Systematic knowledge discovery for all diseases - for clinicians and patients
Underpinned by the needs of patients, clinicians and researchers, the Disease Atlas is an ambitious project involving the generation of systematic, data-driven knowledge across all common and rare diseases. Using newly available nationwide data on 56 million people the Atlas is generating novel comparative insights of the health needs of patients, the care provided, and the research that is carried out (Figure). We believe that the Disease Atlas may change the way we think about and research diseases, inform policy and practice and unlock new ways to improve the health of patients and communities.
The Atlas is based on learning from linked, longitudinal and nationwide data made available to support the COVID-19 pandemic response. A dominant paradigm in health research is to study one disease, or one group of diseases at a time. Moreover diseases which are considered ‘rare’ are usually researched separately from those considered common. Patients and clinicians recognize two limitations of this paradigm: first it misses out on the experience of people (patients), who commonly have multiple diseases, and may take multiple medications; and second, it is not systematic, and many diseases and populations are relatively neglected in research.
The Disease Atlas paradigm is different, because it is systematic: we leave no person, and no disease out. We seek to define and compare all diseases within a data driven framework, irrespective of research vogue, or current funding initiatives. Despite accumulating vast amounts of data, healthcare organisations know remarkably little about which patients have which diseases, in which combinations, and with what outcomes. Nor is it understood how these markers of health need relate to the use of medicines or other interventions. The Atlas seeks to answer such questions, systematically, seeking to discover novel associations between unmet need, care and different types of research.
The Atlas has generated insights from patient data. The British Heart Foundation Data Science Centre and the NHS Digital National Trusted Research Environment have pioneered safe and secure research access to the nation’s health data. The scale (56 million people in England) and the quality of the NHS health data have few if any parallels internationally. We have focused on the International Classification of Disease (10th revision) to define diseases in the Atlas, because it is the only clinical terminology which is widely used around the world. The Disease Atlas seeks to accelerate understanding of the course and interaction of all the common and rare diseases captured in these health data, and for patients across all demographics. In addition, the Atlas has generated insights, for each disease, from public data. Here, we have used terminologies to ‘link’ diseases observed in patient data, to research publications, registered trials and genetic studies.
For more information: please contact Natalie Fitzpatrick email@example.com - Disease Atlas Programme Director
Join the team
If you are a clinician who diagnoses, manages or researches one or more of the diseases shown in the figure and would like to become an ‘Atlas clinician’ of disease, contact firstname.lastname@example.org
Meet the team
The idea and ambitions for the Atlas arose during the pandemic out of discussions between Harry Hemingway and Spiros Denaxas (founders). For the first time in England, and arguably anywhere in the world, research was suddenly possible at a scale of 56 million people -this was due to the BHF Data Science Centre. Prior to this, working with Ami Banerjee and Alvina Lai, we developed an interest in understanding which underlying diseases might make patients particularly vulnerable to COVID-19. We published influential first wave findings on multiple conditions, and on cancers but had limited access to data, and on only small subsets of the nation. Together Professors Hemingway and Denaxas built a coalition of the willing among an interdisciplinary group of researchers at the Institute of Health Informatics. The team named here is ‘self-assembly’ each member has substantive responsibilities at their employing university (UCL) or hospital (UCLH) and have contributed to the Atlas only in a part time ‘above and beyond’ capacity. What brought people together was the opportunity to carry out a new kind of research in nationwide health data for nationwide public good.
After generating the initial body of knowledge and the website (i.e. the Minimum Viable Product), we have more recently been privileged to receive initial funding support from the British Heart Foundation Data Science Centre, the NIHR Biomedical Research Centre at University College London Hospitals and Health Data Research UK (PI’s Hemingway and Denaxas). Only in the last few weeks (September 2023) have we appointed the first full time member of the Atlas team!
From the outset the Atlas has been developed by clinicians, for clinicians, across multiple specialties. Contributors at IHI include Ami Banerjee, Anoop Shah, Laura Shallcross, Rob Aldridge, Rishi Gupta, Tom Lumbers, Mohamed Mohamed, Rui Bebiano Costa and Isobel Braithewaite. Beyond IHI Atlas clinicians include Al O’Brien, Dominic Crocombe, Neil Sebire and Bryan Williams, Riyaz Patel, Aroon Hingorani, Geoff Hall, Mark Lawler and Cathie Sudlow. Harry Hemingway leads the clinician engagement.
From the outset the Atlas has been developed to inform and benefit patients. Working with patients, carers, charities and other patient-facing organisations, we are leading a programme of work to embed patients in the development of the Atlas, by ensuring the Atlas reflects patients’ priorities and needs and has been tested by patients.The Disease Atlas would like to thank the patients and members of the public whose NHS data collected as part of their care and support has allowed the creation of the Atlas. Natalie Fitzpatrick leads patient and public involvement and engagement.
- Informatics, Knowledge Graph and Ontologies
Spiros Denaxas (Lead)
- Computational Pipelines: Patient Data in Trusted Research Environments
Ana Torralbo and Spiros Denaxas (Leads)
Muhammad Qummer Ul Arfeen
- Data Visualizations
Ana Torralbo (Lead)
Johan Hilge Thygesen
- Public Data, Clinical Curation
Jayathri Wijayarathne (Lead)
- Public Data, Computational Curation and Mappings via API
Cai Ytsma (Lead)
Johan Hilge Thygesen
- Phenotyping, Large Language Models and Natural Language Processing
Chris Tomlinson (Lead)
- Front End Design and Development
Spiros Denaxas (Lead)
- Management of the Enterprise
We would also like to acknowledge all data providers who make health-relevant data available for research.