This demonstration provides and overview of Komprise AI Preparation & Process Automation (KAPPA). Bring your custom data function. Komprise automates the rest.
Komprise AI Preparation & Process Automation (KAPPA)
Overview Video
KAPPA data services make the process of defining custom actions simpler and faster. Simply insert a few lines of code for the requested actions per file into a data operation field. Komprise then performs the steps to execute the custom action across a specified dataset as part of a broader AI data workflow or unstructured data management plan.
In this overview, Komprise Field CTO Benjamin Henry introduces KAPPA data services and compares this approach to legacy data integration approaches.
_______________________
KAPPA Demonstration: DICOM Metadata Extraction
Watch this video to see a KAPPA data service in action.
DICOM files have been the standard for medical imaging for decades, but they were designed for storage and transmission, not AI. The rich clinical metadata embedded in every DICOM header — patient context, study type, imaging protocol, scanner model — is invisible to the storage layer, making it nearly impossible to search, curate, or feed into AI pipelines at scale. This demo walks through how KAPPA solves that problem with a serverless approach to DICOM metadata enrichment that requires no changes to PACS or VNA systems and no complex ETL infrastructure. With just a few lines of Python code, KAPPA extracts embedded DICOM header metadata, loads it into the Komprise Global Metadatabase, and makes clinical imaging data searchable by real clinical criteria across hybrid storage environments. The result is medical imaging data that transforms from something simply stored into something actively useful for AI workflows, research curation, and inferencing pipelines.
Learn more at Komprise.com/KAPPA
Read the KAPPA data services solution brief
Read the blog: Komprise AI Preparation and Process Automation for DICOM Files
KAPPA FAQs
What is KAPPA data services and what does the demonstration show?
KAPPA data services is Komprise’s serverless compute framework for unstructured data that allows organizations to run custom metadata extraction functions directly against files in place — without moving them, without modifying the source storage system, and without building or managing any infrastructure. The demonstration shows KAPPA extracting domain-specific metadata attributes from files across a petabyte-scale data estate using a few lines of Python, writing the enriched tags back to the Komprise Global Metadatabase, and making those tags immediately queryable through Deep Analytics and actionable through Smart Data Workflows. KAPPA data services are available in Komprise Intelligent Data Management and require no database setup, no dedicated servers, and no storage vendor dependencies. Watch on YouTube.
What types of metadata can KAPPA data services extract and why does custom metadata extraction matter for AI inferencing?
Standard file system metadata — name, size, owner, last access date — tells you very little about what a file actually contains or whether it is the right input for a specific AI inferencing workflow. KAPPA data services extract the domain-specific attributes locked inside proprietary file formats that standard indexing tools cannot read: DICOM header fields including modality, body region, diagnosis code, and patient cohort; genomics BAM and FASTQ attributes including instrument ID, run date, and sample barcode; ERP project codes embedded in engineering files; and any custom attribute that a few lines of Python can extract from any file format at petabyte scale. These enriched attributes are written back to the Global Metadatabase as searchable tags, making data that was previously opaque to AI inferencing precisely queryable by the clinical, research, or operational criteria that any given AI use case requires — without moving the underlying data or re-running the extraction on files already processed.
How does KAPPA data services connect to Smart Data Workflows and AI data pipelines?
KAPPA data services is the enrichment layer that makes Smart Data Workflows genuinely intelligent rather than just automated. A Smart Data Workflow that identifies files purely by age, file type, or directory path is applying coarse-grained criteria; a Smart Data Workflow built on KAPPA-enriched metadata can identify all chest CT studies for female patients over 60 with a specific diagnosis code, all genomics files from a specific sequencing instrument within a defined date range, or all engineering documents tagged with a particular project code — reducing billions of files to exactly the right cohort for a given AI inferencing pipeline in seconds. Because KAPPA tags persist in the Global Metadatabase regardless of where the underlying file moves, the enrichment work is done once and reused across every subsequent workflow, analytics query, and AI inferencing request that touches the same data — making KAPPA data services the investment that compounds in value across every AI initiative that follows.
_______________________