Doctors and researchers are equipped with objective tests to detect and measure many serious illnesses. But when it comes to mental illness, no such tests exist.

Nationally, some 20% of the population will experience a mental health disorder during their lifetime, and globally these disorders cost the economy $2.5 trillion every year.

Yet there are no objective tests in use that can diagnose these disorders, says Leanne Williams, a professor of psychiatry and behavioral science at Stanford. Instead, the “gold standard” for psychiatric diagnosis is a verbal interview, asking patients how they feel, etc.

“Imagine if you were diagnosing and treating diabetes without tests, without sensors. It’s really impossible to imagine, yet that’s what we’re doing for mental health right now,” says Williams, who spoke about the research at Stanford’s recent Reunion Homecoming Weekend festivities.

Williams and her colleagues are working on a project called Mentaid, which aims to understand mental health by finding measurable links between brain activity and the production of certain hormones. Ultimately, the researchers aim to develop wearable devices that will measure brain activity related to emotional distress or disorder.

The science underlying that goal is complex.

The researchers are aware of six different circuits that the brain engages in. These circuits control particular types of activity. When the activity concludes, the circuit should switch off. But stress and anxiety can disrupt that cycle and a circuit that should switch off stays on, resulting in states like hyper anxiety or an inability to focus.

Zhenan Bao, a professor of chemical engineering at Stanford who is working with Williams, says that the presence of cortisol, a type of hormone that humans routinely excrete in sweat, is an important indicator of stress. The researchers are examining cortisol’s relationship to factors such as heart rate and skin conductivity and the six brain states they have observed.

The group is developing an early prototype of a wearable that would collect information on those variables and give doctors and the wearer insight into their mental health.

Their work is funded by the Stanford Catalyst for Collaborative Solutions, an initiative launched in 2016 to inspire campus-wide collaborations to tackle some of the world’s most urgent challenges.

Transcript
Mark Horowitz: Our next talk is going to be, again, in wellness, but in a slightly different area and it has to do with mental health which is, as everybody knows, a fairly large problem in the country. This was a really interesting collaboration from people in psychiatry as well in engineering to look at what we could do to get better assessments in people who might be in, you know, in situations that may need help before all the symptoms and everything else come out. So it’s for an effective, scalable, and affordable strategies for mental health. And it’s a large team but we’re gonna have two presenters today. So Leanne Williams will start and then Zhenan Bao will finish the presentation.

Leanne Williams: We are here talking about mental health and how to find new solutions through this collaboration. The numbers are staggering and they’re worth restating. So for years, we’ve known that cancer was a silent killer but in our lifetime, through transformative research and interdisciplinary collaborations, we have a new way to understand cancer as one example in more precise treatments.

Sadly, clinical depression has now become the number one silent killer. And that’s especially true for young people. In the time that I’ll be talking to you along with Zhenan, there’ll be 15 people who take their own lives by suicide, one every 40 seconds. It’s the number one cause of death in our young people age 15 to 24, more than all other diseases combined. And the rate of suicide as you may be hearing in the media has doubled in the last four years. So we need new solutions and that’s what we’re here to achieve. And we need them urgently.

One in five of us in total will experience a mental health disorder in our lifetime. I’m sure this has touched you, your families, your young ones, loved ones, students, many of us. Around the world that means for depression anxiety, we’re talking 600 million people — double the population of the United States — who are experiencing this right now. Finding a solution is also an economic no-brainer. We’re talking 2.5 trillion dollars lost from the global economy because of these disorders. That’s due to lost work productivity as well as a health care spend. Our goal is to change this situation.

We already know that one dollar spent on mental health will bring a four dollar return. So in terms of the science it’s how do we accelerate and find the solutions to make that possible? It’s a privilege to share how we’re approaching that with you today and to invite you to partner with us on making this possible.

This is the current situation. One of the biggest barriers for psychiatry in the field that I work in, is that we don’t have objective tests. So the way that we assess mental health conditions and risks for them is by interview and asking you how do you feel. And that’s with questions like the ones here. Imagine if we did this for cancer. How do you feel? That’s how we’re gonna define how we’re gonna test you and diagnose you and treat you. Imagine if you were diagnosing and treating diabetes without tests, without sensors. It’s really impossible to imagine, yet that’s what we’re doing for mental health, right now.

In the project that we’re working on together called MENTAID, the goal is to change that and harness the insights we have. We have new ways to look at mental health through understanding the brain, and that means going to the source. And by doing that, we have an objective anchor for understanding what is going on. And we know that through our collaboration we can bring this information together, from brain to sensors.

