The origin and label of metastatic samples

[peakhong]

Could you indicate whether the label of the metastatic sample you used was consistent with the primary site or with the section location? If the primary colorectal cancer has metastasized to the liver, do you use images from the colorectal or liver? What is the label when using slices from the liver? I would appreciate it if you could answer me!

[ciaobladoo]

I am not the author. But if you read their paper, the goal is to predict the primary origin. So if a colorectal cancer metastasized to liver, the label would be colorectal cancer. I am not sure if I understood your second question "If the primary colorectal cancer has metastasized to the liver, do you use images from the colorectal or liver?" But where the image come from is dependent on the patient. If a tumor is found in liver, then the biopsy or surgery will be performed at the liver. So the image will be from liver. Whether it is originated from liver (i.e., liver cancer) or colon (i.e., metastasized colon cancer) can be determined (or more strictly speaking, inferred) by a pathologist. This algorithm is developed to do just that.

[peakhong]

Thank you for your answer. The reason why I have questions is that there are too few data labeled as
"Metastasis" in TCGA database, and the biopsy site of the images labeled as "Metastasis" is from the primary site. For example, from the colorectal to the liver, where the biopsy site is colorectal but labeled as "Metastasis" . So I would like to know the author's use of the biopsy sites labeled as "Metastasis" images. As in TCGA, the primary site is used for prediction, then the algorithm is of little significance. By the way, do you know where I can get the transferred image?