Sean Mackey, M.D., Ph.D.

I had some young grad students and others who said that they thought they could do this. I told them how you would do it. And I said, it won't work. And I'm going to pay you. I'm going to give you money to go scan people. And you're going to learn how it doesn't work because failure is a great lesson in life. And they came back and they showed... they could do it.

I had some young grad students and others who said that they thought they could do this. I told them how you would do it. And I said, it won't work. And I'm going to pay you. I'm going to give you money to go scan people. And you're going to learn how it doesn't work because failure is a great lesson in life. And they came back and they showed... they could do it.

And it was all through developing patterns in the brain and using machine learning models to then take that pattern, that signature, and be able to predict in other people whether they were experiencing pain. I didn't think we could do that because of the hugely individual nature of pain. It's so different from person to person.

And it was all through developing patterns in the brain and using machine learning models to then take that pattern, that signature, and be able to predict in other people whether they were experiencing pain. I didn't think we could do that because of the hugely individual nature of pain. It's so different from person to person.

But it turns out that there are core patterns in the brain that represent that experience of pain.

But it turns out that there are core patterns in the brain that represent that experience of pain.

FMRI? Functional magnetic resonance imaging. It is a standard MRI that people go into, but we do some fancy pulse sequences. We play some physics tricks where we can see where nerves, the brain's being activated. And we've taken this, we and others have taken this from being able to determine whether somebody is in a state of pain to predicting their long-term trajectory.

FMRI? Functional magnetic resonance imaging. It is a standard MRI that people go into, but we do some fancy pulse sequences. We play some physics tricks where we can see where nerves, the brain's being activated. And we've taken this, we and others have taken this from being able to determine whether somebody is in a state of pain to predicting their long-term trajectory.

We're working right now, a big grant that I have is to actually create composite multimodal biomarkers to predict their future state. We're getting there.

We're working right now, a big grant that I have is to actually create composite multimodal biomarkers to predict their future state. We're getting there.

you'd see rather dramatic increases in activity in brain regions such as the thalamus, the posterior insular cortex, the anterior cingulate cortex, the dorsal anterior cingulate cortex, and a number of other areas.

you'd see rather dramatic increases in activity in brain regions such as the thalamus, the posterior insular cortex, the anterior cingulate cortex, the dorsal anterior cingulate cortex, and a number of other areas.

You would also see it in S1. You're right. And indeed, what we've learned through this, that there's no one single pain region in the brain. That's another mistake that was made along the way. We all thought we were going to find a brain region. Then we can knock it out, right? Just go cut it out. And it turns out that didn't work. It's not one brain region. that generates the experience of pain.

You would also see it in S1. You're right. And indeed, what we've learned through this, that there's no one single pain region in the brain. That's another mistake that was made along the way. We all thought we were going to find a brain region. Then we can knock it out, right? Just go cut it out. And it turns out that didn't work. It's not one brain region. that generates the experience of pain.

It is a distributed network. It's all of these regions coming together and working in harmony. Doing what? Generating the experience of pain and then generating typically a response to that. Let me be very clear because there was a lot of controversy when we and others initially published our papers. We are not trying to take away the autonomy of the patient and the self-report.

It is a distributed network. It's all of these regions coming together and working in harmony. Doing what? Generating the experience of pain and then generating typically a response to that. Let me be very clear because there was a lot of controversy when we and others initially published our papers. We are not trying to take away the autonomy of the patient and the self-report.

I don't need an fMRI to see a patient and know if they have pain. I can just ask them. I can use self-report measures to get it. That's another part of the research. Where we're working to build these objective markers, this objectifying pain, is not to see what they're in now, but can it give us useful information? to predict treatment to a particular therapy?

I don't need an fMRI to see a patient and know if they have pain. I can just ask them. I can use self-report measures to get it. That's another part of the research. Where we're working to build these objective markers, this objectifying pain, is not to see what they're in now, but can it give us useful information? to predict treatment to a particular therapy?

Can we use it to predict their future state? Can we use it to predict their vulnerability to an injury or surgery? Those are things that just asking a patient right now probably not going to get there. I don't know if you want to go back to the brain.

Can we use it to predict their future state? Can we use it to predict their vulnerability to an injury or surgery? Those are things that just asking a patient right now probably not going to get there. I don't know if you want to go back to the brain.