Product

One platform transforms raw notes into intelligence

Three systems: a Data Engine that structures free text, Striata for analysis and reports, and Voyager for the questions nobody thought to ask.

Amorphous Data Engine

Reads clinical notes in any European language and writes structured, coded records. Typos, abbreviations and local shorthand included.

Data Engine 01 / 06

Segmentation & translation

Splits the document into anchored sections and translates where needed, preserving the original source text.

  1. Segmentation & translation

    Splits the document into anchored sections and translates where needed, preserving the original source text.

  2. Extraction

    Parallel agents identify clinical entities by type, capturing the exact words and their position in the note.

  3. Merge

    Brings the extractions together, removes duplicate spans and resolves conflicting entity types.

  4. Attribute enrichment

    Adds timing, negation, certainty and clinical context, alongside details such as dose and severity.

  5. Graph construction

    Connects entities to one another: a treatment to its indication, a finding to its evidence, a condition to a body site.

  6. Coding

    Maps each entity to the appropriate clinical ontology, producing structured records linked back to the source.

Captured Entities

Conditions, medications, procedures, symptoms, family history, allergies, care plans and more. Each entity carries detailed attributes.

“Admitted for acute CHF exacerbation.”

Congestive heart failureICD-10 I50.9
Timing
Current
Certainty
Confirmed
Status
Active
Severity
Acute

Global Standards

Source termType 2 diabetes mellitus
SNOMED CT44054006ICD-10E11ICD-10-GME11.90

Coded to SNOMED CT, ICD-10, RxNorm, LOINC and ATC. National and custom ontologies can be added — a national cancer ontology, for example.

Delivered asFHIR R4OMOP CDM

Amorphous Striata

Ask questions in plain language. Striata selects cohorts, tests associations, maps care pathways and checks guideline conformance, then writes a report with editable charts you can forward.

Clinical codes and SQL can be inspected. Reports detail the methodology and data used.

StriataQuestions

Ask Striata

Operational, financial, and clinical questions — answered in minutes.

Striata handles questions across clinical domains

01

Haematology

What are the primary bottlenecks in the acute myeloid leukemia pathway that push length-of-stay above the national average?

02

Cardiology

Which patient cohorts drive readmissions for CHF, and which interventions correlate with improved outcomes without increasing cost?

03

Orthopaedics

How does cost-effectiveness of minimally-invasive versus open hip replacement compare, adjusted for post-op rehab duration?

Amorphous Voyager

An autonomous health-data analyst that works without being asked. Voyager studies your hospital’s context, proposes hypotheses and investigates your data, then sends each finding to the person responsible.

  1. 01

    Studies your context

    Learns what matters to your hospital from national guidelines, hospital strategy, and policy and planning.

  2. 02

    Proposes hypotheses

    Generates clinical and operational questions worth investigating, guided by your priorities and data.

  3. 03

    Runs analyses

    Tests those hypotheses against your structured clinical records using statistical analyses.

  4. 04

    Investigates each signal

    Asks follow-up questions and runs further analyses until the finding is clear.

  5. 05

    Delivers to the right person

    Routes the finding and supporting evidence to the person responsible for the problem.

Different tools do different jobs

Amorphous combines clinical data extraction and analysis. See how that compares with tools built for individual parts of the workflow.

Clinical data and analysis capabilities compared
CapabilityAI document assistantsQ&A over notesClinical NLP APIsText extraction & codingCohort & BI toolsAnalysis of structured dataAmorphousData Engine + Striata
Turn notes into coded clinical data
Extract contextual attributes
Trace clinical findings to source notes
Build cohorts using clinical criteria
Compare outcomes across cohorts
Ask questions in plain languageVaries
Output FHIR R4 and OMOPVaries
Support EHDS readiness

Signature to Striata

  1. Week 1Technical workshop

    Priority tasks, data sources, success metrics.

  2. Week 2Integration

    A secure connection to your records through your existing EMR.

  3. Weeks 3–4Pilot

    Your teams complete real tasks on selected cohorts.

  4. Month 2Full deployment

    The Data Engine structures continuously and Striata rolls out across departments.

Deploy Striata now.Let's talk.

For partnerships

hello@amorphous.health
Get in touch