In October 2018, the German Ministry of Health (“Bundgesgesundheitsministerium”, short “BMG”) asked for blockchain use cases in the public health sector. Unibright contributed its proposal for “scaled blockchain networks for medical registers”.
Besides the interesting political and social relevance, this project shows why a unification-oriented solution like Unibright is so important when it comes to integrating existing systems (that were established over years and decades) into different blockchains.
With our proposal, we address two key issues in establishing medical registries: interoperability issues (heterogeneous system landscapes, isolated solutions that have grown over the years, interface issues) and unresolved details in rights and identity management (to motivate patients to participate in registry building studies).
We are presenting a reference architecture for scaled blockchain networks for merging and using medical registers.
Interoperability problems of medical registers
The landscape of clinical registers is heterogeneous, regulated differently by each federal state and barely networked across countries. Anyone who wants to get an idea of the incidence and prevalence of cancer in Germany, faces a problem: there is no single central cancer registry.
In April 2013, the Federal Ministry of Health (BMG) launched the “Krebsfrüherkennungs- und -registergesetz (KFRG)” — the “Cancer Prevention and Registration Act” to set up clinical cancer registries in the federal states. The aim of the law is to ensure systematic and consistent data collection in order to ensure valid statements on the occurrence, treatment and course of tumor diseases.
However, various opinions commissioned since then by the GKV-Spitzenverband (for example, https://www.gkv-spitzenverband.de/krankenversicherung/qualitaetssicherung_2/klinisches_krebsregister.jsp), have shown that the establishment and, in particular, restructuring of existing regional cancer registries is progressing only slowly. One of the reasons for this is the Federal Cancer Registry Act, which came into force many years earlier (1995), which gave the federal states a lot of room for maneuver, especially in the implementation of the registration process as well as the nationwide coverage.
Thus, the challenge of standardizing a centralized medical registry is not just the fundamental question of “how to build” such a registry. It is also to deal with the customized solutions and state-specific IT structures that have grown over the years, posing unresolved interoperability issues.
Unresolved rights and identity management
The secured protection of all personal data (to which all persons or institutions involved in the treatment process refer) is essential in order to motivate patients to participate in clinical trials which contribute to registries. The current structure of medical registers does not provide the patient with comprehensive and trustworthy rights and identity management.
The willingness of patients to use their data is what empowers a medical register in the first place.
The Unibright approach
The basic idea of the Unibright approach, is the development of a reference architecture for scaled blockchain networks, which serves as the basis for all other medically relevant registries in Germany. We refer to the example of the uniform cancer registries required by the federal government.
The application of blockchain technology requires an integrative approach, which deals with the different system landscapes of all types of registries of federal states.
So before any new technology can be used, structural requirements are needed that allow for the benefits of this technology. Blockchain itself can not replace a missing structure. Existing systems must first be logically combined with each other.
Therefore, the solution we propose consists of two steps:
- Integration of all data sources
- Scaled blockchain networks to ensure data integrity and legitimate usage
The following picture shows the overall concept, after that we will dive into details!
First step: Integration of all data sources
In order to fully benefit from the advantages of blockchain technology, you have to create conditions that make their implementation possible in the first place. This structural requirement is the first (unambiguously blockchain-related) approach to our concept.
The framework provided by Unibright is based on an integration platform that makes it possible to individually solve all interface problems at the system boundaries of the individual actors. Unibright’s “SmartAdapters” form the bridge to all IT systems and fit seamlessly into the process to be integrated.
The advantage: Existing systems do not have to be adapted, which would be extremely time-consuming and costly. The integration platform ties in with the existing IT landscape of the individual federal states and supplements them with the segments that previously lacked existing stand-alone solutions and thus did not meet parts of the eligibility criteria. Not to change existing IT landscapes, but to tie in the state of previously established registers, is a key point.
Only a global integration unit can complete processes there and complete them in favor of a uniform standard, where it has so far failed due to excessive structural differences. Depending on the degree of unification desired, it can address the different habits of data collection and maintain well-established data collection mechanisms to both maintain the local process while still ensuring the quality of all data.
The Unibright Framework is designed precisely for such a heterogeneous system landscape in order to ensure the interoperability of all IT systems and to facilitate the integrated exchange of unified data. Here we rely on our visual system components that frees the user as much as possible from implementation questions, and automatically generates the required technical objects (such as smart contracts, connectors and evaluation views).
Second step: Data integrity and legitimated use of the data by scaled blockchain networks
A medical registry is only as good as the quality of the data it manages. This data quality stands or falls with the willingness of patients to make their data available for general clinical use in the first place. Given the fact that the relevant data on which statistics, treatments, competence centers, health insurance funds, institutes and many other professions are ultimately personal patient data, data protection must be given the highest priority — data ownership is a legitimate requirement of patients or subjects who are heard in a consistent implementation using blockchain technology.
