More Accurate
Mission-critical use cases (defense, medical identity, payment validation) cannot tolerate the false-accept rates of conventional biometrics.
ParallaxA RESEARCH PRESENTATION BY REYNOLDS DELGADO, MD
HF QRS as a
Biometric.
A signal from inside the heart, unique to every individual, resistant to lift attacks, and originating only from a living heart.
150–250 Hz
Signal Frequency Band
Aliveness
Required by Design
THE NEED
We need better
biometrics.
As digital identity becomes the gateway to every consequential transaction in modern life, the biometrics we rely on are no longer adequate. Mission-critical applications demand higher accuracy, attestable trust, and a measurement that scales to consumer hardware.
Attestable Trust
Authentication has to satisfy regulators, auditors, and adversarial review, not just succeed at the prompt. The trait should be one a counterparty can verify without taking the user's word for it.
Genuinely Practical
Whatever the trait, it has to be measurable in milliseconds on consumer hardware. Anything slower or bulkier never reaches the point of payment, the door, or the cockpit.
THE TECHNOLOGY
What is
High-Frequency QRS?
The QRS complex is the sharp deflection in an electrocardiogram produced when the heart's ventricles contract. Embedded within it, in the 150 to 250 Hz band, is a far denser signal: High-Frequency QRS. It is the electrical fingerprint of the Purkinje system, the branching network of fibers that drives ventricular depolarization.
The Purkinje system itself is a fractal: self-similar at differing scales. Purkinje anatomy varies markedly between individuals, and HF QRS captures that variation.
ORIGIN
How the idea
arrived.
Over a decade ago, as a research collaborator with NASA Johnson Space Center's human space flight physiology laboratory, I encountered HF QRS as a tool for detecting heart disease: cardiomyopathy and ischemia. The signal stayed with me as something extraordinarily dense, structured, and revealing.
Two years before this presentation, the idea came to me in an unremarkable moment: I was trying to open my phone. It struck me that the HF QRS signal (this picture of the Purkinje system, a tree-like fractal) was already a unique identifier. Self-similar at differing scales, individually different in every person.
No two individuals are the same.
Which means the same signal we use to detect ischemia in a hospital can, in principle, authenticate a person at the point of payment, the door of a secure facility, or the steering wheel of a car, with one essential property no other biometric offers: it requires a living heart.
THE FRAMEWORK
Seven requirements,
measured.
The biometrics literature converges on seven properties any candidate trait must satisfy. Five are met by the basic physics of HF QRS. The remaining two, distinctiveness and persistence, are addressed by the published evidence in the sections that follow.
Universality
Every individual in the target population should possess the trait.
Every living person has a heartbeat. Coverage is intrinsic.
Distinctiveness
The trait must sufficiently differentiate between any two persons.
Supported anatomically by the individual variation of the Purkinje conduction network; remains the empirical question this program is designed to quantify.
Persistence
It must be sufficiently invariant against the matching criterion over time.
NASA-led research shows HF QRS waveform features remain stable across months and years (see below).
Collectability
It must be easily obtainable or measurable with practical hardware.
Two points of skin contact are sufficient: a finger on each hand, or wrist and opposite finger.
Performance
High recognition accuracy and speed under varied operational conditions.
The signal is robust given the absence of high-frequency electrical, magnetic, or motion artifact.
Acceptability
Wide public acceptance; the measuring device must be harmless.
The measurement is passive, harmless, and analogous to widely accepted ECG sensing.
Circumvention
Spoofing the trait through fraudulent methods should be difficult.
An internal, live signal is extremely difficult to lift from a surface or reproduce from a static recording.
Artifact analysis
Motion and muscle artifact, the chief sources of noise in surface ECG, are predominantly low-frequency. HF QRS lives at 150 to 250 Hz, well above that range. The frequency separation gives the signal a natural immunity to the most common forms of contamination.
Framework source: Jain, Ross, Prabhakar, An Introduction to Biometric Recognition, IEEE Trans. Circuits Syst. Video Technol. 14(1), 4 to 20, 2004.
PUBLISHED EVIDENCE
What the
literature shows.
Persistence and distinctiveness are the load-bearing biometric questions in the framework above. Three published papers anchor the case for HF QRS as a stable, sensitive, person-specific signal.
