HDPHealth Data Portal


Prof. Edwin van den Heuvel
Groene Loper 3, 5612 AE Eindhoven
Catharina Hospital (CH), the Maxima Medical Center (MMC), Kempenhaeghe Epilepsy and Sleep Center (KH), Eindhoven University of Technology (TU/e) and Royal Philips Eindhoven (RPE)

The Health Data Portal (HDP) is a unique scalable collaboration platform that builds on existing initiatives to provide an infrastructure where researchers can bring together and work safely with medical data. The portal serves commercial partners, medical institutes and academia and enables a fast track to high-tech health innovations using engineering, data science, AI.
The design of the HDP has set the highest priority to GDPR-compliancy without compromising research requirements, a.o. by using a trusted organization as independent third party to process and handle data requests.
Making large datasets of medical information available between the members enables scientists to analyze data, discover and develop new hypotheses about human health. The gradual buildup of rich data sets enables new AI-applications for machine learning to revolutionize healthcare.
The HDP has the potential and ambition to become a national platform that can lead to faster life sciences & health innovation.

The HDP is a national collaboration platform that builds on existing initiatives and standards to provide an infrastructure where researchers can bring together and work safely with medical data.

The HDP presents an integrated work environment for researchers. They can request access to the ever-growing collection of data in the portal through a standardized workflow that by design enforces compliancy with ethical, medical and privacy regulations. For clinical and observational studies, virtual research workspaces are created in which data is safely shielded against unauthorized distribution, yet allowing researchers to efficiently work together, make use of scalable computational resources and a rich suite of software packages to analyze data.

The HDP environment consists of local infrastructures (in each institution), an independent entity for requesting and handling of sensitive medical data from the local infrastructures (the Trusted Third Party, TTP) and a central cloud-based system (the HDP) that allows researchers to access and analyze data.

The nature of medical data can be diverse and may include electronic health records (e.g. demographic information, diagnosis, treatment, prescription drugs, laboratory tests, physiologic monitoring data), but also administrative and workflow data, disease registries, surveys and clinical trials data. Homologation of the data is an important prerequisite for successful use of the HDP. It is a scalable infrastructure that would allow growth towards all kinds of other data sources.

The local infrastructure can be used to collect and de-identify data feeds and forward them through the TTP to workspaces that are part of the central HDP, where the data can be analyzed and used for research activities.

The TTP is an essential governing platform that keeps up a metadata catalogue of medical information that itself is distributed over the institutes, and maintains control of the data exchange for approved medical studies. The TTP is legally and technically separated from the workspace environment, and is set up to maintain a uniform judicial and governance structure for the processing of data and to enable effective data protection through GDPR compliance.

The HDP will offer virtual workspaces to perform data analytics research, thereby enabling a workflow in a confined environment and eliminating the need to expose patient data to uncontrolled environments. The HDP management takes care of data authorizations and regulatory audit logging.

Research facilitated by the HDP
Big data in the health domain offers opportunities to build better health profiles and better predictive models around patients, so that we can better diagnose and treat diseases. The gradual buildup of rich (meta)data sets in the HDP enables new AI-applications to revolutionize healthcare in terms of patient outcome, performance and cost.
With more data on treatments, on DNA, proteins, tissues, cells, metabolites, we can collect ever more information on multiple scales for what constitutes a disease. The use of data can help us improve our models, e.g. in statistical modeling or machine learning, so that they become more predictive, better understanding and lead to new insights.
New medical devices and research grade wearables that are being developed a.o. within the e/MTIC consortium offer sources of data that provide a way to longitudinally monitor the state of a person in many dimensions, so that researchers can develop algorithms that learn to detect deviations from a baseline to predict diseases state.
Institutions usually collect data to meet an operational need rather than to power analytics. As a result, data is fragmented across many institutions and data capture systems. The distribution of patient data over multiple institutions in itself is a pertinent problem; with the HDP we provides secure and practical solutions for medical studies that require combining datasets from multiple institutions & sources.

Aansluiting bij strategische ontwikkelingen
Life Sciences & Health
Gezondheidszorgonderzoek, preventie en behandeling
Meten en detecteren: altijd, alles en overal