Preventive care. Early disease detection. More efficient and superior care. Personalized care. Hospital-grade home care. FHIN creates a new reality.

FHIN started as a unique collaboration of 9 founding hospitals, dedicated to make data-driven healthcare a true reality for all interested stakeholders in the healthcare ecosystem. FHIN aims at providing technological tools for all partnering care providers to primarily align their data towards an unambiguous standard and secondly to facilitate multi-centric data-driven healthcare. We do this within a strong ethical framework, governed by a steering committee. FHIN guarantees that all its partners keep full control of their own data. 

The FHIN consortium is open to any project proposal from any stakeholders in healthcare that requires multi-centric high-quality data. Each project will be evaluated by the FHIN steering committee, where each FHIN partner has the freedom to decide upon participation for each proposed project. 

We believe in an open-source, open-science community and welcome everybody who's interested in true data-driven healthcare. 

About Us

FHIN is a pioneering alliance of non-profit healthcare organizations that are committed to the advancement of healthcare through data harmonization and standardization. Our mission is rooted in the belief that unified, high-quality data is the key to unlocking groundbreaking insights, enabling earlier disease detection, enhancing decision-making in disease treatment and true preventive healthcare. We are open to ALL non-profit healthcare organizations.

FHIN history

FHIN was initiated by a funding scheme from the Belgian government called "Data Capabilities". During spring 2023, 9 hospitals joined forces and submitted a proposal to this funding scheme. This proposal was rewarded with 1M€ and allowed FHIN to kick-off from 1 July 2023. The 1M€ was shared to foster true collaboration, in an atmosphere of open-science and equal partnership. This holds true for the current and future FHIN partners.

Our Approach

We focus on transforming disparate health data into a coherent, standardized format. Among the different standardized formats in the world, we chose to build on the open-source and open-science approach called the Observational Medical Outcomes Partnership (OMOP) Common Data Model from the open source community called Observational Health Data Sciences and Informatics (OHDSI). This framework not only facilitates data comparability and compatibility across our network but also enhances the reliability of health data analytics and adherence to a standard that is ready for international collaboration. OMOP was mainly chosen as the basic CDM because it fully adheres with our believe that data-driven healthcare should be fully transparent, via an approach that fosters intense collaboration, aiming at data-driven healthcare for all of us independent of your background. Further enhancements on the basic OMOP CDM allow more granularity, as well as centralizing other data forms such as images, scans, ECGs, EEGs, voices etc.

Our Technology

To support this data transformation, we provide cutting-edge technology solutions that make our harmonized data readily accessible. We ensure that these technologies are scalable, secure, and user-friendly, empowering a wide range of users—from healthcare providers and researchers to medical technology and pharmaceutical companies. 

Collaborate with Us

We believe in the power of collaboration to drive innovation. Therefore, we invite all stakeholders, including patients, academic institutions, medical technology firms, and pharmaceutical companies, amongst others, to propose projects that utilize our high-quality and truly harmonized data to improve patient care. Each project proposal is rigorously evaluated by our consortium governance to align with our core values and objectives: enhancing patient care and outcomes; for all of us.

Our Democratic Belief in Data

We champion the democratic nature of data-driven solutions. In our view, the basic requirements to access and analyze data—a computer and internet connectivity—are far more widely available than specialized medical equipment. This accessibility allows a broader range of contributors to participate in data-driven healthcare innovation, driving forward our collective goal of better health outcomes through data.

Visit our consortium to learn more about how we are making data work for healthcare and how you can be part of this exciting journey.

More context

Data-Driven Healthcare 

What is data?

Data is the interpretation of a physical quantity. There are two main categories of data:

• Machine generated data: sensors convert physical quantities (temperature, light, wind, radiation absorption, heart muscle activity...) in voltages or currents. These electrical quantities are subsequently transformed in human comprehensible units.

