Twitter
Advertisement

COVID-19 can be detected by smartwatches before symptoms show up, here is how

The method detected 20% of COVID-positive patients two days before the beginning of symptoms and 80% of cases on the third day of symptoms.

Latest News
article-main
COVID-19 can be detected by smartwatches before symptoms show up, here is how
FacebookTwitterWhatsappLinkedin

Smartwatches and other wearable technology like activity trackers have gained popularity in recent years due to their capacity to keep tabs on our health.

In the case of COVID, an increased respiratory rate (or breathing rate) has been shown to be a valuable indicator for early diagnosis. Photoplethysmography  , an approach of estimating respiratory rates, involves just a single point of contact and may be performed noninvasively (for example, your finger or wrist).

Also, READ: Future Apple iPhone models to have a USB-C port? New EU law may force Apple to ditch ‘iconic’ Lightning port

Ambient light, pressure, and motion may all interfere with a photoplethysmograph's accuracy. Most research into this method's potential for spotting COVID has thus concentrated on following participants while they sleep.

The electronics manufacturer Fitbit studied the night-time respiration rates of thousands of its customers to see whether this metric may assist in the identification of chronic obstructive pulmonary disease.

They discovered that some persons with COVID had at least one measurement of increased respiratory rate within a seven-day window (beginning one day prior to symptom onset or one day prior to a positive test for participants without symptoms).

Commercially available wearables may provide a non-invasive means to identify probable COVID infections and have them evaluated, since this was found in around one-third of symptomatic COVID sufferers and one-quarter of asymptomatic patients in this research.

Further research investigated whether or if a fitness tracker made by the US company WHOOP might be used to foretell the occurrence of COVID.

An algorithm was trained using data on respiratory rate and other markers of heart function from a cohort of persons with COVID.

The model was then applied to an independent sample of individuals, including both those with and without COVID but sharing comparable symptoms.

The method detected 20% of COVID-positive patients two days before the beginning of symptoms, and 80% of cases on the third day of symptoms, all based on respiratory rate during sleep.

Other forms of digital detection

Wearables are only one example of how digital technology might be used to monitor for the spread of the COVID virus. The widespread availability of devices equipped with high-quality microphones has paved the way for audio analytics.

Symptoms of COVID often include a hoarse or raspy voice when the virus attacks the upper respiratory system. It has been proven that a smartphone software trained on hundreds of audio samples from persons with and without COVID can correctly identify the presence of the virus in a person 89% of the time.

Tracking illness

The feasibility of using smart technologies and wearable devices to monitor persons with a COVID infection has also been investigated.

One group, for instance, monitored high-risk patients with COVID at home by taking their oxygen saturation, respiration rate, heart rate, and temperature every 15 minutes using an in-ear device.

Patients who would benefit from specialised medical attention were pinpointed with the use of this data, which was constantly checked by an experienced staff. During the early stages of the pandemic, it was suggested that cellphones may be used to detect hypoxia through the user's fingertip.

Some people with COVID who have more advanced illness develop hypoxia, or low oxygen levels in the tissues of the body, without ever showing any outward symptoms.

The widespread effects of COVID have also been mapped with the use of wearable technology. Instances when sleep patterns shifted during the epidemic were shown by data from hundreds of Fitbits (early in the pandemic people were generally sleeping for longer, for example).

An extra line of defence

Wearable and other technologies being tested for their ability to detect COVID depend heavily on artificial intelligence (AI) methodologies, especially machine learning and deep learning.

Effectively scanning enormous amounts of data in great detail, AI can find significant patterns in body signals to recognise the health state of interest.

But there may be practical limitations to these AI models, since patterns of biological signals might vary greatly within and across individuals. It's also important to remember that generic wearables aren't always built for real-time monitoring of infectious illness symptoms.

This suggests that there may be room for development in both the technology and the algorithms. In addition to the continuing research needed to solve these obstacles, we must also carefully examine any potential privacy problems related with the collection of biological data for this purpose.

However, digital technology such as wearables might serve as a second line of defence in the fight against COVID and other infectious illnesses.

(With inputs from PTI)

Find your daily dose of news & explainers in your WhatsApp. Stay updated, Stay informed-  Follow DNA on WhatsApp.
Advertisement

Live tv

Advertisement
Advertisement