Ingestion of Biomedical Images

Learn how you can use TileDB to load biomedical image data efficiently into a TileDB-BioImaging dataset with optimized read performance.

Ingestion is the process of pulling out image data and metadata from image files, transforming the data into an analysis-ready format, and loading it into a TileDB-BioImaging dataset. This section describes the basic ingestion process given by the TileDB-BioImaging open source code, best practices for optimizing read performance on large datasets, and how to use TileDB’s scalable ingestion to automate ingestion.

Basic ingestion

The TileDB-Bioimaging open source code has basic ingestion functionality for the following file formats:

• TIFF/OME-TIFF
• OpenSlide supported formats
• OME-Zarr
• PNG

Ingestion on TileDB

Scalable ingestion on TileDB fully automates ingesting biomedical image files. TileDB gives not only a programmatic API for ingesting biomedical datasets at scale but also a UI for simplifying the process of bringing your images into TileDB as shown in Catalog: Biomedical Images.