00:00We did a nice pilot study. It was controlled with an active stimulation of M1 alone and with sham. So we said, okay, now we saw this nice acute effect. Let's see if we can follow the paradigm that's been used for depression, give 10 sessions over two weeks, and then follow that by a maintenance period and see what happens. And I hypothesize that we'll see similar nice effects after 10 sessions and as well after maintenance and booster. And we were quite disappointed. Oh, no. That's how science works sometimes. Welcome to Stimulating Brains. 01:10Hello and welcome to Stimulating Brains. My name is Nathan Morelli, and I'm your guest host for this episode. Like many of you listening, I currently work on the invasive side of brain stimulation as a global DBS study manager for a medical device company. However, while I was in my academic appointment, I ran a cognitive motor neuroimaging lab where I investigated neural correlates to clinical gait and cognitive dysfunctions as well as the effect of noninvasive brain stimulation. Stimulation specifically trains cranial direct current stimulation on clinical outcomes. Our guest today, Professor Jesri Hostorf, is one of the leaders in the field of gait balance, cognition, and brain function in age and neurodegenerative disease. Professor Hostorf is the director of the Laboratory of Gait Analysis 02:02and Neurodynamics at Tel Aviv Swarovski Medical Center, professor at the Sackler Faculty of Medicine, and director of research. at National Parkinson's Foundation Center of Excellence, Tel Aviv Swarovski Medical Center. Professor Hostorf has a rich research agenda that has led the way in her fundamental understanding of clinical gait and cognitive dysfunctions, their interactions, and how to treat them in aging and Parkinson's disease. Moreover, his recent work has attempted to elucidate potential brain mechanisms that contribute to these deficits using novel technologies, including our topic of focus for today, transcranial direct current stimulation, or TDCS as you'll see it in short. On a personal level, Professor Hostorf and his collaborators' work had a profound influence on my own research agenda during my PhD and while I was starting my own lab. 03:02If you have a background in physiotherapy or rehabilitation medicine as I do, you will likely be aware of a line of research he has had in collaboration with Brad Minoy, who is a professor at Harvard, where they investigate the intersection between cognition, gait, and balance deficits to reduce falls and improve mobility in older adults and patients with Parkinson's disease. I implore you to read his recent and previous works on the matter as I have found them fascinating and innovative. I had an absolutely wonderful time speaking with Professor Hostorf and I hope you enjoy our conversation as much as I did. Okay, that's enough from me for now. Let's kick it to the interview. Enjoy. Welcome, everybody. Today we have a special guest, Professor Jeff Hostorf, Tel Aviv University. 04:00We're very excited to have him as he's had a lot of experience using non-embracing brain stimulation in some of his research works. And I thought he would be a unique, unique guest. He's a unique addition to this podcast because he can comment on patient populations that routinely have DBS as a therapy, but this is a nice alternative therapy as well. And he's going to give us some background regarding TDCS, its applications, mechanisms of action, and how he's currently using it in his labs and works. So, Professor Hostorf, welcome. Thanks. Great to be with you. Awesome. So, I guess, you know, I think we can start really with the mechanisms of action with TDCS and some of the more technical aspects of it. So, if you could just comment on some of them. What are the initial proposed mechanisms of action of this intervention that's kind of led to its imagination into clinical research? 05:00Sure. Actually, TDCS has a really long history. It's back in ancient Greeks to use TDCS with a fish electric to induce an electric current. And the idea is basically that a weak current is inserted into the brain at different locations. It's electrodes placed typically at different locations on the brain where an anode receives the positive electrode and moves through the brain and goes through the cathode. And it's a relatively weak current, only a few milliampers, one or two to maximum, maybe four. It doesn't, in contrast to other modalities like transcranial magnetic stimulation, where it actually excites the neurons. TDCS, because the current is so weak, it just changes the excitability. So, basically, it changes the background. It changes what's going on. And the idea is that by changing the excitability, 06:00it makes the neurons more flexible. And when called into play under certain circumstances, then they're able to, especially when they're in a very stable state, respond more appropriately. That's kind of a general overview. There clearly are many different mechanisms that are going at the same time in terms of release of neurotransmitters, cerebral perfusion, low-term potentiation, short-term potentiation. But the general idea is that it really a change in cortical excitability so that the neurons that are stimulated can respond better, either more strongly, more quickly, more appropriately, when they're, when they're in a stable state. When they are stimulated. Now, is there any evidence that EDCS has a differential effect on certain neurotransmitters that might then have downstream effects on their ability to induce plasticity, pain, short-term or long-term potentiation? And how does that then relate to different medication states 07:01for potential patient populations that it might be applied to in that interaction? Yeah, so that's a good question. There's actually a pretty recent review looking at the influence of different medications on response to TDCS. And clearly, because TDCS is changing excitability, it does have an influence basically on, I would say, on all or many of the neurotransmitters that are key to brain function. A lot of our focus is on Parkinson's disease, and TDCS also