-
nVidia GTC
- March 20-23
- https://www.nvidia.com/gtc/?ncid=GTC-NV0MZR1P
- Free!
- Medical Imaging Talks (35 on medical)
- Monai
- Monai for enterprise
- Monai for inference
- Generative AI - generate medical images on the fly
- Synthetic contrast series
- Auto 3D segmentation
- Large language models (radiology reporting)
-
Large Language Models and Medical Imaging
- https://www.pcgamer.com/nvidia-predicts-ai-models-one-million-times-more-powerful-than-chatgpt-within-10-years/
- datascience.cancer.gov talk today on this
- M3GAN
- Singularity: https://en.wikipedia.org/wiki/Technological_singularity
- Hospitals being their own "AI Factory"?
- local data / protocols
- personalization
- Does this require cloud computing?
- Labeling is hard
- need to integrate into radiologist workflow
- ChatGPT is good and summarizing and translating
- Vendors don't have a good way of tracking Algorithm performance at a site
- "Model drift"
- "Data is the new source code"
- Data is still messy/dirty - standard procedure codes
- Trying to standardize procedure codes in canada
- Wide variety of procedure codes today - lots of creativity
- Total Segmentator
-
Poll - who is going to figure out AI/ML In medical imaging?
- Cloud Vendor(s)
- Large radiology group
- Academic Medical Center
- Startup
- Major Vendor
- Researcher/PostDoc
- All of the above?
- Other?
-
Backup facilitator(s)
- email me at [email protected] if interested
-
floating point pixel data
- Parametric Map SOP Class - only thing that does this
- QUICKR created one - Andre F, David Clunie - Steve Pieper can find this
- https://groups.google.com/g/comp.protocols.dicom/c/S8utOHyhmdk/m/PQUOpYjpCQAJ
- https://groups.google.com/g/comp.protocols.dicom/c/ui02tDJ9doQ/m/6ik0MnOi_DIJ