01Overview
Respiratorydeviation present
Inflammatorydeviation present
Renalwithin personal range
Biological state, deviations and evidence sufficiency in one patient view.
Replica / Biological Twin Engine
The biological patient model beneath healthcare AI.
Replica integrates longitudinal clinical data into an individual representation of biological state, trajectory and uncertainty that evolves through time.
Inside Replica
One patient model. Multiple views of state and evidence.
A clinical console surfaces biological state, evidence sufficiency, longitudinal trends, relationships across systems and uncertainty.
02Systems map
RespiratoryPatient stateRenalInflammatory
A dynamic representation of biological state and relationships across systems.
03Modules
Respiratory4 observations
Renal2 observations
EndocrineEvidence unavailable
The underlying observations, reference context and missing evidence for each system.
04Insights
Investigation grounded in evidence
Clinical question What changed, and what evidence is still missing?
Source · rationale · limitationQuestions derived from the record, linked to source evidence, rationale, limitations and next investigation.
05Trends
longitudinal evidence
Longitudinal measurements interpreted within the wider patient and system context.
06Couplings
Respiratory
↔Inflammatory
ACTIVE · BIOLOGICAL CONTEXTInspectable relationships across systems presented as biological context, not as hidden scoring.
Illustrative interface using synthetic data. Replica remains in research development and is not a diagnostic system.
Architecture
A persistent foundation for focused clinical applications.
Applications address specific use cases while Replica maintains a coherent representation of the patient across evidence, systems and time.
Application
Focused clinical and research workflows
Anali-X and future applications for specific clinical domains
Interpretation
Evidence, provenance and confidence
Inspectable outputs linked to supporting observations
Biological state
An individual patient model through time
Observed facts · estimated state · trajectory · uncertainty
Longitudinal data
Clinical context aligned over time
Events · measurements · interventions · care context
Patient state journey
From clinical records to a state linked to evidence.
01Collect
Fragmented records
02Align
One patient timeline
03Map
Biological context
04Reconstruct
Current patient state
05Explain
Trajectory + evidence
System view
One model across biological systems, evidence and time.
Replica provides a reusable patient representation for clinical and research applications built above it.
Clinical teams
Researchers
Life science partners
questions
INTERACTION LAYER
FOCUSED APPLICATIONSApplications for specific clinical domains, built on one patient model
Anali-XICU research programme
contextual query
BIOLOGICAL MODEL
REPLICABiological state of the individual patient
Current stateWhat the evidence supports now
TrajectoryHow the state is changing
UncertaintyConfidence and unresolved evidence
ProvenanceSources supporting each interpretation
state reconstruction
TIME LAYER
LONGITUDINAL ALIGNMENTOne patient timeline linked to evidence
Clinical recordsevents · diagnoses
Laboratorybiomarkers · trends
Vital signscontinuous signals
Medicationsexposure · response
Imaging metadatafindings · context
Wearables & contextbaseline · behaviour
Interpretable output
Observation, inference and trajectory remain distinct.
Observed
A recorded measurement, event or fact, linked back to its source and time.
Estimated
A modelled interpretation of current biological state, clearly separated from fact.
Trajectory
A bounded view of plausible change, accompanied by uncertainty and supporting evidence.
Designed for clinical scrutiny
Every interpretation retains its context.
Individual patient
The model describes this patient’s state and history, not only a population average.
Longitudinal
Order, recency and change are essential parts of the representation.
Evidence provenance
Interpretations retain a visible relationship to the observations that support them.
Explicit uncertainty
Confidence, missing information and the boundaries of interpretation remain visible.
From engine to programme