TL;DR
Paris-based Raidium launched its AI-native radiology platform at Moffitt Cancer Center. Its Curia model automates tumor tracking and reduces reader variability by 3-fold.
radioa radiology startup based in Paris and Silicon Valley, has launched its native AI imaging platform in the US. at Moffitt Cancer Center, one of the nation’s leading cancer 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 based on Curia, Raidium’s proprietary core model trained on more than 200 million CT and MRI slices from 150,000 exams. Instead of overlaying AI tools on top of an existing PACS viewer, the company built the viewer from scratch with the built-in model. Curia performs automated, organ-independent RECIST measurements, the standardized method for tracking tumor response to treatment, at multiple time points. Raidium says this reduces variability between readers by a factor of three.
The practical problem that Raidium is solving is tedious and important. Oncology radiologists manually track lesions through sequential scans, extracting measurements from previous studies and comparing them to new images. This workflow is time-consuming and inconsistent across readers. Raidium Read automates it: the system scans high-volume image inputs, detects and segments lesions into anatomical regions, 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 one.
“For twenty years, standard PACS scopes have resisted evolution,“said Paul Herent, CEO and co-founder of Raidium. Dr. Cesar Lam, a radiologist at Moffitt, said the platform enables research projects that “It would have seemed impossible not long ago.“The system requires no backend integration, making implementation faster than traditional PACS installations. AI has already shown that it can outperform biopsies in classifying rare cancersbut most of those tools remain research prototypes. Raidium’s bet is that building the viewer around the model, rather than fixing the model to the viewer, is what finally brings AI radiology into the daily clinical workflow.
