Periodic Reporting for period 2 - HELICAL (HEalth data LInkage for ClinicAL benefit)
Reporting period: 2021-01-01 to 2023-06-30
The overall goal of HELICAL is to establish a European training network to prepare the next generation of Health Informatics Researchers. Advances in information science and artificial intelligence provide unprecedented opportunities for using these datasets to study the complex biology of these disorders, how this is influenced by environmental triggers, and to personalise their management. Currently, exploitation of these opportunities is severely limited by the shortage of researchers with the required informatics skills, knowledge of requisite data protection and stewardship principles, and understanding of the clinical context.
HELICAL addresses this unmet need by developing a trans-sectoral and interdisciplinary training programme that builds on the expertise and existing collaborations of its partners. It uses autoimmune vasculitis as a paradigm because it is scalable, the analysis of different forms will allow the influence of age and gender to be assessed, and comprehensive biological and clinical datasets are already available.
The HELICAL training programme focuses on three complementary objectives:
1. application of informatics to such datasets to gain new biological insights
2. translation of biological into practical clinical outputs
3. identification of the novel ethical constraints imposed on such studies and development of strategies to manage them.
The programme was delivered through a multidisciplinary and trans-sectoral partnership of world-leading researchers from Academia and Industry with expertise in biomedical research, epidemiology, statistics, machine learning, health data governance and ethics.
HELICAL addresses this unmet need by developing a trans-sectoral and interdisciplinary training programme that builds on the expertise and existing collaborations of its partners. It uses autoimmune vasculitis as a paradigm because it is scalable, the analysis of different forms will allow the influence of age and gender to be assessed, and comprehensive biological and clinical datasets are already available.
The HELICAL training programme focuses on three complementary objectives:
1. application of informatics to such datasets to gain new biological insights
2. translation of biological into practical clinical outputs
3. identification of the novel ethical constraints imposed on such studies and development of strategies to manage them.
The programme was delivered through a multidisciplinary and trans-sectoral partnership of world-leading researchers from Academia and Industry with expertise in biomedical research, epidemiology, statistics, machine learning, health data governance and ethics.
HELICAL began its work on 1/1/19, focusing initially on assembling an excellent cohort of 15 ESRs, aligned across four scientific work packages. It leveraged the large European vasculitis registry and biobank resources developed over the last decade to learn about the environmental impacts on both vasculitis onset and flare (WP1). By providing training in cutting edge computer science and artificial intelligence techniques across academic and industry sectors, we are moving towards predictive physician tools that can provide advance warning of flare. Further application of such advanced analytical approaches by the three leading European investigators in large vessel vasculitis also informed our knowledge of the genetic drivers of this disease, and highlighted novel therapeutic targets (WP2). In parallel, innovative proteomic, transcriptomic and histological image analysis techniques were used to identify novel biomarker ensembles to assist in management of systemic vasculitis (WP3). Underpinning these three work packages, and generating a new ethical framework for application of such data science techniques in a wide range of chronic diseases, we explored novel data governance approaches in the GDPR era and as we move into realm of the Data Governance Act and European Health Data Space (WP4).
The programme ended on 30/6/23, with all milestones and deliverables being met despite severe disruption during the Covid-19 pandemic. Outputs from the consortium include 29 publications (all open access via the Zenodo HELICAL community), with another 8-10 manuscripts in preparation. Open source software developed by consortium members has been made available on Github. All ESRs received extensive training across the four years of the action; six have been awarded a PhD and a further six expect to receive a PhD before the end of 2023. The key legacy of the HELICAL ITN is that all ESRs were highly competitive for follow on positions; some ESRs have stayed with their host institution as postdoctoral scientists, some have moved to postdoctoral posts elsewhere and some have moved out of academic science into industry, science communication or EU science policy.
