A whole-slide foundation model for digital pathology from real-world data. (https://pubmed.ncbi.nlm.nih.gov/38778098/)

These scientists wanted to create a special computer program that can help doctors look at very detailed pictures of tissues to find diseases like cancer. They used a lot of pictures from patients in a big health network in the US. They trained their computer program on over a billion tiny image tiles to help it learn how to recognize different types of tissues and diseases. Their program, called Prov-GigaPath, is really good at identifying different types of cancer and other problems in these pictures. They tested their program on many different tasks related to studying diseases in tissues, and it performed better than other methods in most cases. This study shows how important it is to use a lot of real-world data to train computer programs to help doctors make better decisions.

Xu H., Usuyama N., Bagga J., Zhang S., Rao R., Naumann T., Wong C., Gero Z., Gonzalez J., Gu Y., Xu Y., Wei M., Wang W., Ma S., Wei F., Yang J., Li C., Gao J., Rosemon J., Bower T., Lee S., Weerasinghe R., Wright BJ., Robicsek A., Piening B., Bifulco C., Wang S., Poon H. A whole-slide foundation model for digital pathology from real-world data. Nature. 2024 May 22. doi: 10.1038/s41586-024-07441-w.