Clarification about mask for evaluation

Hi

Just wanted to confirm if the masks provided are being used during evaluation or not. Because, according to this link, all the pixels are used while calculating metrics.

Dear abhiroopt,

Thank you for reaching out to us.

You indeed discovered a mistake in the code that you are referring to. The score_patient function should indeed not create a mask of ones, but read the provided mask. I will correct this mistake on this github page.

However, the score_patient function is only written to make the code for the metrics readable for the participants. The evaluation on the grand challenge platform consists of a bigger code reading in the patient data. I double checked, and the error was not in the official code, and for the mask we read in the provided mask.

I hope this answers your question.

Good luck with the preparations.

Best regards, Evi