Kalabre is currently seeking funding to complete the final design phase, however you can subscribe to our mailing list and we will let you know when it is available.

Kalabre is a medical device technology based on the discovery that sublingual and pulmonary data, acquired from an individual upon waking and correlated by a unique algorithm, can determine the onset of an illness, days before symptoms or clinical signs including fever.

Tap or hover over the blue icons

Why Early Detection?

Antibiotics and antiviral drugs are most effective at the onset of infection, and most treatments are more effective the earlier they are started. Kalabre is used immediately upon waking and provides a colorful easy to understand graphic report, which can be sent to both the patient and their physician.

Kalabre’s Graphic Display

While you can use your desktop or mobile app to view results, you can also keep track of your tests with the graphic display on your Kalabre.


Kalabre early detection monitor

Kalabre's Front Lights

Kalabre's graphic display

The left column's colored lights show that you may have a non-respiratory infection. The middle column's lights say that you may be having a respiratory episode, which means that you may have trouble breathing. The right column's lights show that you may be developing a respiratory infection.

The Early Detection Algorithm

A proprietary algorithm correlates basal metabolic data and a lung function percentage (PF%). Basal metabolic data is acquired near the sublingual artery, and PF% is measured by blowing through a scroll/turbine configuration.

Kalabre Exploded View

Example Patient Graph of a Viral Infection

Actual graph of a patient on the Kalabre system, experiencing a COVID-like viral infection.

Graph Display

Pre-clinical Trials

Pre-clinical trials using the algorithm were completed shortly after the SARS coronavirus epidemic at five health clinics in Hong Kong, and at Tianjin Haihe Hospital in China. An IRB Approved Research Study was also conducted in a church community in the United States.

The data from all studies was combined and analyzed, and the accuracy of the algorithm was determined to have a sensitivity of 96.7%, and a specificity of 96.8%.

We continue to refine the accuracy of the algorithm through the discovery of pre-symptomatic signs and ongoing research of non-identifiable aggregate patient data.