-
A user-focused, multi-functional, flexible and extensible medical imaging library focused on acquiring, preparing and recording data for use with machine learning.
-
Designed to eliminate the complicated and time-consuming but necessary preprocessing steps at different levels of granularity (dataset-level, scan level, facility-level, patient-level, slice-level)
-
Envisioned as a bridge between various data formats with functionality to account for each modality's specific requirements.
-
An object-oriented programming approach to managing 3D volumes of medical data.
-
Task-oriented and test drove on real-world problems, designed to encourage active experimentation and rapid prototyping in notebook environments.
Demonstration of how to use the library and it's different capabilities are available in the functionality demonstration here.