Clinical data warehouses

20% structured data. What about the other 80%?

Clinical information goes beyond the fields in your databases.

Much medical knowledge lies dormant in documents. CLARYS Intelligence makes it usable, structured and traceable, without adding new sources.

The problem

Information missing from the coding is not missing information.

Your warehouses mainly use coded data. Conditions, complications and social factors described in free text slip through the analysis.

From source to variable
VariableContextCodeSource
Type 2 diabetesConfirmedSCT 44054006Consult note · p.1
Diabetic nephropathySuspectedSCT 127013003Consult note · p.2
Diabetic retinopathyRuled outSCT 4855003Ophthalmology report · p.1

Each variable remains linked to its original passage.

Clinical knowledge, largely held in documents
Structured data
20 %
Directly usable in information systems.
Document-based data
80 %
Reports, PDFs, clinical notes and free text.
The value for your CDW

More clinical information. A more complete view. More relevant analyses.

Enrich variables

Information drawn from documents, beyond coded data.

Preserve clinical context

Affirmed, negated, suspected: every mention read for what it is.

Quality, documented

Precision, recall and coverage measured on your corpus.

Maintain traceability

Each variable tied to its supporting passage.

Semantic interoperability

Exchanging data is not enough. Its meaning must be shared too.

Formats enable exchange. Terminologies align concepts. CLARYS Intelligence adds clinical context.

  • FHIR for exchange. The formats your ecosystem already uses.

  • Terminologies for consistency. Expressions aligned with reference terminologies such as SNOMED CT.

  • Intelligence for context. Negation, suspicion and traceability preserved.

  • Longitudinal for pathways. Dated information, linked to events.

One concept, several languages
FRdiabète de type 2
ENtype 2 diabetes mellitus
DEDiabetes mellitus Typ 2

Reference concept

SNOMED CT 44054006ICD-11 5A11
Use cases

Enrich your data. Broaden your horizons.

Study feasibility

Find eligible patients in clinical reports.

Enriched cohorts

Inclusion criteria enriched by free text.

Real-world data

Better-characterized comorbidities, treatments and events.

Longitudinal analyses

Event sequences ready for care pathway studies.

Your warehouse, our intelligence

Add clinical understanding to your existing infrastructure.

CLARYS Intelligence makes previously out-of-reach information usable for research, population characterization and care pathway analysis.

  1. 1

    Access the content

    Through your systems and preparation pipelines.

  2. 2

    Extract and contextualize

    Concepts, context, terminologies.

  3. 3

    Produce traceable data

    Linked to their sources, ready for validation.

  4. 4

    Power your use cases

    Warehouses, research, analytics applications.

The same warehouse. Broader clinical knowledge.

Integration architecture

Your CDW organizes the data. CLARYS Intelligence extends its clinical reach.

Accessible clinical content

Reports
Letters
Clinical notes
Your data preparation pipelines
CLARYS IntelligenceExtraction · Context · Normalization · Traceability
Your existing CDWEnriched, traceable variables, integrated into your models
ResearchFeasibility and real-world studies
CohortsStructured and text-based criteria
ApplicationsReusable clinical context

CLARYS Intelligence sits between your content and your warehouse, without replacing your pipelines.

Key takeaways

Most clinical information sits in documents. Make it usable.

  1. More variables

    Conditions, complications and social factors drawn from reports, beyond coded data.

  2. Context preserved

    Affirmed, negated, suspected: every data point keeps its meaning and its source.

  3. Your CDW stays yours

    CLARYS Intelligence fits into your existing pipelines without replacing them.

Getting started

One clinical question. Data to rediscover. Value to measure.

A clinical question, a representative corpus: let’s identify what your structured data misses and measure the benefit for your CDW.