Skip to content

[Feature] Differentiable Radiomics (End-to-End Deep Learning Integration) #3

Description

@helloerikaaa

The Idea: Because fastrad uses standard PyTorch tensor operations, we can theoretically preserve the computational graph (requires_grad=True).

Why it's cool: This would allow researchers to embed standard radiomics features inside a deep learning neural network. For example, you could train a CNN where the loss function directly optimizes for GLCM Contrast or Shape Sphericity, calculating backpropagation straight through the fastrad feature extractor!

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions