Visian

With growing dataset sizes, a key limiting factor in training AI models on medical images is human annotation capacity. To show the AI what they are interested in, medical researchers often need to paint structures of interest, such as tumours, on a large number of images. This is especially challenging for 3D images such as MRI and CT.

Visian is a web-based annotation tool primarily focused on MRI and CT scans to support researchers in working faster and more efficiently. It does so by offering a uniquely intuitive user interface, combined with powerful “smart brushes” and volumetric rendering technology.

We created the identity, user experience, and design. Additionally, we led the technical architecture of the product and implemented key technical components such as Visian’s real-time volumetric ray-tracing.

The project originated from a Bachelor’s project at the Hasso Plattner Institute but continued long after its completion. It partnered with the Max Delbrück Center for Molecular Medicine and Charité in Berlin, Mount Sinai in New York City, as well as the AI4Health Focus group of the World Health Organization. Now, the development of Visian is continued by Data4Life, and it is still helping medical researchers navigate volumetric data every day.