Hello In a previous thread, Dezso said: "If a cancer is growing slowly, even if it is there and it is visible, it might not be diagnosed within a year." We observed this situation during the challenge for cancer patients with several exams where the scores of previous exams were higher than usual During training we labeled these exams as negative since cancer was diagnosed after two o more years. But now we are unsure about this. So this opens the following question, should our algorithms learn when the cancer should be diagnosed or when the cancer is visible? In the first case our labeling is correct, in the answer is the second our labeling is not This question is also related to how radiologists are evaluated.... If the cancer is there but the radiologist does not diagnose it after two more years, is it a false negative of the radiologist at that time? What do you think?