has an impact on dopamine release, neurotransmitters. So clearly, the background state, for example, if someone, you know, has a high level of TDCS, they're likely to have a higher chance of getting TDCS. So, you know, if you're looking at the influence of different medications, you know, if someone is with a patient with Parkinson's and is treated with dopamine, if they're off medication state, that can at least theoretically change the response of TDCS. So there's a lot going on in the background. You know, it's not just we take the electrodes, place them on the scalp, and then see the same response today 08:02and tomorrow and the next day. But there's a lot of variation related to the state of the brain, related to the state of neurotransmitters, and in part to medication state. But also, you know, there are even studies that try and make sure that food intake is constant, coffee, which can have an impact on neurotransmitters, for example, could be a co-confounded. So there's a lot going on. We know the brain is not static at all, and definitely medication state can play an important role. And most of the studies in Parkinson's disease have actually given the TDCS, the TDCS, the dopamine release, the dopamine release, the dopamine release, the TDCS. So when patients are on the on medication state when patients are in their, you know, have taken their dopaminergic medications relatively recently, a half hour, a half hour, an hour before the TDCS. And this is kind of consistent with many intervention studies in Parkinson's disease where the idea is to look at behavioral outcomes. 09:01Although things get a little complicated because we know at the same time that there is some suggestion with some outcomes that the TDCS can really impact on the brain. So we're really looking at areas like heart disease. But there could be actually too much dopamine can interfere with certain aspects. For example, cognitive function, there's an idea that cognitive function might actually be better in the off-medication state in Parkinson's disease to some degree. And it's overstimulated due to too much dopamine in the on-medication state. So that kind of gets complicated then when we're doing TDCS in the on-medication state where there's more DOPA available. And then we're looking at effects that are both mediated by both motor and cognitive function. So it's a little tricky to interpret what goes on there. But clearly, more generally speaking, medication definitely has an effect on multiple neurotransmitters. Now, do you believe that that interaction stems from the concentration perhaps? 10:02With dopamine receptors in the areas underneath that cathode? For instance, prefrontal cortex has a higher expression of dopamine receptors and perhaps M1. Do you think that might be cause for those differences you see in cognitive function benefiting actually from the off state as opposed to on? Yeah, definitely. I think, I mean, that's sure. One of the many complicating factors to TDCS is that it's not, again, the neurotransmitters. The neurotransmitters that are affected are differentially concentrated in different brain areas and different brain networks and therefore respond differently to TDCS. So, and clearly, I think that's maybe why, in part, why stimulation of the dorsal lateral prefrontal cortex has been relatively successful, both in Parkinson's disease and in older adults in general. It's been a relatively widely used target to aiming for the TDCS. 11:02Aiming for cognitive function. We know that the dorsal lateral prefrontal cortex is responsible for many functions, but in particular, it plays an important role in attention and executive function. And that has many implications for behavioral functions like the ability to walk and talk, which is actually how we got into the area of TDCS. So, back to answer your question. Yes, definitely. The concentration, the local concentrations probably play a very important role. It's an interesting dilemma to face. I know being an interconnected network, modulating one node might have causal influence over wide arrays of networks that underlie different processes. And we'll get to some of your single, versatile task findings here in a bit. But I was hoping you could describe some of the setup parameters and what it might look like if somebody is doing a TDCS intervention, if they want to do it for a research application. 12:00Or, you know, I know there's a. A number of now commercial devices that are out there. You gave some of the currents. You talk about current density. What does it actually look like and then how long are a lot of these interventions? Are there certain time requirements and then what's routinely seen in some of the literature? Sure, that's a good question. So there are a number of factors that. Influence the delivery and the efficacy of TDCS. One is the location of the network. And then the location of the network. And then the location of the network. And then the location of the network. And then the location of the network. And then the location of the network. And then the location of the electrodes. So typically in the past, two large electrodes, like 35 centimeters in area, were placed in two sides of the brain. And they were used to stimulate actually relatively large areas of the brain in a quite diffuse way. So one parameter is the electrode size. And today, in general, there's movement towards using much smaller electrodes instead of 35 centimeters. 