With its ability to provide an increasingly deeper understanding of the complex interactions in the human body, the application of artificial intelligence (AI) medical science is revolutionizing healthcare on many levels. For example, AI is accelerating the discovery of novel drugs, enabling earlier disease detection, and facilitating vital breakthroughs in patient care. Rx for AI is a real opportunity for the healthcare industry.

Another important field being transformed by AI is remote patient monitoring. As the information generated by a growing variety of wearable devices continues to expand, AI has the potential to enable the analysis of biomarker data at a scale not previously possible. AI-enhanced insights could help healthcare professionals by improving the understanding of everyday biomarkers, revealing important clues about an individual’s health, which could enhance the ability to treat chronic illnesses.

This could be a game-changer for all of us — simply because we generally don’t do a good job of monitoring our health. When we go to the doctor and get lab tests, for instance, the results provide only a tiny snapshot of our body at one moment in time. A far better approach is to monitor our health continuously and then extract actionable information from that data.

Creating a more complete picture with Rx for AI

Periodic snapshots provide only limited information. If you are like most people, you probably go in for a checkup only once or maybe twice a year. As part of that checkup, you might discover that your vitamin D levels are low. They may have been low for many months, but there was no opportunity to know this until your checkup.

Measurements on such an intermittent schedule can’t keep pace with the fluctuations that your body may be experiencing. Moreover, even when you get tested, you’re being measured for only a very small set of biomarkers. As a result, the data captured is far too limited to provide a clear picture. It’s like viewing a still image from a digital camera with 3 broad pixels versus a video with 3 finely tuned megapixels.

By leveraging the power of AI, we can get a much better view into one’s health. We can continuously track health data and make connections between a patient’s health and their daily life — connections that could never be discovered when people are getting measured at a clinic annually. For example, we could correlate the impact of diet and exercise on an individual’s health by monitoring multiple biomarkers on a routine basis over an extended period.

Top AI ML Projects: Scientists Use Adaptive Algorithm to Fine-tune Acoustic Levitation Technology for Contactless Pick-up

Getting regular, everyday access to biomarker measurements could make all the difference. Many people have health patterns and conditions that emerge only when asleep, at work, or in stressful situations. These may not be caught during a visit to a clinic or doctor’s office. That’s why the explosion of new wearable devices brings so much promise. With more functional and more efficient wearables, we can capture those patterns and conditions routinely and non-invasively. AI then empowers us to interpret that data and act on it.

The macro and micro effects of wearable devices

To be effective, wearable health monitoring devices must be small and must provide non-invasive measurement capabilities. In other words, they must be woven into the fabric of a person’s life. There are already several smart watches and wristbands that provide basic biomarker tracking, and more advanced technologies are being developed that could significantly expand what these devices do. For example, one promising new technology is a wearable photonics-based sensor that could measure multiple biomarkers non-invasively. Such wrist-worn solutions could provide hassle-free, clinical-grade data about a variety of key indicators.

Collecting data from wearables is not new, of course.

The problem is that this data collection has until now been limited by the number of sensor modules that can non-invasively look beneath the skin to analyze various biomarkers. Existing sensors aren’t good enough. We desperately need more advanced sensors that do more than measure how many steps we take in a day or what our skin temperature is. Sensors need to monitor a vast range of biomarkers, such as core body temperature, blood pressure, body hydration, alcohol, lactate, glucose, and many more.

Once we expand the scope of wearable sensing solutions, we’ll be able to see deeper into a person’s health. We’ll be able to create a spectral fingerprint that is unique to every individual. That’s how we will start to measure the known unknowns and the unknown ones — and use that information to identify indicators of various conditions and diseases.

This data could potentially be used at both a macro and micro level. At the macro level, data captured from millions of people can be anonymized, and patterns extrapolated from it. At a micro level, data can provide deep insights into an individual person’s condition to show exactly what’s happening at any given time.

The combination of next-generation wearable sensing devices with AI makes it possible for healthcare professionals to extract dramatically powerful insights hidden within billions of data points. Theoretically, this would make it possible to detect the early onset of diseases like Alzheimer’s by tracking sets of relevant biomarkers. This type of early detection could provide enormous benefits to individuals, to clinicians, and to the healthcare system as a whole.

AIThority News and Analysis: Biomed Expert Backs AI’s Role in Improving Human Longevity

Better drugs through wearables

Another arena in which wearables and AI could be extremely useful is drug development. Today, the development of new pharmaceuticals is astronomically expensive. Bringing a new drug to market can easily cost over a billion dollars, taking into account the cost of failed candidates.

However, if pharmaceutical companies could get access to continuous, routine biomarker data drawn from wearables worn by subjects during clinical studies — data that could be obtained both at a clinic and outside the clinic as the subjects lead their daily lives — the companies might make important discoveries and reduce the time and expense of the studies.

A project that in the past might have cost a billion dollars and years of effort — and failed — could instead become a success because the combination of wearable data and AI might help the pharmaceutical companies understand why the drug works in some people but not in others, simply by analyzing biomarker information. The net result could help bring drugs to market more quickly and decrease the cost of drugs for patients.

In the not-too-distant future, it is easy to believe that medical-grade insights will be generated by consumer-grade hardware. The more we can miniaturize the hardware and put an entire medical clinic on a patient’s wrist, the better we can access valuable medical information. In other words, healthcare can get bigger by getting smaller. AI will be a critical part of that process, as it harnesses the power of large datasets to help people achieve their health goals.

[To share your insights with us, please write to sghosh@martechseries.com]