Atrial Fibrillation (AFib) are characterized by irregular heart rhythm, known as arrhythmia, and can lead to fatigue, dizziness, or even strokes. Five million Americans suffer from AFib, and the Center for Disease Control estimate that 12 million people will be diagnosed with Atrial Fibrillation in 2030.
The main issue with AFib is not to treat patients- blood thinners can significantly decrease the risk of strokes associated with AFib– but the diagnosis.
In recent years, smartwatches such as the Apple Watch, the Samsung Galaxy or the Fitbit Sense, and Charge 5 integrated sensors and algorithms to detect arrhythmia. The CheckMe DuoEK S or the AliveCor Kardia also offer such features. The British National Institute for Health and Clinical Excellence (NICE) even recommend the latter as a home device to help detect Atrial Fibrillation in suspected patients.
Recently the results of the eBRAVE-AF trial (NCT04250220) were published in the Journal Nature Medicine and concluded that “digital screening technology provides substantial benefits in detecting AFib compared to usual care.”
Even if these devices appear accurate and reliable for helping detect arrhythmia, they require using specific hardware and actively launching an analysis cycle.
Cardiokol or the device-free AFib detection company
The Israeli startup Cardiokol may have found the ultimate way to identify unintrusively AFib in what they call “revolutionary affordable, scalable, long-term and age-friendly monitoring solutions.”
The company developed a proprietary algorithm that will extract acoustic cues from the users’ voices. In an abstract initially published for the 67th Annual Conference of the Israel Heart Society, Cardiokol presented the results of preliminary studies conducted on 513 recordings in 86 patients. After training the algorithm on 34 patients, the system, just by listening to the patient’s voices, could distinguish pathological Atrial Fibrillation from normal Sinusal Rhythm with high specificity and sensitivity.
Cardiokol also recently published an article in the Journal of Cardiovascular Electrophysiology showing the results of another study conducted on 158 patients and showing an AFib detection sensitivity of 95% and 82% specificity.
The principle underlying the technology is that acoustic signals from the voice pattern can be used to estimate the heart rate, hence detecting abnormal heartbeat patterns.
Cardiokol aims to use the technology to screen the older population that may not be comfortable with the AFib wearable technologies but will somehow use a smartphone or a landline.
Over-the-Phone voice analysis is a very innovative way to detect atrial fibrillation in non-tech savvy but at-risk populations.
The algorithms and principles supporting Cardiokol technology are patented, and the company was recently selected as one of the 12 finalists in the MEDICA Fair startup competition.
To Conclude
The technology and principles supporting CardioKol are innovative. They offer new ways to extract heart rhythms from all the voice recording devices we are all surrounded by, from connected cars to mobile phones and even landlines. In addition, cloud-based processing is very innovative and answers the needs of an aging at-risk population that does not want the burden of having to carry an EKG device or smartwatch but could benefit from frequent AF detection.
CardioKol is a company to follow in the coming years, which may have found a unique way to keep us healthy. Of course, the regulatory road will still be long before the algorithms listen to our hearts, but such companies make the future of Digital Health exciting and worth following.
