A patient admitted to emergency at 11pm. Their oncology history is in the EHR of a hospital 40 kilometres away. Their recent lab results are in a community software system incompatible with the hospital's infrastructure. Their anticoagulant treatment isn't recorded in the admission file. The on-call physician makes a decision with 60% of the information they need. This scenario isn't a hypothesis: according to DREES, it represents the daily reality for one third of emergency admissions in French hospitals in 2023.
Health data interoperability — the ability of systems to exchange medical information in a fluid, secure and intelligible way — is the technical undertaking that would prevent these situations. And in 2026, it is no longer merely a strategic ideal: it is a regulatory obligation, a clinical emergency and a competitiveness lever for French hospitals.
Galeon has made it the backbone of its architecture since 2016. Today deployed in 19 hospitals including two major university hospital centres (CHUs), with more than 3 million structured patient records and over 10,000 healthcare professional users, Galeon demonstrates that real interoperability — sovereign, secure, GDPR-compliant — is not only possible but operational at scale.
This article analyses why 2026 is the pivotal year for French hospital CIOs, what the concrete obstacles are, and how the most advanced approaches — combining FHIR and blockchain — make it possible to build a genuinely interoperable hospital information system.
Interoperability in healthcare is not a new topic. Hospital CIOs have been discussing it since the early 2000s. What changed in 2026 is the convergence of three simultaneous pressures that make inaction impossible: regulatory pressure, clinical pressure and technological pressure.
The Ségur du Numérique en Santé programme, led by the French Digital Health Agency (ANS), has required FHIR standard adoption by all health software vendors funded under the scheme since 2021. By 2026, successive Ségur waves have covered virtually the entire portfolio of hospital applications: EHR, laboratory, imaging, pharmacy and secure messaging.
Hospitals that have not updated their solutions to comply with ANS interoperability standards now face a dual penalty: loss of eligibility for public funding and progressive non-compliance with the European regulatory framework. The European Health Data Space (EHDS), whose regulation was adopted by the European Parliament in 2024, further reinforces this trajectory with a 2027–2028 horizon.
FHIR interoperability is no longer optional for French hospitals — it is a condition of access to public funding.
The fragmentation of hospital information systems has a direct, measurable and documented clinical cost. The average French hospital HIS is composed of 50 to 200 distinct applications (ANS, 2023), a significant proportion of which do not communicate natively with one another.
The consequences include:
A non-interoperable HIS is not merely inefficient. It is clinically dangerous.
The technical debt of hospital information systems accumulates at a rate that outpaces teams' capacity to catch up. Every year without migrating to interoperable architectures adds another layer of bespoke connectors, conversion scripts and fragile middleware to the existing foundation.
Hospitals that begin their interoperability project in 2026 can still benefit from Ségur funding, a mature market offering and active ANS support. Those that wait until 2027 or 2028 will face exponentially higher migration costs, with no equivalent public support.
Health data interoperability refers to the ability of heterogeneous IT systems to exchange medical data, interpret it correctly and reuse it without loss of information or meaning. It breaks down into three levels.
The reference model distinguishes three progressive levels, each more demanding than the last:
Technical interoperability is the foundational level. It ensures that two systems can physically exchange data — that the connection exists, that file formats are compatible, that communication protocols are shared. This is the level at which most HL7 v2 integrations still in production in French hospitals operate.
Semantic interoperability is the critical level. It ensures that both systems interpret the data in the same way. A CIM-10 code for a diagnosis, a unit of measurement for a lab result, a drug name in INN rather than brand name. Without semantic interoperability, data arrives at its destination but may be misinterpreted.
Organisational interoperability is the rarest level. It ensures that the workflows of both institutions are compatible — that prescription, validation and sharing processes articulate correctly from one system to the other.
The majority of French hospitals have achieved partial technical interoperability. Very few have achieved complete semantic interoperability. Almost none have achieved organisational interoperability.
FHIR — Fast Healthcare Interoperability Resources — is the standard developed by HL7 International to address precisely the semantic interoperability challenge. Published in a stable version (R4) in 2019, it rests on a logic radically different from its predecessors.
FHIR treats every element of medical data — a patient, a medication, a clinical observation, a lab result — as a web resource, accessible via a standardised API. This means any FHIR-compatible system can query, receive and interpret these resources coherently, without bespoke development.
The concrete benefits for a hospital CIO include:
FHIR is the common language the global digital health ecosystem is adopting. Hospitals that don't speak it will gradually be left behind.
FHIR solves the problem of data structure and exchange format. But it does not answer a question every hospital CIO and CEO faces: who controls the data once it starts circulating?
A FHIR exchange network is technically efficient. But without a traceability layer, it remains opaque on several critical dimensions:
These are not details. They are the prerequisite conditions for an inter-hospital network to be politically acceptable, legally sound and clinically reliable.
Blockchain is a distributed and immutable ledger. Every operation on medical data — consultation, transfer, modification, use in an AI algorithm — is recorded in a timestamped block, cryptographically signed and replicated across the entire network.
Concretely, this provides three guarantees that no centralised architecture can simultaneously offer:
In an inter-hospital network, blockchain transforms trust from a political question into a technical guarantee.
Galeon has pushed this logic further than mere exchange security. With Blockchain Swarm Learning® (BSL®), the blockchain is not used to transfer medical data from one hospital to another. It is used to train AI algorithms on data that remains physically on each institution's servers.
