dazhangyu123 / ACMIL

Attention-Challenging Multiple Instance Learning for Whole Slide Image Classification (ECCV2024)
MIT License
66 stars 8 forks source link

PathGen-1.6M compared with CONCH and UNI #9

Closed NBitBuilder closed 4 months ago

NBitBuilder commented 4 months ago

Hi, thank you for your contributions to ACMIL and PathGEN-1.6 M!

Are there any results of PathGEN-1.6 M compared with CONCH and UNI?

I didn't find it in the PathGEN-1.6 M paper.

Thanks!

dazhangyu123 commented 4 months ago

Although we applied for usage permissions for the CONCH and UNI checkpoints on Hugging Face, we have not yet been granted access to use them. As a result, we are currently unable to make comparisons with CONCH and UNI.

dazhangyu123 commented 4 months ago

If we are granted access to use CONCH and UNI checkpoints, we would be happy to update our results accordingly.

NBitBuilder commented 4 months ago

Thank you for getting back to me. When will you release the PathGen-1.6M model and dataset? Any timeline please?

dazhangyu123 commented 4 months ago

We plan to publicly release the PathGen-1.6M model and dataset in the middle of next month. Currently, the data is undergoing an internal approval process, so the exact release date will be finalized once this approval is completed.

NBitBuilder commented 4 months ago

thanks for your updates