Reimagine the way to share computer vision algorithms within the bioimaging community
Client
Headed by Timo Betz, BetzLab is on a mission to study how mechanics is involved, and sometimes even responsible, for the correct and robust interaction between different biological components such as cells and tissues.
The lab develops new measurement methods to quantify forces and tension in 3D tissue, membranes and filaments. They use fluorescence microscopy image data as one of the main sources of information for the quantification. Hence, the lab works on developing robust computer vision algorithms to process and analyse microscopy images.
Challenge
The BetzLab developed an algorithm to automatically estimate traction force field from image data. The algorithm infers highly valuable information to study cellular mechanics. However, the team had a challenge to share the algorithm with collaborators.
In the community studying cell biomechanics, researchers often come from different backgrounds with a variety of expertise. That is why having an algorithm without an intuitive interface can reduce the value of the algorithm and hamper research.
Our approach
To address the challenge, we decided to design a web-based application with a minimalistic and intuitive interface facilitating exchange of the algorithm among stakeholders from technical and non-technical background.
We also setup a goal to come up with a reusable interface that our team could use for deploying other bioimaging algorithms and promote collaboration within the scientific community.
Conducting research and preparing mockups
Our first step was to design a flawless and intuitive workflow that even inexperienced users could understand without any prior knowledge about how exactly the algorithm works.
The workflow should also enforce a set of robust controls and validation mechanisms to ensure the correctness of input data and prevent any unexpected user actions that could lead to erroneous computation results.
After conducting the research, we came up with a step-by-step workflow that limits the amount of information user needs to consider in each action, thus reducing the user’s cognitive load.
“I was impressed by the capabilities of the algorithm and the potential value it could bring to the scientific community. As soon as I saw the algorithm in action, I said to myself: wow, what if we made a design that could be accessible to every researcher, regardless of their computer science skills. Since then, my #1 goal has been to deliever the easiest, most intuitive bioimaging tool to the market.”
Prototyping and creating design system
After defining the workflow, we implemented a prototype to test it with actual users of the algorithm. We collected feedback on how the first users interacted with our prototype and analysed usage data to refine and optimise the workflow.
To facilitate prototyping, we developed a proprietary design system that helped us reduce the time required to incorporate for the user feedback and refine mockups. We especially focused on developing interactive forms, inputs as well as widgets to visualise 2D and 3D fluorescence microscopy images.
Developing web-based dashboard
When we obtained positive feedback about the state of our prototype, we refactored it into a production-ready web-based application that is ready to scale depending on the algorithm usage.
We designed the application in such a way that it requires a minimum effort from the algorithm developer to integrate the algorithm into the application. It makes it also possible to reuse the application for other algorithms.
Results
Within the first 3 months of our collaboration with BetzLab, we designed an intuitive user interface, validated it with users, and successfully launched the first version of a web-based dashboard to run the algorithm for estimating traction force field from image data.
After proving the prototype to be successful, we reworked it into a production-ready solution that BetzLab uses internally and plans to release publicly in the coming months.
The application has taken the ability to distribute the algorithm to the next level. Removing the hassle of installing the algorithm locally and trying to figure out how to run it, users can run the algorithm and share results using our web-based application accessible on desktop, tablet and mobile devices through a user-friendly interface.
We’re proud to work closely with Timo Betz and be a part of the journey of his lab in understanding how biological systems perform their daily functions.
“I was impressed by the capabilities of the algorithm and the potential value it could bring to the scientific community. As soon as I saw the algorithm in action, I said to myself: wow, what if we made a design that could be accessible to every researcher, regardless of their computer science skills. Since then, my №1 goal has been to deliever the easiest, most intuitive bioimaging tool to the market.”
Get in touch
Don't hesitate to take the next step towards your data-driven future.