NASA technical report · persistence
Month-to-Month and Year-to-Year Reproducibility of High Frequency QRS ECG Signals
Niles J. Batdorf · Alan H. Feiveson · Todd T. Schlegel
National Space Biomedical Research Institute · NASA Johnson Space Center · 2006
“We conclude that 12-lead HF QRS electrocardiograms are sufficiently reproducible for clinical use.”Conclusion, NASA Technical Reports Server, document 20060023352.Read the original report
Peer-reviewed literature · corroborating persistence
Temporal stability of the high frequency QRS waveform in normal subjects
Ary L. Goldberger · V. Bhargava
Journal of Electrocardiology, 14(2), 153 to 157 · 1981
In serial HF QRS recordings of 10 normal subjects taken approximately 100 days apart in lead aVF, the authors reported a high degree of reproducibility of waveform amplitude and morphology within each individual.Summary of the paper's central finding. DOI 10.1016/S0022-0736(81)80050-7. PubMed PMID 7276784.Read the original paper
Peer-reviewed literature · physiological sensitivity
High frequency QRS electrocardiography in the detection of reperfusion following thrombolytic therapy
T. Aversano · B. Rudikoff · A. Washington · S. Traill · V. Coombs · J. Raqueno
Clinical Cardiology, 17, 175 to 182 · 1994
HF QRS amplitude in the 150 to 250 Hz band increased significantly within roughly an hour and a half of successful coronary reperfusion in 23 of 25 patients, while remaining flat or declining in patients whose thrombolysis failed.Summary of the paper's primary finding. Demonstrates that HF QRS responds to fine physiological change in real time. Biometric entropy is a separate question outside the scope of this paper. DOI 10.1002/clc.4960170407.Read the original paper
BEYOND THE BASICS
Properties no other
biometric offers.
HF QRS satisfies the seven baseline requirements, but it also brings a set of properties that conventional biometrics structurally cannot.
Aliveness required
The signal exists only while the heart is beating. Authentication implicitly verifies the user is alive and present.
Anatomically embedded
The Purkinje conduction network is internal anatomy, not a surface trait. It is not stored on the body, photographed, or surface-readable.
Multi-signal validation
Pairs naturally with bioimpedance, skin temperature, and other live-body signals for layered verification.
Analog-domain processing
Can be processed entirely in the analog domain on-device, with software resident locally; no waveform need leave the sensor.
Active probing supported
Supports a sub-threshold electrical primer ping to actively elicit and validate the signal in real time, within published safety thresholds for surface stimulation. Research-stage.
APPLICATIONS
Where this
applies.
Anywhere identity matters and aliveness is non-negotiable, HF QRS is a candidate. The hardware is modest; the trust it produces is not.
Wearables & Phones
Watches, phones, computers, and other wearables that pair to nearby devices.
Payment Validation
Authorize transactions with a biometric that cannot be lifted, copied, or spoofed.
Medical Identity
Verification of patient identity at the point of care, where errors carry clinical consequences.
Operator Certification
Steering-wheel authentication for vehicles and machinery where the operator must be a specific, qualified individual.
Physical Access
Door locks and secure facility entry where presence and aliveness must be proven, not assumed.
Mission-Critical Roles
Military, defense, and critical-infrastructure scenarios that demand the highest certainty available.
PATENTED TECHNOLOGY
Protected by
international patents.
The use of high-frequency QRS in biometric identification is protected by granted patents in the United States and the European Union, with additional coverage across Australia, China, and Canada.
United States
Patent No. 10,595,753
High frequency QRS in biometric identification
Granted: March 24, 2020
View patent details European Union
Patent No. EP3860441B1
High frequency QRS in biometric identification
Granted: June 11, 2025
View patent details GET IN TOUCH
Contact the
inventor.
Reynolds Delgado, MD
Medical Director, Mechanical Support Devices in Heart Failure · Texas Heart Institute
Founder, Procyrion, Inc. · Founder, Parallax Biometrics LLC
Dr. Delgado is a practicing cardiologist and a recognized leader in the field of heart failure. At the Texas Heart Institute he has directed the mechanical-circulatory-support program since 2003. He is the founder and Chief Medical Officer of Procyrion, Inc., the cardiovascular medical-device company developing Aortix, a catheter-deployed circulatory support pump for heart failure patients.
He is co-author of two New England Journal of Medicine articles that contributed to the FDA approval of the HeartMate II LVAD, principal investigator of the ORQIS MOMENTUM trial, co-founder and president of the Houston Heart Failure Society, and a long-standing research collaborator with NASA Johnson Space Center's human space flight physiology laboratory.
He is the inventor of the heartbeat-biometric patents (US 10,595,753 · EP3860441B1) that anchor Parallax Biometrics LLC.