Wearables accelerate shift to personal health devices - Embedded.com

Wearables accelerate shift to personal health devices

The growth of wearables over the last decade has primarily been based on sports and fitness devices tracking general activity levels. This provided consumers a taste of what’s possible with a relatively inexpensive, easy-to-use device, but now consumers want more – more insights, more guidance, and ultimately they want to know how to live longer, healthier lives. The data behind these insights has become known as patient-generated health data (PGHD).

PGHD is driving the next phase of growth in wearables toward personal health devices. Accenture found in a seven-country, 8000-person survey that consumers’ use of mobile health apps more than doubled in 2 years to 33% and 1-in-5 survey respondents were asked by a doctor to use wearable to track their health. “Digital tools are empowering patients to take charge of their health and interact with the (healthcare) system on their own terms,” said Kaveh Safavi, M.D., J.D., who leads Accenture’s health practice globally. Frost & Sullivan expects the market for wearable technologies in clinical and consumer health to reach $18.9B by 2020.

Making this transition from tracking general activity levels to providing meaningful health insights will require many things, large and small, to happen across the wearables and healthcare industries, but four of the most important requirements for product designers in the near term will be:

  • Biometric sensing is now a “check-box” feature – Measuring not just how much the body is moving, but how the body is responding to a person’s daily activities

  • Accuracy is critical – Achieving clinical-grade accuracy in devices that people actually want to wear for long periods of time

  • Consumers are demanding form-factor diversity – Maintaining flexibility to integrate the technology in a variety of different devices and form factors

  • Data integration must be seamless – Wearable data needs to get to the right place at the right time in clinical workflows

Let’s look at each of those in more detail.

Biometric wearables
You are starting to see more biometric wearables, primarily using optical heart rate monitoring technology, on the market for a reason: biometric wearables provide more insights into how the body is responding to a person’s activities. Counting steps is no longer enough, because it only tells you how much a person has moved. The body’s reaction to that movement varies significantly by fitness level – 10,000 steps per day puts significantly less strain on a highly trained marathon runner than it does on a couch potato.

Biometric wearables not only offer insights on a person’s overall health and fitness levels, but when used over time show trending data and pattern recognition that can provide guidance and recommendations for how to improve one’s overall health. For example, if a regular trend is identified that the user’s hear rate stays elevated for too long following exercise (a well-known “silent biomarker” for cardiovascular disease), pre-screening can be used to suggest that the user should see a doctor.

Clinical-Grade Accuracy
Data used in a healthcare scenario obviously needs to be accurate enough for the healthcare provider to have confidence in the quality of the data to safely make a healthcare decision. And, of course, the devices used to collect the data also need to be clinically validated for the medical claim they are making. In fact, PGHD data integrity and validation is one of the top five concerns identified by healthcare providers.

High-performance wearable biometric sensor systems are now demonstrating performance levels suitable for medical use cases. One of the breakthrough technologies that has enabled highly accurate biometrics to be measured in wearables is something called active signal characterization, a process that proactively identifies the biological, motion, and environmental signals as they come in from the photodetector and categorizes the data sets in the context of physiological models. This enables optical heart rate monitors using active signal characterization to provide only the relevant data to the signal processing methods that calculate the biometric measurements such as heart rate and respiration rate.

The active characterization of the signal data is important, because the noise removal is much more complex than “cancellation”. You need to know what noise to remove and how to remove it. It's important to characterize the type of noise being sensed so that right noise is removed the right way. Otherwise, you may end up “throwing out the baby with the bathwater”, accidentally attenuating important biometric signals that you may want to extract. Another important benefit of characterizing the noise is that the noise can provide critical context information that can be used to accurately deliver higher-level biometric assessments, such as cardiac efficiency, VO², R-R interval, and blood pressure.

Technology Integration
Something in human nature makes us very selective about what we put on our bodies, because the things we wear say something to the world about us as a person. This is highly relevant to wearables as well, because not everyone wants to wear something that looks like a fitness band. Indeed, the marketplace has been trending towards new form-factors such as biometric earbuds, jewelry, clothing, and many more things that people want to wear. Moreover there is now substantial diversity for wrist-worn biometric wearables, as traditional watch brands, such as Fossil and Tag Heuer, are now entering the wearables space.

However, making biometrics work in devices and form factors of all kinds is extremely difficult. There are at least 10 things you need to consider in building biometric wearables to improve your chances of success, including optomechanical expertise, advanced signal extraction methodologies, extensive testing protocols that match your use case, and manufacturing expertise. Don’t underestimate the complexity required to achieve the levels of validated accuracy required for biometric wearables, particularly in different form factors and for highly demanding use cases like healthcare.

Data Integration
Wearables are a means to an end. People don’t necessary want to wear a device, but they want the insights that wearable sensor data can provide. It reminds me of Theodor Levitt’s famous quote:

People don’t want to buy a quarter-inch drill.
They want a quarter-inch hole!
— Theodor Levitt

Therefore, if the biometric data doesn’t get to the right place at the right time in clinical workflows then the wearable is far less valuable to everyone involved. There’s still a significant amount of work to be done, because the right place and right time in clinical workflows is primarily dictated by electronic medical records systems and standard of care processes that don’t move as quickly as consumer technology trends and capabilities. This is a key point of friction that will have to be addressed at an industry level, as well as with individual technologies. Of further concern, data integration is not yet truly seamless with data collection hubs or cloud applications. There are many reasons for this, and the unfortunate end result is that cloud-based healthcare systems that analyze biometric data from wearables may not be compatible with a critical mass wearable products in the marketplace. Overall, this weakens the value proposition of both the healthcare systems and the wearables that feed them.

Summary
High-performance consumer biometric sensor technology has advanced to the point where it is accurate enough for some key medical purposes and new advancements are happening every day. In some ways “consumer” technology is outpacing “medical” technology in this area. One point of friction is merging the chaotic exploration and rapid innovation of consumer wearables with the methodical discipline of clinical validation. Companies that can address these requirements are well-positioned to continue the shift from consumer wearables to personal health devices.  

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.