Coming Soon · Join the waitlist
Cardiology Data Privacy

The waveform is clean.

The report header has never been.

Cardiology data flows from ECG machines, echo labs, and cath labs into AI analysis platforms every day - carrying patient identity in report headers, device fields, and HL7 wrappers that most de-identification pipelines never touch. CardioIQ changes that.

CardioIQ · 12-lead ECG report · header processing
Raw report header (Muse / Cardiostation)
Patient Name████████, ████████
Patient IDMRN-████████
DOB██/██/████
Ordering PhysicianDr. ████████████
LocationWard 4B, Bed 12
Study Date2024-08-14
Ventricular Rate72 bpm
PR Interval162 ms
QRS Duration86 ms
QT/QTc380/411 ms
After CardioIQ processing
Patient NameREDACTEDstripped
Patient IDPSEUDO-A4F2B1pseudonymised
DOB1962pseudonymised
Ordering PhysicianREDACTEDstripped
LocationREDACTEDstripped
Study Date2024-08-14retained
Ventricular Rate72 bpmretained
PR Interval162 msretained
QRS Duration86 msretained
QT/QTc380/411 msretained
Clinical measurements fully preserved · header PHI removed · waveform signal untouched38ms
The gap nobody talks about

AI vendors want your ECG data.
Your patients' names are in the header.

Cardiology AI platforms consume 12-lead ECGs, echo reports, and Holter readings to train and run their models. Most receive files exported directly from GE Muse, Philips Cardiostation, or DICOM-wrapped SCP-ECG - all of which carry patient identity in report headers, device fields, and HL7 message wrappers.

Typical cardiology export coverage
DICOM header fieldscovered
SCP-ECG report headermissed
GE Muse proprietary XMLmissed
HL7 V2 message fieldsmissed
Echo report PDF textmissed
Device serial numbersmissed
How it works

The CardioIQ pipeline.

Clinical measurement data is preserved in full. Only identity fields are removed.

01
Study Ingest
ECG, echo, Holter, or cath data received via DICOM, SCP-ECG, HL7, or file drop
02
Signal Isolation
Waveform and measurement data separated from header and metadata layers
03
Header Scrub
All patient-identifying fields across all report formats processed per profile
04
Policy Check
Recipient allowlist enforced. DUA compliance verified for each AI vendor.
05
Clean Output
De-identified study forwarded with clinical data intact. Audit entry written.
Capabilities

What CardioIQ covers.

Signal-preserving de-identification

CardioIQ separates identity from clinical signal at the format level. Waveform data, measurements, interval calculations, and rhythm interpretations are preserved in their original precision. Only patient-identifying fields are modified.

Format support
SCP-ECGDICOM WaveformGE Muse XMLPhilips XMLHL7 ORU^R01PDF reports

Multi-modality coverage

12-lead ECG, Holter monitor, echocardiography reports, and cath lab data all handled natively.

Echo report text scrub

Echocardiography PDF and structured reports parsed for PHI. Clinical findings fully preserved.

Device field handling

Device serial numbers, station IDs, and location fields processed - not just patient demographics.

Audit trail per study

Every de-identification decision signed and logged. Exportable for HREC and TGA review.

Technical specifications

Built for cardiology lab workflows.

CardioIQ integrates with your existing cardiology information system and AI vendor pipeline - transparent to clinical staff, automatic for compliance teams.

ECG formats
SCP-ECG · DICOM Waveform (MFER) · GE Muse XML · Philips Cardiostation XML · HL7 ORU^R01
Echo formats
DICOM Echo SR · Structured report PDF · XML export
Holter formats
EDF / EDF+ · HL7 · proprietary Spacelabs / NovaBay formats
Cath lab
DICOM SC / XA · hemodynamic XML reports
PHI surfaces
Report headers · device fields · HL7 wrappers · station IDs · referring physician · location
Throughput
Real-time · echo reports async · Holter bulk mode
Deployment
On-prem agent · CIS / ECG system direct integration
Compliance
OAIC APP 11, TGA SaMD, DICOM PS3.15, HL7 FHIR R4 de-id
Coming Soon

CardioIQ is in development.

Join the waitlist. Cardiology labs, cardiac AI vendors, and health networks get first access.