13:01And today, in general, there's movement towards using much smaller electrodes instead of 35 centimeters. And today, in general, there's movement towards using much smaller electrodes instead of 35 centimeters. Now more like three square centimeters, which allows for a much more focal stimulation. So size and placement, of course, is super important. The idea to use TDCS is to target specific brain regions or specific networks rather than stimulate the whole brain. And so placement is super important. And again, typically, for example, we often target the dorsolateral prefrontal cortex, but one of the electrodes. the positive electrode near the forefront of the brain. So the second parameter is location. And there also can be combinations of electrodes. So in the past, again, typically there was two electrodes, input and output, and the currents streamed from the input to the output, and that was it. But today things are getting a little bit more sophisticated, and there's what we call multifocal or multisite stimulation, where there's an array of electrodes, six or eight 14:04or more electrodes, and the electrical flow can be quite complex. One of the challenges with TDCS is that not only do we have to care for and make sure that it targets the areas we want to improve with excitation, for example, the dorsal lateral prefrontal cortex, but also the current goes in and also has to come out somewhere. We want to make sure that it comes out in the right area. And does it? It does. And it also inhibits brain networks and brain regions that are important to the function that we're trying to target. So a real trick is to not only just excite the brain areas that are important, but also to make sure that we don't ad hoc inhibit those other brain networks and regions. So again, electrode size and placement is critical, the combination of electrodes. And then there are many parameters in terms of how the electrical current is delivered. Whether it's the magnitude of the current, whether it's... 15:03And typically for TDCS, it's transcranial direct current stimulation. So the current is turned on. It ramps up slowly so as to not... So the subject gets used to it relatively quickly, then it remains in a typical session for 20 minutes. That's kind of a non-standard standard. And then it ramps down quickly. And then it's a non-standard standard. And then it goes back to the offsetting. That's kind of the... And again, you can modulate the amplitude of the current. And also with more advanced modeling techniques, we can try to... For example, we can look at the current concentration in different areas and how it's kind of split among different regions, different targets. So for example, if we want to limit total current flow to four milliamps, we might use some of that to target one brain region or network and another portion of that to target 16:04a different brain region or network. And safety is obviously critical. But one thing that is really, I think, becoming clear with TDCS, at least in the research setting, is that it's quite safe. The side effects or the adverse events are quite minimal. Usually the transients may be a tingling or a little bit of itch, but it's really nothing major. So one emerging lesson really is that if there are necessary precautions, we don't... We typically don't give TDCS to someone with a background of epilepsy, but that's really as a precaution. And there's the side effects in our hands and in others, it's really quite minimal. So that's really something to put minds at ease. So there's a lot of room for playing with different parameters. And we'll call that... DR. DR. DR. DR. DR. DR. DR. DR. DR. DR. DR. DR. DR. DR. DR. DR. DR. DR. DR. DR. DR. DR. DR. DR. DR. DR. DR. DR. 17:00DR. DR. DR. DR. DR. DR. DR. DR. DR. DR. DR. DR. DR. DR. DR. DR. Part of that is the montage, so there's lots of, again, current modeling, where it goes, what it targets, how deep it targets, which brain region and brain networks, and there's a lot of effort to improve that because we know the brain is not just simple isotropic solution, but rather there's different areas of different thickness, different cortical thickness, different concentrations of neurons, et cetera. So the current flow modeling is quite challenging, and I think that's also one of the challenges and one of the reasons why we may or we often see differences across different subjects and different study participants because we typically model it on a general brain, which that's the easiest to do the modeling on, but obviously there are major changes from subject to subject. That's a great point, and I've seen some of the data that shows. Stimulation modeling across a number of participants with the same parameters and the tissue activation 18:04is substantially different, which is what I would surmise has led you to use those more complex arrays in order to try to shape and level and focus the intervention to a specific area. So because of some of these, maybe you could call them drawbacks with TDCS, what can we then infer about overall cortical function? Yeah. Yeah. Yeah. So can you really infer about overall cortical function if you could even infer maybe subcortical function from some of these findings from TDCS? That's a good question. So I think there's a lot of a series of studies by our group and others that, I mean, I would say just maybe to back up for a second. So in our hands and in the hands of others, we use TDCS, I guess, for two general purposes. One is really to gain insight into the underlying mechanisms. In our case, it's looking into the mechanisms of the control of gait and the relationship of cognitive function, specific types of cognitive function. 19:07So one area is really using TDCS to target specific brain region networks and see the behavioral response. The second use of TDCS is really to say, okay, maybe we can use it clinically. Maybe we have a patient with Parkinson's disease, have deficits in their walking pattern. Maybe they have freezing of gait. Maybe they have tremor. Can we use TDCS as a clinical tool? So I think in the first case, there's been lots of successful studies giving us insight into some mechanisms, into the role of the dorsolateral prefrontal cortex, for example. I think, for example, we've been very interested for a number of years in looking at the phenomenon of freezing of gait, which takes place in about 60 to 80% of patients with advanced Parkinson's disease. In that case, they report that their feet are glistening. They're not looking at the ground. They don't really understand why, but it happens relatively suddenly. 