HORUS Dataset Catalog CHUV

The HORUS Dataset Catalog CHUV is the product of a close collaboration between the Data Science team of the DSI Department (DSI-DS) at CHUV and the Data Stewards Biobmed Unit (DSBU).

• The DSI-DS develops and maintains the technical infrastructure of the catalog, ensuring a secure and interoperable environment for sensitive data.

• The DSBU plays a role by supporting researchers during the dataset submission process and by ensuring high-quality curation of metadata related to sensitive clinical datasets.

This catalog is specifically designed to host sensitive CHUV clinical datasets that cannot be anonimized in a secure and restricted-access environment, protecting the data in compliance with legal and ethical requirements. While the datasets themselves remain protected, their metadata is automatically transferred externally to the Zenodo FBM UNIL-CHUV Community, making it FAIR and internationally visible.

This dual-system approach ensures both data protection and scientific visibility: sensitive data stays in a closed environment, while metadata is made accessible for discovery and potential collaboration.

Process Overview

  1. Dataset Submission & Metadata Entry Researchers are supported by the DSBU to upload their dataset into the HORUS Dataset Catalog and fill in all necessary metadata fields describing the dataset.
  2. Metadata Review & Curation A data steward (curator) from the DSBU carefully reviews the submission to ensure the metadata is complete, accurate, and compliant with FAIR principles.
  3. Metadata Publication to Zenodo Once validated, the metadata is automatically published to the Zenodo FBM UNIL-CHUV Community, making the dataset discoverable on an international level, even though the data itself remains protected on CHUV internal servers.
  4. Data Access Request via Metadata When other researchers read a publication referring to a dataset hosted in HORUS — or when they browse Zenodo and find the associated metadata — they gain a detailed understanding of the dataset’s content. If they are interested in using the data, they can contact the Principal Investigator (PI) through its e-mail address listed in the metadata.
  5. Access Evaluation & Agreement The PI, possibly with a designated review committee, evaluates whether the data access request is legitimate. Criteria include:
  • The resquester comply to the Data Sharing Requests and Procedures
  • The requester works in a relevant field,
  • Their study is clearly aligned with the dataset’s subject and goals,
  • A clear scientific justification for the data reuse is provided.
DataSharing Request and Procedures

Researchers interested in accessing data can request the necessary Data Transfer and Use Agreement (DTUA), via the Principal Investigator. Upon request, the scientific committee will evaluate the application. If accepted, the data will be extracted from the Horus CHUV Dataset Catalog at DSI-CHUV, under the supervision of the Principal Investigator.

Requirements for Data Access

Researchers who wish to analyze, reproduce, or gain further insights from the data must submit a request for de-identified patient-level data to heads of the current project. Requests must clearly justify the use of the data and specify the required variables. External users or collaborators should demonstrate a proven track record in the field, provide a detailed research protocol, and explain how they will handle the data ethically and scientifically.

The protocol should include:

  • Collaborators and their affiliations
  • Potential conflicts of interest
  • Project funding details
  • Inclusion/exclusion criteria
  • Scientific questions to be addressed
  • Detailed analysis plan
  • Peer-reviewed publication in the field

Approval Process

The heads of the current project will review the application, recommending any necessary protocol adjustments to enhance quality and ensure alignment with project objectives. Before the data is released, both parties will sign the DTUA based on the Swiss Personalized Health Network (SPHN) template (link). This agreement will define responsibilities related to data protection, management, authorship (in accordance with international standards), and the proper citation of sources and funding. Additionally, it will specify the required date for secure data destruction after analysis is completed, with confirmation of destruction submitted to the project leads.

6. Collaboration & Data Transfer and Use Agreement (DTUA)

    If the PI and committee approve the request, a collaboration agreement can be established. Data can then be shared with the external team under the terms of a CHUV DTUA (The template used for this agreement is based on the one published by SPHN).

    SPHN DTUA link