Clinical depression happens when stress becomes chronic and uncontrollable, and it will cause short circuits, effectively, in the brain, that are hard to rectify. So we need to identify them and we need then distal sensors, like, measures all that stress that are objective-based in cortisol that can peacock when the risk is occurring.

In terms of the brain, we’re using one of our new ways. It’s the first way that we have developed in the center that I lead, Center for Precision Mental Health, to assess the brain using direct measures and generate types of depression in each person. These are examples. There’s six different ways that we can see circuits in the brain engaged and then become disrupted and form particular types of symptoms that you might experience. For example, rumination is one circuit in the brain that is essentially your brain in idle. You’ll see a set of brain regions cooperating together and usually they switch off when you’re engaged in a focused task. If you end up with some staying on because of continual stress, it means it’s almost impossible to engage focused attention and you’ll find that the loop of wiring and brooding and what we call rumination will start. So it’s one example.

You can think of something like, say, threat disregulation is when the brain stays in alarm mode after acute stress and you will feel heart racing, palpitations, sometimes panic attacks and that’s the brain not switching off from that mode. This is innovation within bringing into development of sensors. We do that by assaying the brain directory through functional MRI and we have advanced ways that we do that here at Stanford. And then imagine, which is where we are heading, that you could take this brain information and have a real-time, real world way to now understand that stress response.

On that note I will hand over to my wonderful collaborator and leader.

Zhenan Bao: Here, the aims of the project is to develop measurements that can be done routinely, can be done on a daily basis so that we can access parameters that are related to the brain type that Leanne and her colleagues are studying. We would like to understand what the correlations are between these measurable parameters and the different biotypes. We then want to build a wearable that can allow us to measure these parameters continuously. And finally, we would like to, within this project, validate the wearable that we build.

Here you see the group of participants, the faculty participants in this project ranging from engineering to psychiatry. Again, these are the brain types. Based on the neuroscience understanding and the extensive literature review that Leanne’s group has done, here is the hypothesis on what the correlation might be between heart rate, the skin conductance, and also very importantly a chemical, cortisol, that’s a type of hormone that we all generate everyday. That hormone is strongly related to stress level. The hypothesis shows that potentially there is different pattern between these parameters and the biotype.

We are very excited that just recently Leanne’s group is able to show for the first time that there is such correlation that they could observe experimentally by using the big machine you have seen earlier using stationary equipment that has to be located in the lab. Of course, we want to turn this into wearables. So one of the most challenging problems that we face on the engineering side is to try to measure this hormone that can be found in our sweat because it’s quite readily collected but the challenge is the amount of this hormone in sweat is a very, very low concentration. So this is nanomolar concentration and it’s equivalent to, take your fingers and pick up two small pinches of salt and put it in ten million gallons of water. That’s the concentration we’re trying to detect. It’s possible to do that in a well-controlled laboratory setting and that has been done, that’s how it’s done in a hospital but we need to do that in a simple wearable in a continuous fashion.

And on top of that, this molecule is very, very small and we need some kind of biological receptor that can recognize this molecule in order to detect it very specifically. So this molecule is about a thousand times smaller than the receptor that we need to use to detect it. So it’s equivalent to you standing in front of a hundred-story building. The binding event will cause very little change to the surrounding environment and making it very difficult to detect. We are developing methods to allow this to happen.

What we do is we attach the receptor onto the surface and then when there’s cortisol generated in the sweat and when it binds to the receptor, it will lift up this giant building, essentially, from the surface and then it will cause a very, very dramatic change so that we can detect it electrically. So this is the method we’re using and we’re able to show that this assay that we develop is now possible to detect the trend that we collected from patient sample from saliva and then from sweat and showed the same correlation as the correlation that we would measure using a laboratory equipment.

We are also building, starting to learn how to build this wearable. This is already the second generation we’re building. Each generation we’re incorporating and adding more functionalities into the wearable. And this also allows us to start to learn what the design will need to be in order for users to start using this kind of wearable. For example, the users would tell us that they all would like to have the intervention but they would also like to know that they are starting to develop high level of stress early on and we also find that for engineers, we would like to, of course, put the wearables in locations where it’s easy for us to detect but most of the users we spoke to, they wanted the sensors to be located in the green region so that the sensors are not readily visible.

In summary, now with the two years of the project, now we’re able to show for the first time that there is indeed a correlation between these measurable parameters and the biotype and we demonstrated the visibility of measuring the hormone cortisol from sweat that correlates to the measurement that’s obtained in the laboratory.

Going forward, with these two important pillars, we will be able to start doing some clinical validation and start the design and the user interfaces studies and even going further forward, we need to start thinking about engaging different stakeholders and thinking about the ethics-related issues.

Source: Stanford University