Wearable impedance cardiography
Thorio is building an AI-based wearable ICG patch that measures cardiac output continuously — beat by beat, in motion, outside the hospital. The single most diagnostic number in cardiology, finally available to someone who isn't lying in a hospital bed.
The problem
Heart rate tells you the heart is working. It does not tell you how well.
Apple Watch, Whoop, Oura and Garmin all read electrical or optical signals — PPG and wrist ECG. Neither can produce stroke volume or cardiac output. That is a physical limitation of the sensor, not a software gap.
Impedance cardiography exists — on hospital carts, or in sedentary patches built for patients lying still. Portable rigs need a belt-worn box and six tethered leads. None survive real movement.
Heart failure decompensation is signalled by falling stroke volume and rising thoracic fluid weeks before symptoms appear. Athletes cross anaerobic thresholds before heart rate reflects it. Nobody is capturing either.
The device
A lightweight, sweat-sealed chest patch with an integrated four-electrode tetrapolar array. It passes a low-amplitude, high-frequency current across the thorax and reads how impedance changes as blood is ejected — turning the chest into a continuous hemodynamic sensor.
Highlighted — the four no consumer wearable can produce at any price
Ultra-light mechanics designed to travel with the chest wall, minimising skin drag and motion artifact.
The chassis maintains constant skin-electrode impedance contact under maximum perspiration.
On-device filtering plus AI analysis converts raw impedance into meaningful trends and alerts.
How it works
ECG and PPG wearables track electrical impulses and pulse waves — proxies. ICG extracts the mechanical performance of the myocardium itself.
A low-amplitude, high-frequency alternating current is passed across the thorax through a tetrapolar electrode array.
As aortic volume fluctuates with each heartbeat, thoracic impedance shifts. The patch samples that shift continuously.
Standardised Kubicek modelling converts impedance change into stroke volume and cardiac output, cycle by cycle.
AI models turn the raw hemodynamic stream into trends, thresholds and alerts the wearer can act on.
Defensibility
ICG is a known physical principle. The unsolved problem was making it survive movement in a form factor people will actually wear.
Our answer — IMU + on-chip DSP
A 9-axis inertial measurement unit models body movement in real time. On-chip digital filters subtract it from the impedance stream, preserving clinical-grade ICG metrics through sprint and maximum-shear states.
Every day the device is worn, it generates real-world hemodynamic data no competitor holds — and that data trains the next generation of models.
Where we are
The development kit is validated and the full signal acquisition and processing solution is built. We are testing on real users now.
Team
Get in touch
Hardware manufacturing partners. Clinical and regulatory advisors. Sports science and cardiology validation sites. If that is you, write to us.
shkhasher@gmail.com