This invention describes a system for training artificial intelligence to identify specific markers, such as radiopaque markers, within angiography images. It works by feeding the AI a large set of angiography images along with "ground truth" data that precisely indicates the markers' locations. The AI then learns to automatically detect these markers and improve the alignment of medical images.
Why it matters: The fundamental challenge of acquiring and meticulously labeling vast amounts of angiography image data for ground truth, as required for training, remains a significant hurdle since 2024.
AI gives you a few directions you could take this. Pick one, and we check whether your version is different enough to patent, then write the filing.
Reinvent this with AI