20:01There's no advanced warning. It's not like with an epileptic seizure that the patient knows about to happen. And inexplicably, they're not really able to continue. This is a terrible symptom. It causes a lot of falls, loss of independence. And we and many others have been trying to untangle this mystery of freezing of gait for many years. And the good thing is there are lots of theories about what causes freezing of gait. multiple competing theories and some suggest that maybe all of them are correct and some at different times or maybe they're different subtypes of freezing or different people with different types of freezing but the beauty of using tdcs is it allows us to probe some of these mechanisms and kind of state specific hypothesis about the role again for example of the dorsal prefrontal cortex stimulate it and see what happens we can you know do that acutely and get a response on a group level relatively quickly and see test specific hypotheses and see whether or not the 21:03data supports them so that was i think one thing that we've been able to get out we and others about from tdcs really for example in freezing of gates i think there's some nice cause and effect evidence now of the role of this these uh attention networks executive function networks which relate to the uh the role of the dorsal prefrontal cortex which there were certainly some i would say anecdotal anecdotal and behavioral evidence to support that but i think the tdcs evidence really helped to nail that down to a large degree so an exciting role of tdcs you know is really to be able to probe different mechanisms and understand what's going on there in terms of clinical outcomes as far as i understand there's no as yet fda approved outcomes for tcs we know i know there are a number of medical centers in the us and elsewhere 22:02that do use tdcs on a clinical basis for a variety of outcomes whether movement disorders like parkinson's disease cognitive function depression changes but again um i'm not sure exactly why i can't but i think we're still lacking large-scale randomized controlled trials that you know with the evidence to support the clinical utility you know maybe some of the trick or the challenge is really some of those person-to-person changes that when we you know take a large cohort um some of those person-to-person changes cause uh things to work out a little bit more complex way than we hypothesized um so again there's i think there's a lot of potential for the clinical targeting as well but uh i don't think we're quite there yet you know it's a fantastic point about freezing of gait and cognitive uh performance being enhanced 23:02specifically with tdcs and parkinson's diseases those are two symptomatologies that are not as efficacious with dbs particularly as you on in the disease progress where tdcs could show some potential to be of therapeutic benefit for these patients now i know you've done some research looking at dosing whether it's a single dose of tdcs seeing if there's benefit there versus more chronic application with perhaps some boosters so if you don't mind just commenting on the findings between an acute singular balance of tdcs versus the more chronic phases that might be a more realistic application to the clinical setting sure so early work in tdcs kind of asked the question okay if we give about 20 minutes of tdcs how long does that last and there's some nice early work from 2000 or so to showing that 24:02you know obviously it definitely depends on many many parameters again the electorate side sites the targeting new regions etc but in general i think it's fair to say that effects were retained up to two hours after a single 20-minute vow the question is how to translate it to something something that's meaningful clinically and meaningful in the day-to-day. So, again, for other outcomes, for mood, which is, I guess, has the mood and depression, which has the strongest background in TDCS, their standard is 20 minutes a day over 10 sessions, typically given over two weeks. And there is some evidence that that is retained. There's typically that's followed by maybe a maintenance phase and then a weaning phase. And there's suggestion that that really impacts mood and reduces depressive symptoms on a relatively repeatable basis. 25:04So kind of 10 sessions over two weeks, followed by some once a week and then weaning that down afterwards. And, again, there appears to be some long-term retention in that kind of paradigm. Okay. And I guess the idea is, you know, in some sense, you know, we're targeting the neurons in the brain and, you know, getting mechanisms for long-term retention and plasticity. But it's kind of like, you know, in a sense, like, you know, we know about for other interventions or, for example, if you want to improve muscle function, single-bound isn't going to do it. We have to do it over and over again. When you go back to, you know, Malcolm Caldwell's, you know, the 26:10! So we had a really nice pilot study led by a PhD student, Maria Dagan, along with some collaborators at Harvard, led by Brad Menor. And we found basically exactly what we would like to see. We stimulated the dorsal prefrontal cortex and the motor area, or M1. And we found that that combination improved the reduced freezing of gait in a group of patients with Parkinson's disease. We found that it improved cognitive function or specific aspects of cognitive function, suggesting that it targeted the areas that we're targeting, 27:01both the motor and the cognitive function. We found it improved testing mobility, suggesting, again, that the combination was getting where it was supposed to, it was working as we hypothesized. And we compared that to sham, which we saw basically no effects, and also just targeting of M1. And we found some clear advantages to the multifocal stimulation, which targeted both the cognitive areas and the motor areas. So we said, wow, great. This is, you know, we did a nice pilot study. It was controlled with an active stimulation of M1 alone and with sham. And people's guesses were this, they couldn't tell the difference between sham or M1. So the placebo effect was really pulled out quite well. So we said, okay, now we split this nice cute effect. Let's see if we can, you know, follow the paradigm that's been used for depression, 28:00give 10 sessions over two weeks and then follow that by a maintenance period and see what happens. We hypothesized that we'll see similar, you know, nice effects after 10 sessions and as well after maintenance and booster. And we were quite disappointed. Oh, no. That's a good point. That's science work sometimes. Hypothesis and it's well founded in pilot study. But, say, the randomized control trial, which we did with Brad and our colleagues in Boston, over 70 subjects with freezing, which is very challenging to collect them and to identify them, took quite a long time to conduct a study. And when we finally unblinded, we were, our mouths kind of dropped, not in despair, I would say, was the counter. So, yeah, in terms of impact of freezing, actually self-report improved to a degree. 