Home Personal Health Train IN German Chapter Workshop Report

On 12 February 2019 the GO FAIR Personal Health Train (PHT) Implementation Network (IN) German Chapter workshop took place in Berlin.

Organization CommitteeOya Beyan, Oliver Kohlbacher, Stefan Decker, Matthias Löbe, Toralf Kirsten, Holger Stenzhorn, Anja Busch
ChairsOya Beyan, Oliver Kohlbacher, Matthias Löbe, Lukas Zimmermann
Related linksWorkshop agenda (link to pdf-file)

Objectives

The main objective of the workshop was to introduce the Personal Health Train (PHT) approach for improving the reuse and analysis of sensitive data in distributed settings, and to discuss specific needs and use cases as well as strategies to achieve sustainable solutions on a national level.

The workshop had been designed to serve as a platform to bring communities together including but not limited to Medical Informatics Initiative projects, NFDI (National Research Data Infrastructure) networks, research data infrastructures, European networks, and projects.

The workshop consisted of four sessions. The first session was dedicated to introducing the GO FAIR PHT Implementation Network and the existing PHT concepts, technologies, and reference implementations. In the next session, different communities such as rare diseases, ELIXIR, and FAIR4Health platform presented their perspectives. The third session was designed as a collaborative discussion platform, where participants teamed up in three working groups, namely strategic vision, technical infrastructure, and use cases. The last session consisted of presentations of working groups and discussing a joint vision.

With this workshop, the GO FAIR International Support and Coordination Office established a German PHT chapter. The German chapter will work together with the International PHT implementation network. The office will support all activities that enable the building of a PHT implementation network in Germany. Stakeholders from Germany are invited to actively join the network and collaboratively work on future solutions for the PHT German chapter.

Content

Chairs opened the workshop, welcomed the participants and addressed the agenda.

Session 1: Introduction of the GO FAIR PHT Implementation Network

Roman Siddiqui on behalf of TMF e.V., Berlin welcomed the participants and presented the goals, membership landscape and working groups of the TMF. The role of the TMF in standardization and quality control was emphasized.

Welcome_GO FAIR_TMF

 

Anja Busch represented the German GO FAIR International Support & Coordination office on behalf of Klaus Tochtermann. She presented the objectives and organizational structure of the GO FAIR initiative, and introduced the GO BUILD, GO CHANGE, and GO TRAIN tracks and rules of engagement to implementation networks. She also informed participants on the available support.

PHT-IN_GO FAIR presentation_Anja Busch

 

Oya Beyan from the Fraunhofer Institute for Applied Information Technology presented the GO FAIR Personal Health Train implementation network. Guiding principles of the PHT and its innovative approach for data sharing were presented. Main concepts, components, and existing implementations were introduced. The German Chapter of the PHT train had been discussed.

GO FAIR PHT IN German Chapter

 

Oliver Kohlbacher from the Centre for Bioinformatics Tübingen (ZBIT), University of Tübingen introduced the current prototype implementation of the PHT within the German Medical Informatics Initiative and demonstrated an application to the demonstrator studies based on the National Core Dataset of the Medical Informatics Initiative.

2019_PHT_IN_2019-02-12_Oliver Kohlbacher

 

Session 2: Perspectives from Communities

Marco Roos from the Human Genetics department of Leiden University Medical Centre (LUMC) reported on the GO FAIR Rare Disease community and transition of FAIR principles. Infrastructure for rare diseases such as the European Reference Networks, and European Joint Program on Rare Diseases (EJP RD) had been introduced and challenges to use data for efficient care and research at the global level had been discussed.
EJP RD FAIR based virtual platform and a vision of connecting EJP-RD FAIR data points to PHT had been explored.

Von Bahnhöfer und Hauptbahnhöfer für seldsame krankheiten – PHT German Chapter 2019_Marco Roos

 

Jen Harrow from ELIXIR, Wellcome Genome Campus introduced ELIXIR and provided an overview of the ELIXIR platforms for data resources, interoperability, tools, computing, and training. ELIXIR infrastructure and services for a federation of human genome data and sharing them across borders had been addressed. The IMI FAIRplus project and its Capability Maturity Model Integration approach had been introduced.

ELIXIR_PHT_Jen Harrow_12Feb19v2

 

Matthias Löbe from theInstitute for Medical Informatics, Statistics and Epidemiology (IMISE), Leipzig University introduced the FAIR4Health project which focuses on the FAIRification of health data and demonstration of the benefits of its sharing and re-use. FAIR4health approach for the FAIRification of local data sets and the platform for delivering data-driven services had been discussed.

2019-02-12 FAIR4Health – GO FAIR PHT Berlin_Matthias Löbe

 

Session 3: Working Groups

Strategic Vision

The Strategic Vision group evaluated the current landscape of the projects, identified challenges and set goals for the next five years.

