Empatica and the importance of wearable devices in getting the full patient profile

By Liza Laws

- Last updated on GMT

© Getty Images
© Getty Images

Related tags Wearables Remote patient monitoring Virtual clinical trials Patient centricity Decentralized trials patient engagement

Matteo Lai was part of a panel at DPharm in Boston this year, it was called working together to accelerate the adoption of digital biomarkers in clinical trials.

The company recently revealed the Empatica Health Monitoring Platform which offers enhanced compliance reporting inside its Care Portal, with both site and study-level views. The company says this is an invaluable feature for healthcare professionals and researchers conducting clinical trials or studies at scale.

OSP spoke to Matteo to find out more about his role and what drives his enthusiasm for monitoring and data, and how Empatica is getting it right.

OSP: Can you explain your role and how you got to where you are?

I have a background in engineering and in design, and I've always been fascinated by the idea of understanding people. I worked at MIT Biological Engineering and then started this company with another MIT professor.

In 2013, we worked on the idea of building algorithms to understand what's going on with people in their daily life outside of the hospital. That was the main idea.

I've always been fascinated with the idea of knowing what's going on because when we're suffering, we often talk about food being contributory and other causes, but people really suffer from a lot of diseases that are completely behavioural, and we prevent most of their effects with a better lifestyle.

It's very difficult for people to manage all the facts and data associated with their condition. When we started Empatica, it was myself, our Chief Technology Officer, Simone Tognetti, and our Chief Scientific Officer, Roz Picard, who still is a full-time professor at MIT.

She invented affective computing, which is the idea of building algorithms using machine learning to understand people’s behaviour and emotions in daily life. What I wanted to do was focus on medical conditions where we don't know much about the patients.

One example is epilepsy, where our first products were implemented/used. We actually don't know when a seizure will happen, and what happens to the patient beforehand. When prescribing and dosing a drug, doctors have to rely on the patient’s self-report of how many seizures they had. They literally have to count themselves, even if they're not conscious or there are no caregivers around, and then report them in a diary. They then must provide them to their neurologist, and the neurologist can then decide whether it is accurate and adjust the medication or prescribe a new one.

It's a very tough job to do precisely because even if the drugs are effective, you don't know what's happening to the patient and so you don’t know how accurate you are.

I've always been fascinated with the idea of having sensors, and we also build medical devices with one of our partners. Medical devices are able to collect this data accurately so you can then develop and validate algorithms as digital products. So, we have this concept of the algorithm, as a software, as a medical device that's used for diagnostic purposes.

OSP: What kind of what devices do you have at the moment and are there any that you are keen to get into?

We wanted the sensors that we develop to be used in wearable products. So we have a wristwatch, it has similar sensors to some that the Apple Watch and other commercial products use, but with the difference that devices like ours are medical devices.

They can be prescribed by a doctor, and they can be used in clinical trials, they can be used for real clinical research. We developed these devices to have accurate data of all of the raw data coming from the sensors and also to provide this data to the patient.

We collect pulse rate variability, respiration rate, oxygen saturation –  cardiac based and respiratory based measures. We have temperature, we have movements, or any type of activity-based measures, we’re using an accelerometer and gyroscope and, we also have electrodermal activity, which measures autonomic arousal, so the sympathetic response from the autonomic nervous system.

OSP: In the UK, everyone is going crazy for the glucose monitoring wearables, is this an area you’re interested in?

The work that we do, especially in our clinical trials is around neurology, so on the central nervous system. We also work in immunology, some oncology, and some cardiology as well. And we've been used in some diabetes studies. The reason for the benefit of using a smartwatch instead of the patches (in diabetes studies) is the reduced burden on the patient.

OSP: Can you give me some examples of what you have got from any trials and through your monitoring and data groups?

One example of an algorithm used in clinical development is with one of our clients that has a collection of studies about trying to understand pain. And so, for a lot of the treatments used in fields like epilepsy, you really don't know what's going on with the patient. You have the self-reports, but they're not super accurate. They're very subjective to their energy.

