Children’s Hospital of Philadelphia now models a child’s heart in seconds using open source NVIDIA AI, the Nordics must decide tonight whether to build sovereign medical imaging stacks or import them.

NEW DEVELOPMENT Children’s Hospital of Philadelphia runs a cardiac modelling service on open source NVIDIA AI. The service turns CT, MRI and 3D ultrasound scans into anatomically precise 3D heart models in seconds, replacing a four-hour manual workflow. The models are used before surgery to choose the right device for each child’s unique anatomy. TECHNOLOGY The stack is open. MONAI, the medical imaging framework, is Apache 2.0, maintained by NVIDIA, NIH, King’s College London, Mayo Clinic and thirty other institutions. MONAI Label plugs into 3D Slicer, a free research platform, and SlicerHeart, a cardiac toolkit built by the hospital’s own lab. NVIDIA Warp, also Apache 2.0, runs GPU-accelerated physics simulations that cut device simulation time from four hours to near real time. ADOPTION More than twenty children’s hospitals in the US now run cardiac modelling programmes. Boston Children’s Hospital uses modelling in over half of its cardiac surgeries, roughly five hundred cases a year. CHOP expects two hundred modelled cases in 2026. The hospital’s IDEA Lab, part of the Morgan Center for Research and Innovation, is now applying the same tools across multiple disciplines. NORDIC CHOICE The Nordics have the talent, the data and the clinical need. We can import the stack, train our own segmentation networks on Nordic scans, and run the service inside Nordic hospitals. Or we can build a sovereign stack: fork MONAI, add Nordic governance, and keep the data and the models inside Nordic borders. The choice is binary: import or build. POLL

researched · 3 sources

15 Sepreaches everyone