Key Points from the Strategic Vision Working Group Discussion
The overlap between projects and alignment
We observed from presentations that there is considerable overlap between projects. Many projects aim to develop FAIRification guidelines, have a goal to apply federated, distributed analytics solutions. Projects, goals and standardization efforts need to be aligned.In Germany, the NFDI projects such as NFDI4Health might be an umbrella involving GO FAIR Germany.On the European level EOSC (European Open Science Cloud) bringing together current and future data infrastructures. A demonstrator within EOSC4life can be considered. Funding option for short time technical projects through ELIXIR can be explored.Challenges
Following challenges are identified as critical for the realization of the PHT:

  • Governance structure
  • Trust infrastructure & container review/certification, KRITIS (Critical infrastructures) (website in German only)
  • Security by design
  • Legal basis for analysis and data exchange
  • Reproducibility
  • Define requirements for the PHT infrastructure
  • Build an open-source community building a production-ready infrastructure
  • Libraries of trains
  • Quality management of code in the container
  • Data discovery and data harmonization

Midterm Goals
Following action points had been identified:

  1. Running trains within Germany between MII partners.
    1. Existing consortial use cases will not serve the purpose.
    2. Interconsortial use cases should be set. Rare Diseases would be a good demonstrator.
  2. Running trains cross borders
    1. Train rails between the Netherlands and Germany:
      1. Advantage: Joint development within the GO FAIR initiative
      2. Disadvantage: No national initiative exists yet.
  3. Train rails between Switzerland and Germany:
    1. Advantage: The Swiss Personalized Health Network (SPHN) is well aligned with MII.
    2. Disadvantage: Switzerland is not part of the EU.

 

Technical Infrastructure:

The afternoon session on the Technical Implementation of the PHT identified several key challenges and suggested directions for future implementations.

Key Points from the Technical Infrastructure Working Group Discussion
Overcome the Trusted Third Party Restriction
Current implementation efforts include at least one central service, the Docker Registry, that is known and addressable by several other services via HTTP. The operator of such an instance can access all images within the registry, hence the operator needs to be trusted. Avoidance of such trusted third parties leads to a preference for decentralized approaches. A potential Peer-to-Peer approach technology might avoid centralization and constitute a platform for developments efforts.Rich Station Metadata Publishing
A train is associated with a set of stations that it visits, potentially with a certain order. The construction of such routes needs to be assisted by metadata, which is provided by the stations. Station metadata should provide information on data quality and completeness. A metadata server may keep an index of available data resources at stations

Trusted Broker
Knowledge of the stations that a train decided to visit leaks information on the station’s data to the train issuer. Hence, the train issuer is generally not allowed to know the stations being visited by the train. This leads to an anonymization component, which hides the set of stations visited by the train from the issuer.

Data Standards
The diversity of data resources at each station poses significant challenges for implementing algorithms that operate on the data. Hence, the wide adoption of the FHIR communication protocol and common schemata directly enhances the potential of the PHT. Exposition of data as FHIR poses challenges for clinical data warehouses, such as tranSMART. The interoperability between tranSMART warehouses and FHIR resources should be explored.

Improve Data Findability
The development of machine learning algorithms requires data exploration and potential pre-analyses. Pre-analysis trains that can cache results for subsequent trains. The PHT infrastructure can benefit from the ability to store results from such pre-analyses at stations for later consumptions by main algorithms.

Station – Train Communication
In the current implementation, the stations do not have any assumptions on the content of trains. Consequently, stations cannot assist trains in their execution, for instance by adaption computational resources to the need of the station. A semantic communication between train and station should be facilitated ideally before routing. The station should know who runs the analytics task and its intention. There could be multiple trains, each focus on another language, format, technology. The right train for the station should be identified.

 

Use Cases

The use cases working group identified the possible application areas of PHT in healthcare and life science domain.

Key Points from the Use Cases Working Group Discussion
This Group had identified a set of Use Cases that could benefit from the PHT approach, such as:

  1. Text Mining (application): access to medical plain text documents which cannot be shared otherwise, run algorithms or models on corpus, could be useful for training and productive phase (for training, human interaction/acccess is necessary)
  2. Feasibility requests: getting the number of eligible patients, popular task with low data protection problems (only aggregated data)
  3. Support for recruitment: contacting and including a patient as a subject, important for clinical trials and registers, more problematic with regard to data protection due to personal identity and mostly lack of consent to data processing or contacting
  4. Record Linkage (PPRL) to find and count e.g. rare disease patients

The importance of providing FAIR data at the creation had been emphasized:

  1. Instruments should be developed to support FAIR data at creation
  2. Success stories: provide plausible scenarios for utilizing FAIR data and show ways for a data protection-compliant implementation
  3. Simple, easy-to-deploy-environment for PHT to demonstrate the advantages (project “Feldbahn”)

The role of ontologies and vocabularies are underlined:

  1. Phenotyping: is everything but HPO
  2. A mapping between European countries f.i. medical terminologies (e.g. BRIDGE DB)

Next Steps:

  1. Prioritization of use cases and goals
  2. Identify a mapping between goals and existing projects

 

Session 4: Discussions, Summary and Outlook

Working groups presented and discussed their results.

The alignment between ongoing projects was discussed. Suggested collaborations:

  • Establish a knowledge exchange forum.
  • Organize topic focused hackathons.
  • Organize training meetings “How to…”, reach out the GO FAIR training pillar INs

Collaboration with international partners should be developed:

  • In rare diseases community, the commonalities and complementarities between developing the PHT and the EJPRD mixed /federated infrastructure can be identified.
  • ELIXIR interoperability meetings can serve as a platform.

Next steps defined as:

  • Write a German manifesto. The international manifesto has been published. Start building a German chapter and including more parties.
  • Consider organizing a follow-up workshop.
  • Create a platform/website with information.
  • Develop specifications: Possible to join regular International PHT architecture group meetings.

This slideshow requires JavaScript.