29:01It was larger in the control group, in the intervention group than in the control group. But our objective measures of freezing, which did not respond quite as favorably. So it could be there was some signal there, but it was still, and again, maybe there was, could be a challenge in the control group. It could be a challenge in the measures that we used to assess freezing of gait. There's a large, that's also one of the problems with freezing of gait. We know that it's quite difficult to measure and it's not very consistent over time. So we could try to some degree to blame our tools, our evaluation tools. I don't think, I think that might be part of the story, but I think the more honest look at things is that the intervention didn't deliver. What we saw in the pilot study. And, you know, there are lots of reasons we've speculated about why that might be, but clearly, you know, it could be that we need to do a longer dosing. 30:00It could be that two weeks was not sufficient. You know, if you think about, you know, just doing an activity for 20 minutes a day, for example, if you're learning to play a new piano, if you're a relatively beginner at the piano, you try learning a challenging task. 20 minutes a day, it's really not going to make it. You're not going to be like Bach and Beethoven after 10 sessions. So it could be that the retention was not there because we didn't sufficiently stimulate it. It could be we needed to do it for a longer time period. It could be we need to change some of the parameter settings. I think, you know, again, specifically thinking about freezing of gait, I think that the idea is promising using TDCS, but I think, you know, clearly we're not there yet. Now for this study, did you look at TDCS during a therapy versus TDCS perhaps just in a non-activity dependent state, 31:00just perhaps sitting? I know that you've had some papers looking at that. Was that used for this study? Right. So this study, we kind of used a more classic paradigm where the subjects sat and just relaxed while they received the TDCS. We have done some work on this. There's a lot of interesting literature about combining TDCS with the active performance of the behavioral test of interest. And there's three options or actually four options. One is not to combine the two. The other is to deliver the TDCS while the task is being performed or deliver the TDCS before or after. So there's many combinations and there's actually, again, depending on the specific outcome, behavioral outcome, whether it's a cognitive task or a motor task. There's lots of evidence to support the different ideas. Some suggest that TDCS can be, should be delivered after the activity. 32:01The activity is, could give some sort of priming to prime those neurons and networks and then be more receptive to the TDCS. Others suggest just the opposite. We've done a little bit work actually delivering TDCS simultaneously. With a walking cognitive task. And there we saw that the combination did appear, at least in a pilot study, to do better than TDCS delivered while sitting. So that could be something to explore with freezing of gait. Maybe go back to the drawing board and try to do a pilot study where we combine walking or walking in a challenging, cognitively challenging environment with TDCS and see how that affects things. The idea there is that, again, it could be that we know that despite the idea that we use modeling of the current flow, we're using relatively small electrodes these days, it still could be, again, because of these person-to-person changes in tissue and brain structure, 33:04that the current flow might not get exactly where we want it, where we're targeting. So some have suggested that by the active performance of the behavioral task, which presumably is, you know, exciting those neurons that are involved in that behavioral task, I think we can all agree to that. By doing that, it allows the TDCS to be focused in those specific areas where it's needed. And that might lead to better efficacy. Makes a lot of sense. There's some evidence to support that idea. Some evidence that's not consistent with that idea. I think it really much, very much depends on the specific outcome and specific networks and brain regions that are involved. You know, the brain is not homogeneous. Different networks and brain regions function in different ways. So do you still believe that the problem of freezing of gait can be solved, or at least partially mitigated, with the use of TDCS? 34:01I know with DBS, they've looked at expanding past your basal ganglion nuclei, STN-GPI, and going into PPN, pedunculopontine nuclei, as a way of a multi-target array for DBS with specific targets towards the TDCS. So, for example, if you have a TDCS-based gait disability, do you still believe that that's out there for TDCS, or do you think that the ability to focus and find the current flow is just not there with the technology today? Or do you think that's outside the cortex? Yeah, I'm cautiously optimistic. I would say, I mean, again, I think the dorsal prefrontex does appear to be involved in freezing of gait. It does play an important role. The brain. One of the things that we've seen from TDCS is the brain, even among patients with Parkinson's disease, there's positive plasticity that can be