Research engineer at Institut Cochin, Raphaël Braud-Mussi leads the development of OpenCID, an open platform designed to simplify the management, processing and sharing of biomedical data. Initially created to support imaging data workflows at Institut Cochin, OpenCID has progressively expanded to other types of biomedical data and is now being deployed beyond its original environment.
In this interview, Raphaël shares the ambitions behind the project, its benefits for researchers and imaging facilities, and the potential synergies with BioImage Cloud within the France-BioImaging ecosystem.

To begin with, could you briefly introduce yourself?
Hi, I’m Raphaël Braud-Mussi. I’m a research engineer at Institut Cochin in Paris, where I lead the development of OpenCID, a platform developed with an open collaboration model that helps thousands of scientists manage and process their biological data.
I started at the institute in 2023, working on interoperability between OMERO and our in-house image database, and integrating deep learning tools for image segmentation. Before that, I studied mathematics and computer science at Université Paris Cité, then completed a master’s degree in image processing and computer vision at Université de Montpellier.
What’s driven me throughout is a genuine interest in building tools that make researchers’ work more efficient and OpenCID is exactly that kind of project
What are you currently working on?
Right now, the team and I are focused on the next major version of OpenCID. The core of the platform has been entirely rebuilt in Python, which ensures portability across server environments and makes it easier for the broader community to contribute to the codebase.
This new version will also include a redesigned web interface with more fluid and intuitive workflows for researchers. We’re planning to deploy it at Institut Cochin this summer, with a rollout to other institutes starting in September.
My role is to make sure everything runs as expected across environments and to coordinate the deployment with each site. Beyond that release, we already have a roadmap of new features to continue expanding what the platform can do; we’re currently managing over 350 TB of data at Cochin with a data flow exceeding 10 TB per month, so scalability stays a constant priority.
Could you explain what OpenCID is and what need it addresses?
OpenCID started as an internal project at Institut Cochin called CID. The Cochin Image Database is designed to help researchers manage their imaging data. As the user base grew, we expanded the platform to include processing pipelines directly accessible through the web interface, removing the need for researchers to install software locally.
Institut Cochin hosts a large number of core facilities across different disciplines, so once the imaging management layer was stable, we extended OpenCID to handle other types of biomedical data (genomic, proteomic, cytometry…) produced by those facilities as well.
Today, OpenCID is openly available upon request and allows research institutes to manage and process large volumes of biomedical data entirely through a web interface. It’s built to scale, and its open nature makes it easier for the community to adopt and contribute to it.
In practical terms, how does OpenCID work for users?
Three words: simplicity, efficiency, usability
Let me give you a concrete example. You’re a biologist who needs to acquire microscopy images and RNA sequencing data for your project. Traditionally, getting that data from the core facility to your workstation means copying files onto an external hard drive or by using a shared space on your network, a process that eats into your time slot on the acquisition machine and breaks your workflow.
With OpenCID, everything is streamlined. Once your acquisition is done, leave the machine to the next user. Then log into OpenCID, select the machine you used and your data folder directly from the web interface, and import your files into the database in just a few steps, accessible from anywhere, at any time. From there, you can run your processing pipelines (segmentation, deconvolution…) and download your results with a couple of clicks, without ever installing software locally, whether you’re working remotely, attending a congress, or even on the beach.
Who can currently benefit from OpenCID?
OpenCID is currently deployed at Institut Cochin, where it serves users across multiple core facilities, including IMAG’IC for bioimaging. External users who acquire data at the institute can also import and retrieve their own data through the platform, even if they are not based at Cochin. The platform is designed to go well beyond a single institute.
Any research institute interested in deploying OpenCID can reach out to the team at cid.u1016@inserm.fr to get started. The solution is ready for deployment, and we are genuinely happy to support new sites through the process. For those who want to explore first, we can also provide access to our GitLab repository, where the documentation and the full deployment procedure are available.
What are the main benefits of OpenCID for researchers, engineers and imaging facilities?
