TL;DR
Paris-based Raidium launched its AI-native radiology platform at Moffitt Cancer Center. Its Curia model automates tumor tracking and cuts reader variability by 3x.
Raidium, a Paris and Silicon Valley-based radiology startup, has launched its AI-native imaging platform in the US at Moffitt Cancer Center, one of the country’s leading oncology research institutions. The platform, called Raidium Read, replaced Moffitt’s legacy radiomics applications and is currently available for clinical trials and research use. FDA 510(k) clearance is expected before the end of 2026.
The system is built around Curia, Raidium’s proprietary foundation model trained on over 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 the model embedded. Curia performs organ-agnostic, automated RECIST measurements, the standardised method for tracking tumor response to treatment, across multiple time points. Raidium says this cuts inter-reader variability by a factor of three.
The practical problem Raidium is solving is tedious and consequential. Oncology radiologists manually track lesions across sequential scans, pulling measurements from prior studies and comparing them to new imaging. This workflow is time-consuming and inconsistent between readers. Raidium Read automates it: the system scans large-volume imaging inputs, detects and segments lesions across anatomical regions, and maps historical lesion data against new follow-up scans. Corti’s Symphony AI took a similar approach to medical coding, treating an error-prone clinical task as a reasoning problem rather than a labelling one.
“For twenty years, the standard PACS viewers have resisted evolution,” said Paul Herent, Raidium’s CEO and co-founder. Dr. Cesar Lam, a radiologist at Moffitt, said the platform enables research projects that “would have seemed impossible not too long ago.” The system requires no backend integration, which makes deployment faster than traditional PACS installations. AI has already shown it can outperform biopsies at grading rare cancers, but most of those tools remain research prototypes. Raidium’s bet is that building the viewer around the model, rather than bolting the model onto the viewer, is what finally gets AI radiology into daily clinical workflow.
