Aignostics published research results for a new pathology foundation model developed in collaboration with Mayo Clinic Digital Pathology. The model, called Atlas, was built in two months using 1.2 million diverse slide images from Mayo Clinic and Charité – Universitätsmedizin Berlin.
The company said the model establishes a new state-of-the-art for performance across more than 20 public benchmarks. In addition, the model is computationally efficient, enhancing its usability relative to existing image-based models.
Now Mayo Clinic is highlighting its deployment of NVIDIA DGX SuperPOD with NVIDIA DGX B200 systems, an infrastructure, provides state-of-the-art AI compute capabilities to support digital pathology and other areas of development. The Blackwell infrastructure to accelerate pathology slide analysis and foundation model development, reducing four weeks of work to just one, according to Mayo Clinic.
【MORE】