Joseph Paul Cohen

I asked the account for license information. (https://mobile.twitter.com/josephpaulcohen/status/1244669496960442371) Let's hold off on adding them until it is clear they are meant to be public. The account doesn't list a doctor's...

Sounds cool. Has it been working for you to find candidate papers?

Awesome thanks!

Look in the metadata file.

So these 4 are not labelled: > MERS-CoV-1-s2.0-S0378603X1500248X-gr4e.jpg > auntminnie-2020_01_31_20_24_2322_2020_01_31_x-ray_coronavirus_US.jpg > [email protected] > radiopaedia-2019-novel-coronavirus-infected-pneumonia.jpg I'll need to figure out what there labels should be.

+1

It should be as simple as using this line: ``` model = xrv.models.ResNet(weights="resnet50-res512-all") ``` in this script: https://github.com/mlmed/torchxrayvision/blob/master/scripts/transfer_learning.ipynb Also change the resizing to xrv.datasets.XRayResizer(512) so the images are 512x512

Oh sorry I responded to fast and didn't test the code. The resnet loads an internal resnet model inside so the fc is located at `model.model.fc`. ```python model = xrv.models.ResNet(weights="resnet50-res512-all")...

Hey sorry for my delay in getting back to you. Your code looks correct. I processed the image you posted using this script: https://github.com/mlmed/torchxrayvision/blob/master/scripts/process_image.py It seems the densenet at a...

I also took a look at what the 224x224 densenet model was looking at using the gifsplanation approach (https://arxiv.org/abs/2102.09475) using this code: https://colab.research.google.com/github/mlmed/gifsplanation/blob/main/demo.ipynb The images are not great but they...