evoked. And so, you know, in the past, we thought, okay, you know, a large percent of the neurons have died in Parkinson's disease, 35:00in basal ganglia, striatum, et cetera. And, you know, 68% have died when PD is typically diagnosed, and there's not much to do. But I think, you know, now we know that that's really quite an oversimplification, and there's a lot we can do to maybe not bring about new neurons, but certainly restore the, improve the efficiency and provide resilience of what's left, what's still there. So, and I think one of the lessons from TDCS is also that we and others have shown that there is some positive brain plasticity, even in Parkinson's disease. Whether TDCS of the dorsal epipharynx by itself is the answer, or maybe it needs to be combined with the DBS, that might be another way to go in the future. It'd be really interesting to see. You know, I think one of the lessons, I think, from TDCS and also from DBS is that we're really learning what we knew before, 36:04but the brain is really complex. There's a lot going on. There's lots of, you know, I can remember early models of, you know, the basal ganglia of Chagai Bergman, where, you know, showing some, you know, five or six blocks and the relationship. If now you look at those blocks from Chagai, they're, you know, infinitely more complex and more layers. So it could be that if we want to achieve some clinically relevant benefits for freezing of gait, that we need to think beyond just stimulating at one site or one region, and really target the different regions that are involved and do that in a smart and intelligent way. Maybe it needs to be, you know, in some closed loop response that it's only activated during specific times, specific parts of the gait cycle, 37:01or maybe only when there's a freezing of gait that's impending. There could be issues of habituation as well. So I think there's a lot of open questions. But I wouldn't say that we need to close the door on TDCS for freezing of gait. But I think we're certainly not there yet. Lots of room for improvement and lots of open questions. That's a valid point. Kind of getting back to some of your outcomes data and a potential measure that could shed some light on the dosage and chronicity of freezing of gait. And I guess I don't want this to be just a freezing of gait kind of question. This could be applied to a lot of different things. A lot of different patient populations. But what evidence out there exists describing the after effects of TDCS specifically on cortical structure and function from a neuroimaging stance? Whether that's a mobile form of neuroimaging such as mobile EEG or FNIRS or a traditional MRI, perhaps PET scan looking at either functional structural connectivity or a more task-based bold approach. 38:07Yeah. I think there's a growing literature that has supported, I guess, the TDCS understanding and also to show that it's really been impactful on brain structure and function using all of those different modalities that you've suggested. I think there's some move, maybe not for the outcomes that we're focusing on, but there is some move to use PET to really look at the different neurotransmitters and how they respond to TDCS in different forms. And also, you know, I think there's a lot of evidence out there. But also one advantage of PET is really to get at those subject-to-subject changes. So we have some pilot evidence, again, with our colleagues in Boston. It's showing that with FNIRS, with functional neuroinfrared spectroscopy, that actually the dorsal-lateral prefrontal cortex activation levels are reduced in response to freezing of TDCS. 39:05Again, supporting the idea that those brain regions have become more efficient and learned to work better during certain tasks. So the load on those areas are reduced. And we see that. And that can explain, in part, the behavioral improvements that we've seen in response to TDCS. In particular, coming back to the dual task walking, we've seen in a number of studies and a number of cohorts that, again, the brain is more efficient. And that's a good thing. Yeah. Yeah. And it seems that dual task walking seems to improve. And it appears that it's specifically the dual task walking as opposed to single task walking. So it looks like we're really brain areas related to the dorsal prefrontal cortex, again, are responsible for that splitting of attention and prioritizing an executive function, which is critical to dual tasking. 40:01Those are specifically targeted and improved. Again, there's some evidence from imaging from using FNIRs that those cognitive loads are reduced during dual tasking. And we don't see those in some of our studies. We don't see improvements in the single task, again, which is suggesting that really only when cognitive function is challenged that we see those benefits of the TDCS. And that's, again, supported in part by some of those FNIR studies. EG has also been used to kind of help. And we've been using that to kind of back up and better understand what's going on with non-invasive brain stimulation. I think more of the studies are focused on, say, another, maybe a cousin of TDCS on transcranial alternating current stimulation or TACS. There, the idea is to try and not just use direct current to change excitability. But there, the idea is to try and train and synchronize the brain. 41:02And synchronize different networks or different neurons to different frequencies. We've done a study looking at, for example, theta stimulation. And there's a series of studies using EG both to set the parameters, to provide feedback, and to see the changes during TACS as well as with TDCS. Interesting. So would that be more of an application to modulate neural oscillation patterns? Exactly. Yeah. Interesting. Now, I'd like to touch on some of the findings that you were talking about with dual TACS as opposed to single TACS and only seeing differences in that dual TACS function. And what that can tell us about neurodegenerative disease and the usability of this technology moving forward as that patient population will likely continue to expand. And the implications that's there. 