The principle is as follows: AI algorithms travel from server to server, train locally on each hospital's data, and return only the model weights — never raw data. The blockchain traces each step of this training and distributes the value created in proportion to the data contributed by each institution, via the $GALEON token.
Data never leaves the hospital's servers. This is the founding principle of Galeon's Blockchain Swarm Learning®.
This model simultaneously achieves what no previous architecture had managed: training AI models on unprecedented data volumes (3 million records across 19 hospitals), guaranteeing each institution's sovereignty, ensuring GDPR compliance by design, and equitably sharing the value created.
Positioning interoperability as a priority is legitimate. Claiming its implementation is straightforward would be dishonest. Here are the real obstacles every hospital decision-maker must anticipate.
The coexistence of HL7 v2 and FHIR creates a complex transition period. Most systems in production in French hospitals still operate on HL7 v2. Migration to FHIR is inevitable, but it entails a period of coexistence between the two standards, with the hybrid interface maintenance costs that implies. There is no "smooth" migration that does not require dedicated human and financial resources.
Upstream data quality determines everything downstream. A performance standard as robust as FHIR cannot do anything with poorly entered, incomplete or heterogeneous data. Structuring data at the source — at the moment the clinician produces it — is a non-negotiable prerequisite. This is one of the most underestimated challenges in hospital digital transformation projects.
Resistance to change is often the first obstacle. Interoperability projects rarely fail for purely technical reasons. They fail because clinical teams were not involved from the outset, because the concrete benefits were not clearly explained, or because change management was treated as a residual budget item rather than a strategic investment.
Opening APIs increases the cyber attack surface. The more interoperable a HIS is, the more potential entry points it exposes to malicious actors. Interoperability must be accompanied by a strengthened security policy: strong authentication, end-to-end encryption, real-time flow monitoring. ANSSI recommends integrating security from the design phase, not as an afterthought.
Inter-institutional governance is a political challenge as much as a technical one. Getting multiple institutions to collaborate on a shared data network requires formal agreements on responsibilities, access rights, data use conditions and conflict resolution mechanisms. This level of governance takes time and requires high-level leadership — including CEOs, CIOs and medical directors.
Does achieving FHIR interoperability require replacing the entire HIS? No. FHIR is designed to integrate progressively with existing systems via adaptation layers called FHIR middleware. It is entirely possible to maintain HL7 v2 systems in production while deploying FHIR APIs for new exchange flows. Migration is carried out by successive functional domains — laboratory, prescriptions, clinical reports — without global operational disruption. A data flow mapping audit is the recommended starting point.
Does the Ségur du Numérique directly require hospitals to adopt FHIR? The Ségur requires compliance with ANS interoperability standards from health software vendors funded under the programme. For hospitals, this means that the solutions they deploy or renew must be compliant to be eligible for financial support. In practice, any renewal of an EHR, laboratory or imaging system will be subject to this FHIR compliance requirement.
How can you guarantee data sovereignty in an inter-hospital network? This is the central question that Galeon's Blockchain Swarm Learning® resolves. In this model, data remains physically on each hospital's servers. AI algorithms travel across the network, train locally on each dataset, and return only the model's statistical parameters — never raw data. The blockchain traces every step, guaranteeing full transparency and sovereignty.
What are realistic timelines for an interoperability project in a mid-sized hospital? A FHIR interoperability project for a mid-sized hospital (300 to 600 beds) typically spans 18 to 36 months, depending on the complexity of the existing HIS and the number of systems to connect. Typical phases are: audit and mapping (3–4 months), solution selection and contracting (3–6 months), deployment by functional domains (12–24 months), validation and scaling (6–12 months).
What funding is available for an interoperability project in 2026? Ségur du Numérique en Santé funding remains the main source of public support for interoperability projects. It covers investments related to bringing EHR, laboratory, imaging and secure messaging software into compliance. Amounts vary by wave and institution type. ANS regularly publishes eligibility standards on its official website. Regional funding and FMIS credits (Fonds de Modernisation des Investissements en Santé) complement this scheme.
Is medical blockchain compatible with hospital performance constraints? Yes, provided you distinguish between public blockchains (Bitcoin, Ethereum) and the private or consortium blockchains used in healthcare. The inter-hospital blockchain deployed by Galeon is a consortium blockchain, optimised for performance and confidentiality, with transaction times in the range of a few seconds — compatible with real-time care requirements. It is not comparable in terms of energy consumption or latency to a public blockchain.
Health data interoperability is the most structurally significant undertaking facing French hospitals in 2026 for three converging reasons: intensifying regulatory pressure from the Ségur du Numérique and the European EHDS, a clinical and economic cost of fragmentation that has become unsustainable, and the emergence of technologies — FHIR and blockchain — that finally make fluid, secure and sovereign exchanges possible. The FHIR standard establishes a common language the entire global digital health ecosystem is adopting. Blockchain adds the layer of trust, traceability and governance that FHIR alone cannot guarantee. The Galeon model — 19 partner hospitals, more than 3 million structured records, 10,000 healthcare professional users — demonstrates that real interoperability, sovereign and capable of powering cutting-edge medical AI, is operational today, not in ten years. Hospitals that launch this initiative in 2026 are building the infrastructure that will make all of tomorrow's medicine possible.
For further reading, take a look at this article on blockchain terminology in healthcare
Cour des comptes, Public Hospitals Facing Their Challenges - Report on Public Health Finances, 2022.
