This is a template for multi-modal machine learning in healthcare using the Kedro framework. You can fuse reports, tabular data, and images using various fusion methods (Early & Late fusion. Other fusion methods and graph data are a work in progress). This project is compatible with Kubeflow and Vertex AI.
- For understanding the Kedro platform, please start with the overview 👇
- This is a template repository. Generate a new repository with the same directory structure by selecting the Use this template button ☝️ and use it as a Kedro project.
- Install dependencies
pip install -r src/requirements.lock - Refer to the default pipeline for usage examples.
- Refer to sample data for data format. Prefix model datasets with appropriate model type from image_ , text_ , tabular_ , and bert_. (text_ is for CNN text models)
- Refer to catalogue for inputs and outputs
- See parameters that can be tweaked.
The essential pipelines are in requirements.txt. More details on the components are in their respective repositories 👇 (PR welcome. Read CONTRIBUTING.md in the repositories)
- kedro-tf-image
- kedro-tf-text
- kedro-tf-utils
- kedro-dicom (optional) for processing DICOM images
- kedro-graph (optional) for creating DGL graph from multimodal data.
- fhiry (optional) for flattening FHIR resources.
Familiar guide, troubleshooting help, rules and guidelines, project dependencies, and packaging your Kedro project details stay the same. Make sure to acknowledge the new URLs and project paths.