42:01What are some of the things that this therapy could add? Sure. So we've looked at a number of different patient populations. Older adults. Patients with some cognitive deficits. A patient with Parkinson's disease. And I think the most striking and consistent finding has been that the dual TACS walking ability has improved. In response to targeting of the dorsal lateral prefrontal cortex. We've played with the combination simulation of motor targeting M1 in the dorsal lateral prefrontal cortex. The idea behind M1 is, you know, okay, well, gain is related to motor function. If we target M1, then that improves its excitability and responsiveness. And we might expect to see improvements there as well. And there are some studies. And again, coming back to the TACS. Again, coming back to Parkinson's disease. 43:01There are probably more studies using targeting M1 in Parkinson's disease during, than targeting dorsal lateral prefrontal cortex. Again, because Parkinson's disease for so many years was perceived primarily as a motor disease. And clearly it is a motor disease. But so that's probably part of the history behind targeting M1. And there's some really positive results even on some small RCTs. Also looking at freezing of gait. But the other option, again, as we've discussed, is to target the dorsal lateral prefrontal cortex for the combination. And in our hands, we actually have found that there is no advantage of adding M1, at least in some of our studies. I guess we don't have total consistency, which is part of the research process. But in some of our studies, we find that there's no advantage of the M1. Okay. So there's no advantage of the M1 combination, of adding M1 to the dorsal prefrontal cortex. 44:01And really the key benefits of the TDCS were from the dorsal lateral prefrontal cortex stimulation. And the idea, again, there too is that it could be for some of the questions of interest. And maybe there's healing effects for the usual walking or single task walking without the cognitive load. It could be that the dose needed to improve the single task walking is different than that needed for dual task walking. It could be that different brain regions are involved. Clearly, other brain regions or other brain networks are involved in that. They're not being targeted sufficiently. But I think so that might explain the lack of benefit from M1 by itself. On the other hand, I think the positive benefits seen both in certain cognitive tests, improvements in response to TDCS targeting, the dorsal lateral prefrontal cortex, and relatively consistency across different cohorts and studies, 45:02we found that, again, the targeting the dorsal lateral prefrontal cortex was beneficial for the dual task walking, which, again, not surprisingly, implicates the dorsal lateral prefrontal cortex in dual tasking, but also speaks to the potential of improving resilience and improving plasticity in that brain. I'm curious if, as a role of cognitive reserve plays a causal factor in the response rate to TDCS. For instance, does somebody with a higher underlying cognitive reserve have a better response to TDCS than somebody with lower, or does the individual with a lower cognitive reserve perhaps have a larger ceiling to work with and therefore have a larger change in response? Is that something that your lab has considered? And if so, what do you see? Yeah, we've considered it. It's a great question. 46:00We haven't really teased it out. I think both sides of the argument could be supported theoretically, but we haven't had a chance to really tease it out. We know that cognitive reserve obviously plays an important role in cognitive function and with aging and resilience to mild cognitive impairment, Alzheimer's disease is also important in Parkinson's disease, where, again, those cognitive changes, those changes become quite critical with advanced disease. But its impact on the responsiveness of TDCS, as far as I know, hasn't really looked at yet. But I think that's a really interesting question. Like you said, it could be that sometimes there's ceiling effects and TDCS could only be used in that kind of a sweet spot where people have sufficient cognitive reserve to be able to respond. But if their cognitive function is too good, maybe they won't respond. So it could be that they're not able to get any of the benefits of TDCS. 47:00So I think it's a really interesting question about the mediating or modulating role of cognitive reserve. And it could also explain, you know, we talked about differences of brain structure and brain function that change from person to person. But clearly, cognitive reserve more generally can be one of the explanations behind those differences that we've seen. So I think that's a really interesting question. And I think that's a really interesting question. And to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to to half the brain and the other brain side was the negative electrode. 48:01Now we're using much more focal electrodes. But I think in the future, I suspect the electrodes will become smaller, will be able to do better modeling of the current flow and make it more personalized. There's some work, again, my colleague Brad Menor, he's now doing some work where they do MRIs for the GIFT-TDCS to really understand the detailed personal functioning and structure of the brain before they administer TDCS and combine it. So they're able to tailor the TDCS to really make it personalized. And I think that's great on a research setting. Obviously, that's not going to be accepted widely on a clinical basis, but I think it can give us insights into how TDCS can be personalized and how important those person-to-person changes are. So it's a great segue that you've done for me as I'd like to kind of finish on talking about the future application of TDCS and what that will look like. 49:00I will mention for the listeners, there's a great article entitled Beyond the Target Area, an Integrated View of TDCS-Induced Motor Cortex Modulation in Patients and Athletes that Professor Hostorf was a co-author on, and I would recommend