If you're a pharma company trying to decide what's working and what's not, sometimes it's difficult to do, especially on new programmes.

They developed some digital measures that have been associated with pain to have an internal pain index, where they can see within a collection of molecules what is working and what is not working. Not because they're going to go to the regulator for approval with that compound, but so they can move it from an early-stage development to a later stage of development. There is a bit more refinement in how they can choose what is in their pipeline. These are very expensive programmes, so you can decide in a more accurate way earlier on if something is worth an additional investment or not.

OSP: Measuring pain is so difficult, especially when you’re told it is what the patient says it is - how you do it?

An example, we worked on migraines using a similar approach to our work in epilepsy. Migraines affect 13% of the population overall, but 85% of those with migraines are women. It's a massive drag on productivity and lifestyle for people who suffer. Then you have varying degrees of intensity, you have people that have one a month, people that have one a week, and they have different triggers, like pressure, stress, poor sleep, lack of physical activity, but it's difficult to make that objective. So, when all these patients have to interact with the neurologist, they have a hard time understanding what works and what doesn’t, and that also affects the development of new drugs. So, having an objective way of understanding what’s going on with the patient can benefit both the development of better therapies, but also, the quality of life of patients themselves, which is the main reason behind all our work. 

OSP: So, what are you working on now?

Although it takes a while to develop these, we are actually one of the first companies to have a product like ours approved by the US Food and Drug Administration (FDA). In neurology, we have an algorithm running continuously for epilepsy, doing real time detection of seizures, and providing alerts at night in the case of epilepsy. If people are sleeping, they're measured temporarily, by pauses in the respiratory function in the patient. It's very valuable to provide these real time alerts.

OSP: It sounds like you’re doing very well but aside from everything that you've just said, what would you say is your unique selling point - why should people to come to you?

This is still a new field, there's a lot of talk about using AI (artificial intelligence) in healthcare. If you look at the products on the market, there's very few like ours, but what I do know is just because you have AI doesn't mean that the product works well, or that it will make other problems go away. There’s still a lot of work to do. We've been early with these products, so we've learned a lot of the lessons.

We have been doing this long enough to have a lot of knowledge about the needs of pharma companies. Sophisticated pharma companies today are investing in digital technologies. They think they want to have people that have experience, who've done it themselves. I guess we are among the people that are experienced in the field. We've been more widely used than other data technologies, so we have the expertise, because we've done the work ourselves in deploying these algorithms onto patients. We have patients who use our products every day. So, it's a better fit for making something that also works for clinical trials.

OSP: Have you anything on the horizon, what's coming up for you next?

There are some things we're very excited about in terms of new biomarkers, so new measures. The exciting ones are mostly confidential. I cannot disclose those, but I think there is a lot of opportunity in solving many important problems, like migraine, pain, fatigue, and others that affect a variety of therapeutic areas, not only in neurology, but also in oncology, and immunology as well.

The point here is that we cannot solve the challenges present in the area of developing and implementing digital biomarkers in decentralized trials alone. The collaboration of pharma and biotech companies, regulators, and technology companies is essential - different bodies need to come together to push these innovations forward. But it’s exciting to see that an area that has been a niche for quite some is now gaining more prominence and becoming more widely used. 

Related news

Show more

Related products

show more

Saama accelerates data review processes

Saama accelerates data review processes

Content provided by Saama | 25-Mar-2024 | Infographic

In this new infographic, learn how Saama accelerates data review processes. Only Saama has AI/ML models trained for life sciences on over 300 million data...

More Data, More Insights, More Progress

More Data, More Insights, More Progress

Content provided by Saama | 04-Mar-2024 | Case Study

The sponsor’s clinical development team needed a flexible solution to quickly visualize patient and site data in a single location

Related suppliers

Follow us


View more