According to Statista, the number of IoT devices used across the globe in 2021 was 21.5 billion, with a great number applied in healthcare and patient care facilities. As of 2018, there were about 406 million connected mobile IoT devices with 20.8% projected average annual growth. The global IoT market in healthcare is prognosticated to reach $176.82 billion by 2026 growing by 2.5 times compared to 2020. Wearables are used for fitness purposes, vital signs monitoring, tracking the condition of cancer patients to improve the quality of their lives, as well as for knowing when to hospitalize patients or admit them to the ICU. In this article, we’ll consider how the wearable market is advancing and what possibilities custom medical software offers to healthcare providers.
Forecasts for the medical wearables market
According to the Market Study Report, the wearable medical devices market is the second-largest market after consumer electronics. The highest growth in this segment is forecasted to occur by virtue of devices for fitness tracking and healthcare monitoring. The extensive use of remote patient monitoring devices must also contribute to market growth.
Such groups of IoT devices as wearables for diagnosing infectious diseases, providing TeleMedicine services, and building predictive models for national epidemic strategies deserve a special mention.
The report states that the COVID-19 pandemic boosted people’s interest in these devices because more measures need to be taken to prevent or detect infection. The industry players have updated their patient monitoring apps and devices, especially for combating the pandemic. According to the report, the CAGR in the body and temperature monitors subsegment is expected to reach about 19.4% within the period of 2021-2027. The usage of medical devices at home will increase on average by 23.3% per year over this projection period due to the application of home healthcare devices for remote patient monitoring.
Categories of wearable devices
Wearables are devices that are fixed to the body of an end-user in order to read their health and activity indicators. Today, there is a great variety of such devices – from special helmets and glasses to smart socks and shoes. Specialists distinguish ten categories of smart devices, depending on the way they are attached to the body. Each of these devices can be fitted with sensors, network ports, cameras, data processors, and other components needed to track users’ daily behavior.
By the type of sensors, most wearable devices applied in healthcare and for fitness purposes can be divided into the following groups, based on:
- Photoplethysmography devices (PPG) – scanning through the skin with a light beam,
- Electrical impedance and EKG – electrical impulse sensors,
- Biosensors – sensors using samples of body fluids or secretions.
Today, many of these devices are used for diagnosing and building predictive models for infectious disease dynamics, including COVID-19.
Wearables for diagnosing COVID-19
From the very start of the pandemic, there have been attempts to monitor and detect the foci of coronavirus infection by using modern digital and mobile technologies.
For example, to track the incidence and geographic boundaries of the disease, search query data are used, with the help of improved Google Flu Trends (GFT) algorithms like Weibo COVID-19 Trends (WCT). The algorithms also use data from social media, creating incidence statistics across regions and age groups by analyzing private messages of users, search engine requests about treatment, or direct answers to questions automatically generated by a neural network. This method makes it possible to predict the disease’s development but runs into a problem of discerning positive and false-positive cases.
In some countries, special software installed on mobile devices is widely used for identifying contacts of possible virus carriers and creating tracking maps to comply with the required distancing and hospitalization measures (for example, Singapore’s Smart Nation initiative).
Among all the tools for early diagnosing and predicting COVID-19, wearables provide the highest data accuracy, primarily those that measure heart rate, body temperature, and sleep duration (these are associated with the common symptoms of the infection). In particular, an average adult with COVID-19 has a respiratory rate of around 20 breaths per minute at rest, a body temperature of around 38°C, and a resting heart rate of more than 100 beats per minute. You can obtain this data with a high degree of reliability by using certified medical wearables.
Models for early diagnosing COVID-19
Predictive models based on combined data about resting heart rate, sleep duration, and physical activity are mostly used for the early diagnosis of COVID-19 symptoms. According to Scripps Detect Study, using data from smartwatches and fitness trackers allows for the creation of models with a prediction accuracy of up to 80% when resting heart rate data is combined with sleep duration parameters and self-reported symptoms. The method is based on spotting heart rate increase of an average of 8.5 beats per minute for each additional 1°C of body temperature.
The Warrior Watch Study also considered the possibility of using data on heart rate variability measured with an Apple Watch for early detection of COVID-19. The cardiologist Eric Topol, MD states:
“What’s exciting here is that we now have a validated digital signal for COVID-19. The next step is to use this to prevent emerging outbreaks from spreading.”
Using dedicated neural network algorithms allows for the creation of a predictive model that detects resting heart rate anomalies during the entire course of the infection (7 days after the common symptom expression and 21 days after their cessation) in 92% of infected patients. In 56% of cases, the model helps to identify cases of infection even at the pre-symptomatic stage.
According to the UCSF Preliminary Study, using wearables that regularly measure body temperature helps with early detection of COVID-19 in patients. This includes Oura Ring, a device from a Finnish startup that is put on the user’s finger and interacts with a mobile app. By regularly measuring body temperature and tracking its subtle fluctuations, you can determine a static model that is informative enough even if the disease is asymptomatic.
One more example of using wearables for the early diagnosis and treatment of COVID-19 is ultrasonic technologies for measuring lung activity. In particular, Novosound’s wearable device for lung monitoring allows clinicians to remotely monitor a number of patient parameters. Unlike such dedicated devices as X-ray and CT machines, it doesn’t require the involvement of a large number of specialists.
Using wearable data for building predictive models is only possible if there is the maximum level of medical information exchange through cloud technologies and mass use of EHRs. That is why countries like Israel are leading the way in combating COVID-19.
Therefore, not only wearable characteristics, their sensors’ sensitivity, and monitoring consistency but also the use of custom healthcare software, EHRs, cloud technologies, and advanced algorithms of neural networks processing arrays of statistical data are crucial for early diagnosis and prediction of COVID-19 and other febrile infections.