Open
Description
The goal is to extend MedPipe3D to allow loading and processing of tabular metadata alongside medical images, supporting multi-modal model inputs and use cases like combining imaging with patient data.
Success when:
User can supply a CSV (or DataFrame) with clinical metadata.
The pipeline maps and batches it alongside image tensors.
The model accepts and uses both image and tabular data in forward pass.
Metadata
Metadata
Assignees
Labels
No labels