MetFinder
Stained slices are
tiled and analyzed
A classification
heatmap is generated
Highlighting cancer tissue
MetFinder is a tool for quantification of melanoma metastasis tumor burden in mouse whole slide images of livers and brains.
MetFinder relies on a deep-learning network trained by manual annotations from pathologists to recognize tumor and non-tumor regions in a whole slide, H&E stained image, removing artifacts and other types of tissues from the analysis.
Thus far this tool is trained for the quantification of tumor burden resulting from human xenograft models of melanoma metastasis in mouse livers and mouse brains.
Disclaimer: Due to the diversity of tissues, sample preparations, scanners, and possible artifacts, we provide this tool “as is” without warranty of any kind. This tool was trained on mice samples and is for research purposes ONLY. Once processed, the output of the jobs will be publicly available, including the file names. The slide label image is not shared nor included in the results. DO NOT upload slides from human subjects. Reminder that this tool is not meant for slides from human subjects. If uploaded, users are responsible for consent, anonymization and general handling of such slides which falls under the Declaration of Helsinki and are protected by the U.S. Department of Health and Human Services (HHS) regulations.
