at King’s College London; and Chief Academic Officer at Guy’s and St Thomas’ and King’s College Hospital NHS Foundation Trusts said:

“King’s Health Partners sits at the epicentre between some of the UK’s biggest, busiest NHS Trusts and a world-leading, research-driven university. We encourage innovative thinking and speeding up the implementation of tools that can make a difference to patients’ lives. NHS health records hold rich patient data, but they are often inconsistent and can be difficult to access by busy staff. After its early days as an initial collaboration between our Trusts and top academics to free up staff time and improve patient outcomes, I’m delighted to see CogStack growing further and benefiting the delivery of healthcare at home and worldwide.”

References:

1. Annual report of the Chief Medical Officer Genomics England 2016: Generation Genome.  https://assets.publishing.service.gov.uk/media/5a82b85fe5274a2e87dc2a4a/CMO_annual_report_generation_genome.pdf

2 Awards include: 2021 ‘Artificial Intelligence Health and Care Award’ winner from the NHS AI Lab; 2022 HETT Data Innovation award; 2023  CogX Award for ‘Outstanding Achievements and Research Contribution for Social Good’.

To ensure data security CogStack technologies are deployed within Hospital IT networks or Secure Data Environments, meaning all patient data stays safely and securely within the digital ‘walls’ of the Hospitals. CogStack AI models only contain representations of coded clinical information so will not ever release confidential information and has been in use in major NHS programmes for many years. Hospitals and patients can derive benefits from our pre-trained AI models safely without risk to identifiable data.