Physiological Signal & Device Data

When the monitor
leaves the ward,
the patient data
leaves with it.

Physiological signals from ICU monitors, ventilators, and wearable devices carry patient identity in device headers, alarm logs, and session metadata. SignalIQ™ removes the identity without touching the waveform.

EDF/HL7
Format support
Real-time
Stream mode
ICU / ward
Device classes
SignalIQ · ICU monitor · header processing
Device session metadata
Patient namestrippedstripped
Patient identifierpseudonymisedpseudonymised
Date of birthyear onlypseudonymised
Location identifierstrippedstripped
Care team referencestrippedstripped
Device identifierpseudonymisedpseudonymised
Session timestamppreservedretained
Sampling parameterspreservedretained
Signal channelspreservedretained
Waveform data (untouched)
Physiological waveform · preserved at native resolution
-0.12,0.08,0.21,0.45,1.23,0.89,0.34,0.12,-0.08,0.02,0.05,0.08,0.06,0.04,0.03,-0.01,0.02,0.04,0.07,0.12,0.45,1.21,0.91,0.38,0.14,-0.07,0.01...
The problem

The signal is science.The header is a breach report.

Physiological signal data is among the most valuable in medicine - AI models trained on ICU waveforms are transforming critical care. But every session file carries the patient's identity in device headers, admission fields, and equipment metadata that standard de-identification pipelines were never designed to handle.

PHI surfaces in physiological signal data
EDF header patient field
Name, DOB, gender, admission code
PHI surface
Clinical message wrapper
Patient identity in the message header
PHI surface
Device session metadata
Patient ID, bed, nurse station, ward
PHI surface
Equipment serial numbers
Linkable to hospital asset register
PHI surface
Alarm log patient references
Patient name in alarm event records
PHI surface
Waveform signal data
Clinical signal - contains no direct PHI
no PHI
SignalIQ addresses all five PHI surfaces. Waveform data is never modified.

SignalIQ separates signal from identity at the format level. Waveform data, sampling parameters, and channel configurations are preserved with full precision. Only the fields that identify the patient are modified.

Device coverage

Every device class. Every signal format.

SignalIQ is built around the signal formats that actually come out of clinical devices - not an idealised standard.

ICU monitoring

Devices
  • Major ICU monitors
  • Bedside monitoring systems
  • Central station feeds
  • Multi-parameter monitors
Signal types
ECG (multi-lead)SpO₂Invasive BPEtCO₂Temperature
Formats
Standard waveform exportsECG standardsClinical messaging

Wearable & ambulatory

Devices
  • Holter monitors
  • Patch ECG devices
  • Continuous glucose
  • Pulse oximeters
Signal types
Single-lead ECGCGM glucoseSpO₂Accelerometry
Formats
Standard waveform exportsCSV exportProprietary binaryFHIR resources

Ventilators & respiratory

Devices
  • Hospital ventilators
  • Critical-care respirators
  • Anaesthesia workstations
  • Sleep / respiratory devices
Signal types
Flow / VolumePressure waveformFiO₂PEEP settings
Formats
CSV exportHL7Proprietary RS-232 logs
How it works

The SignalIQ™ pipeline.

Signal fidelity is never compromised. Only identity changes.

  1. 01

    Signal Ingest

    Waveform file or real-time stream received from device, gateway, or clinical system

  2. 02

    Format Parse

    File format identified. Signal and metadata layers separated. Channel map built.

  3. 03

    Header Strip

    All patient-identifying metadata removed or pseudonymised. Signal data untouched.

  4. 04

    Policy Check

    Release rules and recipient allowlist enforced. Research DUA verified.

  5. 05

    Clean Output

    De-identified file forwarded or streamed. Tamper-evident ledger entry written.

Technical specifications

Precision for the signal. Precision for privacy.

SignalIQ preserves every sample, every timestamp, every channel with full numerical precision. The de-identification is applied only to the metadata layer.

Primary formats
Standard waveform formats · physiological signal exports · device-native binary
Stream mode
Real-time clinical message streams · device streams · REST endpoints
PHI surfaces
EDF header fields · clinical message headers · session metadata · device serial · alarm logs
Signal preservation
All samples, timestamps, channel configs, and calibration data preserved exactly
Sampling rates
High-rate clinical signals supported · configurable downsample on output
Throughput
Real-time streaming · session file processing
Deployment
On-prem agent
Compliance
OAIC APP 11 · Privacy Act 1988 (Cth) · ISO 27001 alignment · HREC-exportable audit

SignalIQ™ is in development.

Join the waitlist. ICU research teams, clinical AI vendors, and biomedical engineering departments get first access.