TL;DR
Paris-based Raidium has launched an AI-powered radiology platform at the Moffitt Cancer Center. Its Curia model automates tumor tracking and reduces reader variability by a factor of 3.
RaidiumA Paris and Silicon Valley-based radiology startup, Launched an AI-based imaging platform in the US at the Moffitt Cancer Center, one of the nation’s leading oncology research facilities. The platform, called Raidium Read, replaced Moffitt’s older radiomics software and is now available for clinical trials and research. FDA 510(k) clearance is expected by the end of 2026.
The system is built around Curia, Raidium’s proprietary foundation model trained on more than 200 million CT and MRI slices from 150,000 exams. Instead of layering AI tools on top of an existing PACS viewer, the company built the viewer itself from scratch with an embedded model. Curia performs organ-agnostic, automated RECIST measurements, the standard method for monitoring tumor response to treatment, at multiple time points. Raidium says this reduces reader-to-reader variability by a factor of three.
The practical problem that Raidium solves is tedious and consequential. Oncology radiologists manually track lesions on successive scans, taking measurements from previous studies and comparing them to the new imaging. This workflow is time-consuming and creates inconsistencies between readers. Raidium Read automates it: the system scans large volumes of imaging inputs, detects and segments lesions by anatomical region, and maps historical lesion data against new follow-up scans. Corti’s Symphony AI took a similar approach to medical codingtreating an error-prone clinical task as a reasoning problem rather than a labeling problem.
“For two decades, standard PACS trackers have resisted evolution,“said Paul Herent, CEO and co-founder of Raidium. Moffitt radiologist Dr. Cesar Lam said the platform enables research projects.not so long ago it seemed impossible.” The system requires no backend integration, making deployment faster than traditional PACS setups. Artificial intelligence has already shown that it can outperform biopsies in evaluating rare cancershowever, most of these tools remain research prototypes. Raidium’s argument is that building the tracker around the model, rather than tying the model to the tracker, is what will finally bring AI radiology into the daily clinical workflow.
