Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Detection model in the model zoo, does not support datasets with no nodules #8126

Open
DSRajesh opened this issue Oct 2, 2024 · 0 comments

Comments

@DSRajesh
Copy link

DSRajesh commented Oct 2, 2024

Hello
The Monai bundles are convenient to use because all the config for traning, transforms etc are specifyable easily as JSON and we are fortunate to have such a turnkey way of training the detection pipeline. The reference Lung Nodule detection, which is really the only detection in the model zoo, does not support datasets with no nodules, which would be a very odd thing to do in the medical field, because one would expect several healthy patients with no lung nodule. A medical detection network that is trained only on diseased samples is flawed right in its data input distribution. I am guessing that something like StandardizeEmptyBoxd needs to be added to the _transforms section of the lung_nodule_ct_detection/configs/train.json of the model zoo https://github.com/Project-MONAI/model-zoo/releases/download/hosting_storage_v1/lung_nodule_ct_detection_v0.5.9.zip. Any guidance would be appreciated.

Thanks
Rajesh

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant