Home / Our Algorithms

Breathing Meets AI

Circadia has developed a suite of signal processing and machine learning algorithms, designed to provide clinical decision support to physicians. Continuous analysis of respiration data allows for early detection of patient deterioration. Circadia’s proprietary algorithms alert physicians of adverse events such as bradypnea, tachypnea, and in future also COPD exacerbation (expected early 2020).

Radar Technology

Our device's low power radar locks onto a patient's position and monitors their body motion to derive respiratory rate. Even when the patient has a bed partner, the device possesses the capability to track two people, Currently identifies and tracks the nearest person.

Sleep Analysis Algorithm

Our proprietary algorithms predict sleep stages with an accuracy approaching that of gold-standard PSG and exceeding that of published performance data for existing wrist-worn, consumer sleep tracking devices. See our white paper for details.

Respiratory Analysis Algorithm

Our proprietary algorithms can accurately extract respiratory rate from our devices' raw radar data. Our C100 System is currently being validated against gold-standard PSG respiratory rate data in a clinical study forming part of an FDA 510(k) submission.

Clinical Validation Studies

Performance Evaluation of the Circadia Wireless Respiratory Monitor and Circadia Sleep Analysis Algorithm
Circadia Contactless Technology as a Tool For Predicting Acute Clinical Deterioration
Respiratory monitoring study: the performance validation of a non-contact breath rate electronic device