to readers for a nice, succinct review. And you alluded to it with Brad Menor's research, and I was just attending Nan's conference last weekend, and one of the central themes was big data and AI and how that can revolutionize patient care. And I was interested to get your perspective of how more personalized approaches with larger data sets can help shape and focus and refine the application of TDCS in the future. And is that even possible? Well, you know, there are those who argue that AI is the solution for everything. There's some skeptics. But I think, you know, 50:00one of the challenges, one of the problems with TDCS, we don't have huge studies. You know, a large study is 100 subjects or 50 subjects. That's a relatively large study. So there can be lots of confounders and variations that we're not really taking into account yet. And if we were able to get to larger scale studies, and then, you know, use machine learning or AI to kind of tease out why and better understand why this cluster of subjects were responsive and this cluster were not, that might be one way to go. There's also work on using home-based TDCS. You know, you can go online and buy a TDCS kit. I wouldn't recommend that. But there are research kits that are being introduced to deliver TDCS in the home setting where, you know, you can't really come up with a TDCS kit. So you have to have a very, you know, highly controlled setting where the electrodes are placed properly and where the research assistant can actually see online whether the current is delivered appropriately. 51:00And it could be that one way to get to these large numbers and rather than requiring people to come in for 10 sessions, which can be kind of burdensome over two weeks. And then with maintenance and booster, one way to alleviate that burden and to increase the numbers of people who have access to TDCS clinically, you know, first research-wise, and then clinically is to use some of these home-based solutions. So, you know, again, we have some prior data which really argues to safety. There were no problems, even in older adults, even with a few patients with Parkinson's disease, they're able to use the system typically with the help of a spouse or a caregiver, but they're able to use it in the home setting. And maybe that's one way to get to these large numbers. And then with those large numbers, you know, we have to understand what's going on with using machine learning and AI. So I don't think we're there tomorrow, but it could be in five, 10 years from now, 52:00we'll have sufficient numbers and a much better understanding of the way the brain works in response to TDCS to get us to a point where it can be delivered and used on a clinical basis. Maybe while you're exercising at home and you're on your treadmill or on your bicycle, you can also receive a little bit of steam, a little bit of TDCS to improve your cognitive functioning into target specific brain areas. And then for an older adult, any little added value can maybe take them from a place where they're functioning independently to a place where they're not able to do that. So if we can improve that by combining TDCS with other conventional modalities like exercise, maybe that will help find its place down the road. That's about it. And coming from a world of physical therapy in my background, you know, we have tens units in every single outpatient clinic and particularly even in neuro rehab clinics, 53:02you could potentially see them having TDCS at their availability to use clinically at some day. We're a long way from that, but if you have your future goggles on, it might look something like that. And the last point I'd like to get your view on is right now, one of the main progresses in TDCS is the ability to form closed loop or adaptive stimulation patterns with some sort of sensing technique. And then the stimulus is then modulated accordingly. Is there any current work in the similar guise for TDCS? And if so, you know, what signals could potentially be used in order to create an adaptive or closed loop TDCS? Yeah, that's a great question. There is, there's some recent work that actually has begun to look at that to use to add closed loop to TDCS. I think that's going to be a great way to address some of the challenges that we've discussed in terms of differences across people, 54:05to change the modulation, change the timing, change the targeting. You know, one of the, as you know so well, that location, location, location is critical for DBS, you know, how where the electrodes are. And it could be also with TDCS, maybe you, you might consider in the future kind of an array of a small array of electrodes over a small space. And then kind of through closed loop feedback, playing with the, to adjust which specific electrodes are given when, what exactly is the feedback? That's a great question, whether it's some EG measure looking at synchronization or activation of sets of neurons, or whether it's a measure from ethneers, looking again at, at, at, you know, dynamic response, or maybe it's a behavioral response. I think there is an open question, but I think there is lots of potential for, for closed loop and then really an exciting way to think about it in the future. 55:02Yeah, it'll be interesting, particularly with Parkinson's given the variability and clinical presentation, if you're using perhaps IMUs, inertial measurement units as your sensing components, you know, do you have them where six different IMUs, do you have them where six different IMUs, for monitoring balance and tremor and postural tremor and using all these different signals? And, you know, how do you limit that input down in order to create a, a feasible application? It's quite, sounds like a daunting task, but I'm glad. Actually, maybe machine learning or AI can help be helpful as well. There you go. Yeah. Well, Professor Hossdorf, I want to thank you so much for your time. It's been an incredible opportunity. I'm so happy to have this opportunity to get to speak with you today. And thank you for staying up late to speak with me. I appreciate it. Thank you. Pleasure. 56:19!