Big Data for Medical Analytics

Project Information

BigMedilytics

Grant agreement ID: 780495

DOI 10.3030/780495

Project closed

EC signature date 21 November 2017

Start date 1 January 2018

End date 30 September 2021

Funded under

• INDUSTRIAL LEADERSHIP - Leadership in enabling and industrial technologies - Information and Communication Technologies (ICT)

Total cost € 16 949 408,93

EU contribution € 14 997 306,25

Coordinated by

PHILIPS ELECTRONICS NEDERLAND BV
Netherlands

Periodic Reporting for period 2 - BigMedilytics (Big Data for Medical Analytics)

Reporting period: 2019-09-01 to 2021-09-30

Summary of the context and overall objectives of the project

The healthcare sector currently accounts for 10% of the EU’s GDP. To improve productivity of the healthcare sector, it is necessary to reduce cost while maintaining or improving the quality of care. The most effective way to achieve this, is to use the knowledge that is hiding within the existing large amounts of medical data.
The BigMedilytics project used state-of-the-art Big Data technologies to improve the productivity of the Healthcare sector by at least 20%, by reducing cost to the patient, improving quality through better patient outcomes and delivering better access.
BigMedilytics aims to transform Europe’s Healthcare sector by executing pilots that are distributed over three themes: (1) Population Health and Chronic Disease Management, (2) Oncology and (3) Industrialization of Healthcare Services. The first two themes together cover 78% of all deaths in from non-communicable diseases in Europe. The third theme represents hospital operations and equipment cost, covering around 33% of the healthcare spending.
To scale the Big Data concepts across Europe, the project defined best practices considering (i) Big Data technologies, (ii) new business models and (iii) European and national healthcare data policies and regulations.
The overall objectives are:
• Improve chronic disease and cancer outcomes
• Optimize workflows through industrializing healthcare services
• Guarantee replicability of Big Data concepts for healthcare
• Increase market share through data integration
• Establish secure and privacy preserving cross-border and cross-organisation healthcare services thus strengthening EU’s Digital Market Strategy
• Define Best “Big Data” practices

Two of the key accomplishments of the project:
• A blueprint has been developed that describes how solutions in the Healthcare sector using Big Data can be rolled out at scale. The blueprint takes a holistic approach by considering not just aspects related to the Big Data technologies, but also aspects related to business modelling, policy, regulations, legal, privacy and ethics.
• Demonstration of how certain KPIs can be measured and improved in the three themes.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

1. WP 1: “Customization and Deployment of Core Technologies and Platforms” monitors the technological developments. It oversees the transfer of mature Big Data technologies into the use cases.
The outcome is a Blueprint that provides guidance on the key things to consider when rolling out Big Data solutions in the Healthcare sector. The guidance considers multiple aspects: clinical applications, technology, business models, health policy, legal, privacy and ethical issues. The guidance is provided for different stakeholders, e.g. for patients, hospital decision-makers, clinicians, data scientists, policy makers and legal and privacy officers.

2. WP 2: Big Data Solutions for Chronic Disease Management
The main objective is to assess both the potential impact of Big data technologies in improving the outcomes in health care as well as the reducing costs in different non-communicable diseases. WP2 aims to: improve the stratification of risk based on the electronic health records of a large population, propose advanced design of clinical interventions beyond the current state-of-the-art, improve the efficacy of health-care delivery, and scale up the big data technologies across the entire health care continuum. The project has also demonstrated the benefits that can be gained through the introduction of AI/IoT-powered remote patient monitoring solutions. Many of the solutions are especially important when considering the impact of COVID-19.

3. WP 3: Big Data Solutions for Oncology
The pilots demonstrate how the care of prostate, lung and breast cancer patients can be optimized. For example, decision making can be greatly optimized when utilizing open platforms that are able to integrate multiple modalities of data such as radiology, lab, clinical and even financial data. Clinical data can be combined with publicly available data to assist decision making. Advanced AI algorithms can be used to assist clinicians in identifying high-risk patients and deciding on the optimal treatment.

4. WP4: Industrializing Healthcare Services
Optimizing operational workflows positively impacts patient outcomes, reduces cost, and improves working conditions of care personnel. This WP demonstrates how various workflows can be optimized based on various data sources and technologies. For example, Real-Time Location System technologies are used to characterize the care pathways of stroke and sepsis patients. This information was used to identify bottlenecks and subsequently introduce interventions that help reduce the time to start treatment. The results obtained showed that conventional data sources are not accurate enough to improve such hyper-acute pathways. The same RTLS technology has been used to demonstrate how hospitals can reduce costs by optimizing the use of mobile assets and reduce the burden on staff. AI-driven solutions also demonstrate how radiologists can significantly reduce the time taken to search and analyse images thus not only improving their productivity but also allowing for more consistent quality across different groups of radiologists.

5. WP5: Business Impact
This WP shows how the use of big data applications can lead to an increase in performance. To measure this, we developed a Balanced Score Card and defined KPIs. In addition, this WP supports pilots to develop a business model that enables implementation and replication in other countries across Europe and across other diseases. Parallel to business modelling, we mapped the relevant European and National regulations for the collection, management and use of big data in healthcare.

6. WP6: Dissemination, Communication & Standardisation
This WP developed a communication plan and tools (e.g. project image, project website, newsletter, brochure and social media content), as well as a specific protocol for the generation of high-impact press releases and policies for external communication. The consortium has been contributing to the communication and presenting the project in different events. The project has also involved External Exploitation Partners in a workshop organized in Valencia in September 2019 and in the final BigMedilytics event in September 2021.

7. WP7: Project Management
The overall coordination of the project is performed by Philips. Special attention is paid to data protection, privacy, ethics, and data management taking GDPR into account.

8. WP8: Ethics
Deliverables were submitted, providing detailed information on how ethics is handles in the project. An ethics board was set-up, and an independent external ethics advisor was appointed.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

BigMedilytics focuses on how the latest Big Data technologies can be adapted and applied in novel ways in various healthcare applications. Some examples include:
• Stratification algorithms to help identify high risk patients and optimize patient care.
• Deep Learning algorithms can be used to automatically identify and label tumours.
• Real-time complex event detection enables better understanding of the situational context due to efficient integration of different real-time data and signals.
It demonstrates how to achieve an increase in healthcare productivity between 20% and 63% across 12 different pilots covering the most prevalent and expensive disease groups across Europe.

public-summary.jpg