Our concept is based on the use of scaled blockchain networks. Here, private, semi-private and public blockchains form an overall architecture on which all surveys, storage, analysis and transfer of data and information take place.
Inner level — private blockchain
As an inner layer, a private blockchain is operated between the various participants in a register. By way of example, the various federal states can each build up a node of a private blockchain in order to operate a register in it. In terms of content, key-value pairs for each register are stored in individual blocks.
The filling of the register at the internal level takes place by connecting the classic systems in use to the data standard specified in the register. The classical systems do not have to be adapted or replaced for this purpose, but simply integrated.
Medium levels — semi-private blockchains for data access to the register for third parties
At the middle level, data from the inner register may be aggregated in other blockchain networks in any form to be made available to third parties. The form and degree of aggregation, as well as the underlying protocol (e.g., Hyperledger), can be freely chosen for each network. Participating nodes may be e.g. research institutes or clinics that want to query and evaluate data from the individual registers. For health insurance providers and other participants, seperate, medium levels can be defined as needed.
Technically, each middle level can define its own data standard, whose inventory is gained through aggregation, transformation, and integration of inner-level records. The addressee group, access rights and degree of encryption can also be set for each middle level, e.g. to be able to serve temporary research projects (from regional to European to worldwide).
On the respective middle levels, suitable customizable standard queries are defined by means of an own API, which can query and evaluate the available aggregates as efficiently as possible.
Outer level — public blockchain for identity management
Each participant of a register, be it a patient, clinic, evaluating research unit or health insurance provider, is identified by a digital identifier, which is stored in a (as widespread as possible) public blockchain. The corresponding key is referenced in the inner levels, so that, for example, each participating patient always recognizes in which aggregates and in which evaluation circles his study data has flowed.
In a further expansion stage, this identity management can be extended by a rights management in which, for example, a patient can give (and also revoke) the consent to a specific aggregation by means of a code in his identifier (which might be, modeled as a smart contract).
Why use blockchain over traditional alternatives?
We are convinced that a new technology hast to be successfully integrated into existing processes and system landscapes before it can leverage its own benefits.
Given the scope of the task, a blockchain-based solution can offer the following benefits:
- Proofability: Unlike a central, traditional database, a blockchain offers the “technology-based” advantage of being able to demonstrably pinpoint who provided what information and when.
- Counterfeit security: Also included in the design of Blockchain is the security against falsification of data. Under the keyword “trustless trust” one can understand a blockchain as an automatically manipulation-protected construct, in contrast to centralized, vulnerable data collection architectures.
- Secure and rights-based access channel: Also, the use of a (semi-private or private) blockchain offers the possibility to grant different parties clearly defined access to parts of the data (see “Views”). The blockchain itself represents a defined communication channel through which the various parties can easily and securely access data
- Pseudo-anonymity: If provided in the concept, the concept of pseudo-anonymity contained in many blockchain protocols offers the possibility to store data from different cases or instances pseudo-anonymously, which may be desired in the given context for evaluation purposes or research purposes.
The Beauty of Unibright
In our opinion, the very high technical requirements for interoperability and integration of existing systems and the associated future security are an elementary part of the solution we have proposed:
- There is freedom of choice in the blockchain protocols to be used at the internal level (private blockchain), middle level and public level (public blockchain). Unibright tools can create objects for Ethereum, NEM, Hyperledger, EOS and many more.
- The existing systems, which provide the actual register data, do not need to be adapted, but are integrated.
- By modeling at the process level, selected implementations (e.g., certain blockchain protocols) may also be exchanged retrospectively should e.g. in the coming months and years offer completely new approaches, which is to be assumed in a further young technology in further iteration stages.
The proposed approach of implementing various blockchain networks on different scales, interacting with each other, is a well known concept in software architecture.
Projects like Quant (with their “Overledger” approach) are also dealing with this idea. However, Unibright provides a complete solution: from a purely technical level to a concept and process level as well — providing a unified solution which includes cross-chain and off-chain integration, as well as future proofness.
Unibright offers a unified framework, bringing blockchain technology and smart contracts to mainstream usage. With its “no-coding-needed” approach, smart contracts get generated, deployed and updated automatically into different blockchains. Unibright works with visual, usecase-related templates and also automatically integrates existing IT systems into the blockchain.
Unibright Solutions, a dedicated consulting branch to support blockchain use in business processes, was additionally launched in December 2018.