• Human generated data: humans either sense something (touch, sight, hearing, smell or taste) or get answers to questions. These observations are interpreted. The moment of registering these personal interpretations is when human generated data is born.

What is digital data in general?

Digital data are numbers. The reason is that digital devices like computers, tablets or mobile phones are machines that can only handle numbers, which they are extremely good at. Note that the number-handling capabilities of digital devices doubles nearly every two years (Moore's law). Hence, to make a computer work for us for any type of information, we need a dedicated table that converts unambiguous human-interpretable data into a unique numeric representation. And such tables have to be accepted as a standard by the community of interest. There are a number of internationally standardized coding tables to ensure mutual understanding between computers and humans, mainly for vision and sound. There are basically three categories of coding tables that are globally accepted in the digital world:

• Text: the ASCII table presents the basic characters that we daily use. It consists of 128 characters, numbered from 0 to 127. Additional text characters are defined in the Unicode standard coding tables. Whenever you type something in a digital system (e.g. an 'A'), the digital system is processing the corresponding number (which is 65 for the letter 'A'). 

• Image: colors are represented by numbers via so-called color spaces. An example is the sRGB representation.

• Sound: This is coded according the IEC 60958 standard.

What is digital data in healthcare?

There are two types of digital data in healthcare: unstructured and structured data. 

• Unstructured data consists of basic digital data. There is no dedicated coding table that allows a numeric representation of specific healthcare concepts. For example the concept 'fever' is unambiguously understood by humans but not by computers. In an unstructured manner, it is represented by a sequence of numbers as defined in the ASCII table (namely: [102, 101, 118, 101, 114]). Since there is no unique numerical representation for this concept, it is not possible for a computer to process the concept "fever". Note that textual data typically is unstructured data.

• Structured data consists of unambiguously defined healthcare concepts that have a numeric representation via an internationally accepted coding system.  There are however numerous coding systems in healthcare. For example, the 'Unified Medical Language System' (UMLS) consists of 189 underlying distinct coding systems. As a result, a large number of healthcare concepts have a numeric representation in multiple coding systems. This is an issue when aiming for data harmonization by using the same unique code for a given healthcare concept across multiple care providers. The solution is to standardize across the multiple coding systems. UMLS is such a hyper-standard, maintained by the NIH. A second hyper-standard is maintained by the open-source community 'Observational Health Data Science and Information' (OHDSI), based on 155 underlying coding systems. FHIN follows this OHDSI approach.

What is data-driven healthcare?

Data-driven healthcare is the use of computing power to visualize, analyze or learn from digital data, with the aim to improve patient outcomes and support decision-making in healthcare. The nature of digital data is that it is far more detailed then any other data capturing system that ever existed (compare music, photos and videos on your smartphone with music or video tapes, or photography film rolls from the past). This is the sole reason why healthcare driven by digital data is destined to be revolutionary.

What are the challenges with digital data in healthcare?

• Healthcare providers prefer unstructured data as it allows describing the subtle nuances that are specific for a patient. It is simply not possible to make a unique numerical representation for each subtle difference between patients. Additionally, unstructured data is often easier to register, for example with the aid of voice-to-text tool or predefined templates. However, computers can only work with structured data. This delicate balance between unstructured versus structured registration forms a first major challenge for care providers to step into data-driven healthcare. The balance between sturctured versus unstructured data is fully governed by the care providers themselves and is out of scope of the FHIN consortium.

• But even when care providers are registering in a fully structured manner, there remains a gap between the registered concepts and the concepts in the aforementioned hyper-standards. For example, the healthcare professionals in a hospital register in their local Electronic Health Record (EHR) software. Such EHR software often adopt a proprietary coding system which is not internationally accepted. As a consequence, the local coding systems in use by care providers needs to be mapped to an internationally accepted coding system. This forms the main focus of the FHIN consortium: is it possible to develop the necessary software tools and a way of working, that results in truly harmonized data across its partners to facilitate data-driven healthcare?