UNIFIED MULTIMODAL DATASTORE (UMD)
Genomcore UMD is, probably, the world’s most advanced system for multi-specialty biomedical data storage and retrieval. Our technology enables flexible and efficient interoperability and observability for large specialty data types such as genomics, time-series or clinical biovalues.
HIGH-PERFORMANCE SPECIALIZED BIOMEDICAL DATASTORES
Genomcore UMD allows fine-grain structured storage of precision-health specific datapoints
Genomics datastore supports high performance storage and retrieval of individual variant calls for germline, somatic and trio experiments as well as custom annotations
Biovalues datastore allows efficient storage and retrieval of metabolic biomarkers, supporting on-the-fly normalization using LOINC codes and associated reference thresholds
Image datastore allows metadata extraction from DICOM files and their integration with other entities in the UMD
Metrics datastore is designed to store time-series from activity monitors and biosensors enabling rangedĀ queries with automatic aggregation functions
Clinical Records and patient-owned datasets can be customized with UMD Records system, supporting structured sections and onthology terms.Ā
TYPED RECORDS
UMD Records system features a highly flexible and powerful way to model complex domain-specific data schemas, including primary data storage, associated semantics and type-dependent built-in operations.
Document
Link and store generic files and metadata
Form
Structured information retrieval from end-users
Image
Link and store medical imaging files and metadata
Insight
Structured report findings for results delivery
Ontology Term
Coded, controlled and translatable terms
Patient
Personal details can be linked to an account
Study
Use-case specific multimodal data integration
Task
Launch configurations and results from apps
CUSTOM RECORDS SCHEMA MODELING
- Text
- Numeric
- Link File
- Link Record
- Date and time
- Boolean
- Options
- URL
DISTRIBUTED FILE SYSTEM
For storage of BLOB (Binary Large Objects) files, Genomcore supports standard file systems and S3 objects storage connectors, providing an unified view of files using an emulated file browser with support for arbitrarily large uploads, restricted public URLs sharing and embeded viewers for certain data files.
IMPORTERS, PARSERS AND CONNECTORS