For researchers, OpenCID simplifies the management of an entire team’s data and supports a genuine open science workflow. Every dataset can be shared and downloaded based on access rights defined by the team leader, and an automated open science link generator allows any dataset to be connected to a DOI, making data citation and publication straightforward.
Beyond that, OpenCID is designed to help research teams meet the requirements of Data Management Plans, which are now mandatory for most public funding bodies. By centralizing data storage, traceability, and sharing in a single platform, OpenCID makes it significantly easier to demonstrate compliance with FAIR data principles, findable, accessible, interoperable, and reusable, without adding administrative burden to the team. (https://hal.science/hal-04834159v1/file/INSTITUT_COCHIN_ENTITY_DMP_-_CID_V1.pdf)
For engineers, OpenCID brings structure and accessibility to data management. Processing pipelines can run workflows on a workstation connected to the OpenCID system, allowing engineers to launch analyses on a dedicated machine and continue working on other tasks in the meantime. Results are made available directly on the web interface as soon as the workflow completes. And thanks to our fully Python-based codebase, integrating new workflows is straightforward, either by requesting an addition from the team, or by contributing directly if you are comfortable doing so.
For imaging facilities, OpenCID provides a centralized and reliable solution to handle the constant flow of data produced by acquisition machines, with no need for users to manage everything manually.
Ultimately, every one of these benefits points to the same goal: reducing friction at every step of the research workflow, so that scientists and engineers can focus on what actually matters; science
Could OpenCID be deployed across other imaging facilities?
As mentioned earlier, this is precisely our current priority. OpenCID is already being deployed beyond Institut Cochin, we are actively working with sites such as INRAE Montpellier and Institut Max Planck in Cologne, and more are in discussion.
The community-driven nature of the platform is a key enabler here: it lowers the barrier to adoption and allows each site to adapt the solution to their specific environment. Our affiliation with France-BioImaging also gives us a strong network to reach imaging facilities across the country and beyond.
Building a community around OpenCID is not just a goal, it is part of the project’s long-term vision.
What feedback have you received since OpenCID has been implemented?
It has been three years of sustained development rethinking the entire CID project from the ground up. Now that version 2.0 is ready and continuously updated, the feedback from users has been very encouraging.
A big part of that trust comes from how we handle support. The team commits to addressing any issue within 24 hours, whether we identify a bug proactively or a user reports one directly. That responsiveness has built a real relationship with our user base.
User feedback is also a direct input into our roadmap. We actively seek opinions on the platform, and many of the features we have built or are planning came directly from conversations with researchers and engineers in the field.
What are the next steps for OpenCID?
The next steps for OpenCID are threefold: reinforcing the team, completing the current deployment rollout, and delivering several new features that are already in progress.
On the technical side, we are currently in the testing phase for Zarr format support, which will be a significant step forward for open science interoperability. We are also developing an online 3D object viewer, which will allow researchers to visualize volumetric data directly from the web interface without any local software. Both features are on track to be released soon.
The project is at a stage where it needs additional engineers to sustain both maintenance and new development in parallel, and that is precisely what we are actively working to address.
FBI.data recently launched BioImage Cloud. In the future, do you see possible synergies between OpenCID and BioImage Cloud?
OpenCID and BioImage Cloud have been developed with different but complementary goals. BioImage Cloud focuses on streamlining the transfer of imaging datasets from acquisition to national cloud storage, while OpenCID provides a broader platform for data management, processing, and sharing, regardless of data type or format.
Given that both projects are part of the France-BioImaging ecosystem and share the same underlying objectives, making research data FAIR, accessible, and manageable, I genuinely see potential for synergy. One concrete avenue would be to explore how OpenCID’s data management layer could interface with FBI.data transfer infrastructure, effectively combining the strengths of both solutions. A proof of concept in that direction would be worth considering.
More broadly, I think the community benefits from having multiple approaches being explored in parallel. The most important thing is that researchers end up with tools that actually work for them.
















