Description

Background

MOSAIC is a scientific initiative led by OWKIN and research partners to integrate multimodal and spatial oncology data at scale. Communicating this infrastructure meant making a complex system visible: heterogeneous datasets, non-linear relationships, and analytical processes powering AI models. The project was conceived to make this knowledge infrastructure transparent without oversimplifying its scientific depth.

Objective

Design two interactive visualizations to make complex medical datasets readable and clarify their role in advancing AI solutions in oncology.

Result

Two interactive visualizations define the narrative: one renders the patient dataset tangible, the other reveals how those data drive AI models and advanced oncology solutions.

Year:

2026

Industry:

Associations

The Set-up

The visual system is based on an elemental unit: a graphic particle representing the patient and structuring the entire framework. The six data modalities activate and reorganize this dynamic constellation: selecting a modality or filtering by tumor type reshapes cases, revealing the relational nature of the data. The same language guides the product-focused view: an ordered, non-hierarchical network where different dataset combinations move through analytical layers to generate concrete AI use cases.

Production Process

The design process integrated solutions that amplified its impact. Animations were developed in Rive, enabling a level of micro-motion detail that would have been unattainable through code-only implementation.

In the second visualization, the combinatorial network structure and the transformation of traditional icons into dot-based configurations were enabled by custom AI scripts, generating dynamic systems and reinforcing continuity between the visualizations.

The Backend Technology

The first visualization was developed using Matter.js, a JavaScript library for 2D physics simulation. Each patient, modeled as a point, behaves as a reactive entity, contributing to a dynamic and organic ecosystem.

The solution was delivered as a custom front-end component, embedded into the client’s Webflow site through dedicated integration, ensuring compatibility and full control over the digital experience.