Hi,
This is David, and I have a couple of questions about task 2. In Disease_ID.tsv, there are 624 patients with 16 Rare-X disease names. As disease names in this tsv file did not use external resources like Orphanet or OMIM, I am confused about which disease set I should predict diagnosis with.
1. Is task 2 limited to these 16 diseases? Or is it to predict diagnosis not only limited to 16 diseases above?
2. If task 2 is about predicting a diagnosis of entire rare disease cases, should we be able to convert external disease names to Rare-X data disease names? Or could we directly use external resources for disease names?
Thank you for your help.
Created by Sukwon(David) Kim sukwonk I'm so glad I could help @sukwonk! Hi @mdsage1! Thanks for making task2 clear. It helped me a lot! Hi David,
The objective for Task 2 is to differentiate between between the 16 RARE-X diseases. You can absolutely utilize external data sets to identify relevant features to use in the RARE-X data to differentiate between the 16 RARE-X diseases, but the final predictions must be for one of the RARE-X disease labels.