Big data and digital health

Big data and machine learning in healthcare offer benefits which could reshape how health care is delivered along all stages of the patient journey.  Both private and public health sectors are investing in technologies that can unlock the potential of Big Data and analytics. Computational science can be used to improve patient health care and detect patterns in public health or map biomarkers in vulnerable populations.  Analytics can map surges in epidemics or track disease pathways on a global scale.

Cancer immunotherapy, for example, uses the patient’s own immune system to deter cancer cells from spreading.  Historically, only small subsets of patients benefit from immunotherapy. Machine learning can model pathways for immunogenicity across large populations. The data to facilitate this kind of predictive modelling is scattered across a large range of modalities – genetics, clinical symptoms, demographic patterns, immune response spectrums, and research on checkpoint inhibiters. Computational modelling, using big data sets can draw these strands together and assist in clinical decision-making.  

For all its potential to generate new knowledge and solutions for health care problems, Big Data also brings concerns over privacy, ethics and an ever widening pool of stakeholders. Media companies may hold data that can be used to model public health spending or the impact of contagion on urban populations but this data may be inaccessible to health researchers.  There are challenges to the evolution of the Big Data ecosystem mainly in confidentiality of individual patients, governance of data frameworks for sensitive health services such as HIV or mental illnesses, and contested ownership of large data sets derived from multiple domains, public and private.  Security of both data and devices used to gather and analyse health care data is highly problematic.  Technology alone will not solve the problems of Big Data in health care. It requires governments, companies, patients and a wide range of health care and technology providers to collaborate on setting standards and defining the boundaries of data sharing.

In 2017, an EU funded project (BigData@Heart) was set up to share clinical evidence, novel frameworks and other data about cardiovascular disease.  The intention of the project is to create an open access informatics platform for arterial fibrillation and general heart failure that can be made available to heart specialists around the world. 

Big data and computational modelling opens up new ways of creating knowledge in health care and disease management on a scale that has not been possible before.  As the technologies evolve, the benefits and the risks must also be explored by the wide range of stakeholders involved in the global health care sector.