PathAI is excited to share ContriMix, our new machine learning approach that helps our digital pathology algorithms generalize across staining techniques, scanner types and other kinds of variation without the need for pathology annotations or pre-specifying image characteristics and metadata. ContriMix is a powerful domain generalization technique that has ranked among the top algorithms developed for digital pathology data on the public
Stanford WILDS Camelyon17 dataset. You can read the
full pre-print here.
Access the full article below to learn more about ContriMix, and how we use it to build state-of-the-art pathology algorithms. Tan Nguyen, Dinkar Juyal, Jin Li, Aaditya Prakash, Shima Nofallah, Chintan Shah, Sai Chowdary Gullapally, Michael Griffin, Anand Sampat, John Abel, Justin Lee, Amaro Taylor-Weiner
Access the full article below to learn more about ContriMix, and how we use it to build state-of-the-art pathology algorithms. Tan Nguyen, Dinkar Juyal, Jin Li, Aaditya Prakash, Shima Nofallah, Chintan Shah, Sai Chowdary Gullapally, Michael Griffin, Anand Sampat, John Abel, Justin Lee, Amaro Taylor-Weiner