The programme ended on 30/6/23, with all milestones and deliverables being met despite severe disruption during the Covid-19 pandemic. Outputs from the consortium include 29 publications (all open access via the Zenodo HELICAL community), with another 8-10 manuscripts in preparation. Open source software developed by consortium members has been made available on Github. All ESRs received extensive training across the four years of the action; six have been awarded a PhD and a further six expect to receive a PhD before the end of 2023. The key legacy of the HELICAL ITN is that all ESRs were highly competitive for follow on positions; some ESRs have stayed with their host institution as postdoctoral scientists, some have moved to postdoctoral posts elsewhere and some have moved out of academic science into industry, science communication or EU science policy.
HELICAL was first and foremost a training programme: the ESRs have been equipped with skills that put them to the forefront of an area of expertise that is critical to Europe’s continued success and prosperity, namely clinically meaningful application of informatics skills to link clinical and research data, to bring advances in data science and artificial intelligence to bear on the health of European citizens.
Examples of research progress beyond the state of the art include:
• Development of novel data science approaches for modelling the impact of weather and pollution on occurrence of systemic vasculitis (Kawasaki disease) in Japan
• A tantalising association between sulphur dioxide pollution and occurrence of vasculitis
• Innovative software that links an individual's location to prevalent weather and pollution conditions to facilitate exploration at scale of these conditions on occurrence of disease
• The first epigenetic profiles in specific white cell subsets from patients with vasculitis
• Linkage of genetic signals in the blood vessel wall of patients with vasculitis to disease pattern and severity
• Comprehensive optimisation of isolation of the tiny particles that break off from cells in the body, providing opportunities for generation of novel biomarkers for use in the clinic
• Development of a new artificial intelligence imaging approach to analysing kidney tissue
The action has had significant impact in the field of rare disease research in Europe as it has high visibility and traction in the vasculitis field. Stimulated by explicit links with the European Reference Network for rare immune disorders (ERN-RITA) and the European Vasculitis Society, several HELICAL initiatives have begun to highlight and ameliorate the challenges of innovation using health data from rare disease patients. For example, two EU projects have been built upon the outputs of HELICAL: the FAIRVASC project (https://fairvasc.eu/(opens in new window)) which seeks to create a semantic web platform that supports distributed learning from different regional registries across Europe and the PARADISE (Personalisation of relapse risk in autoimmunity) consortium which builds applies the autoantibody and transcriptomic signatures discovered in WP3 to help build a predictive algorithm to quantify relapse risk.
Examples of research progress beyond the state of the art include:
• Development of novel data science approaches for modelling the impact of weather and pollution on occurrence of systemic vasculitis (Kawasaki disease) in Japan
• A tantalising association between sulphur dioxide pollution and occurrence of vasculitis
• Innovative software that links an individual's location to prevalent weather and pollution conditions to facilitate exploration at scale of these conditions on occurrence of disease
• The first epigenetic profiles in specific white cell subsets from patients with vasculitis
• Linkage of genetic signals in the blood vessel wall of patients with vasculitis to disease pattern and severity
• Comprehensive optimisation of isolation of the tiny particles that break off from cells in the body, providing opportunities for generation of novel biomarkers for use in the clinic
• Development of a new artificial intelligence imaging approach to analysing kidney tissue
The action has had significant impact in the field of rare disease research in Europe as it has high visibility and traction in the vasculitis field. Stimulated by explicit links with the European Reference Network for rare immune disorders (ERN-RITA) and the European Vasculitis Society, several HELICAL initiatives have begun to highlight and ameliorate the challenges of innovation using health data from rare disease patients. For example, two EU projects have been built upon the outputs of HELICAL: the FAIRVASC project (https://fairvasc.eu/(opens in new window)) which seeks to create a semantic web platform that supports distributed learning from different regional registries across Europe and the PARADISE (Personalisation of relapse risk in autoimmunity) consortium which builds applies the autoantibody and transcriptomic signatures discovered in WP3 to help build a predictive algorithm to quantify relapse risk.