we are live now and welcome to another episode of the Machine Medicine um Interview Series on neuromodulation and we're delighted this week to have Professor Leigh Charvet and Dr Giuseppina Pilloni from New York uh with us here today welcome both of you and and i wonder perhaps we could start with you Giuseppina could you um uh give us a brief um overview of how you got into this fascinating field and and uh yeah i'm delighted just have you here can you can you hear us Giuseppina hmm maybe not okay so perhaps we'll start with you Leigh if that's all right and Giuseppina will Giuseppina is sorting out her sound issues yeah would you mind giving us a little overview of how you got into this area Giuseppina are you on yeh I'm on so you can start with your background okay yeah okay so but i cannot hear the other people oh okay i can see you but like i can i cannot see the other people and i cannot hear the other people um are you in the same building yes would it be would it be all right if you went down the corridor and you both sit next to one another i think rather than finding a technical solution that might be yeah so yeah yeah maybe you would start Leigh i can start talking so i'm a neuropsychologist um clinician and i've worked in neurology um primarily people living with multiple sclerosis or MS um but you know really all different types of neurologic conditions with uh cognitive and psychiatric problems resulting from their conditions and so um you know and have done a lot of research around that and always with the research focus on what we can do tools to improve quality of life and reduce reduce the disease burden that they have um and so i was doing i did a large trial of cognitive remediation with people living with MS and that trial was telemedicine and that actually brought me into the world of tDCS because we were very interested to see if we could pair stimulation in the rehab setting to see if we could boost uh benefit of cognitive training to help get greater benefit and shorter amounts of time and so that was about 8-9 years ago that i started partnering with with uh biomedical engineers and and immediately because of my background in telerehabilitation to tDCS at home for clinical application and that's you know uh directed the journey that's been quite an intense journey um this part got you but so you really came to it from a from a very clinical kind of place with a significant sort of tech uh element yeah and Giuseppina be interesting to hear how how you got into this what your general background is and how you got into this area so i'm a biomedical engineer and i started working with the Charvet since 2018 i went here in New York the first time from Cagliari in Italy or as a as a visiting PhD student and then i joined the team last year and um so my main focus was like the PhD was like the quantitative assessment of gate and balance uh with inertial sensor and first platform and so i apply this technology to see if like some like rehabilitative intervention uh can um like improve gait and balance in neurological disorder and specifically in multiple sclerosis gotcha so so yes it's kind of nice um asymmetry so you came to this field from really a very different place i suspect several years ago you were studying systems of partial differential equations um and that kind of thing right and now and now you guys are working together so that's fantastic and and and that's without further ado it'd be great to get on to the the the meat of your of the your this recent publication on the uh on the use of transcranial electrical brain stimulation to manage the acute and the chronic uh Covid-19 symptoms so i think as i mentioned briefly Leigh when we were talking before i think it comes as a surprise to people to learn that neuromodulation and bioelectrical medicine more generally has a role in inflammatory disease but it really does and it could potentially be very impactful indeed right is this what's the best way to sort of think about it um i think absolutely and and i think it's also driven by the accessibility safety and tolerability of these tools so it's a non-drug intervention that patients can have immediately it can be applied anywhere and you know as we just mentioned our our focus is on on home use um and so it and if you look at uh the the the growing literature um there's there's a lot of potential direction that can go for evaluation and potential application of these stimulation therapies and in managing Covid throughout it's you know from acute through um you know persistent symptoms and recovery but if i can play devil's advocate for a moment i think you know certainly when i was studying neuroscience um um i i think the the kind or or even medical science more generally i think the kind of the way that we sort of conceived of the nervous system was as an information processing system and in that regard sort of certainly conceptually seemed to sort of consider the immune system and other sort of related system say the endocrine system as somewhat sort of in parallel but sort of separate and yet and yet there seems to be a premise underlying here that actually by intervening with you know electrical kind of um interventions we can actually modulate the immune system sort of i guess indirectly through the nervous system or so indirectly is key so we're not saying that this treats Covid um i think that it it it treats the effects of Covid on the system and manages you know the sequel of what Covid does to somebody and i think that what it does in the acute infection and the persisting symptoms is really unknown and still being characterized and so it's just catching up to seeing if we can apply some of what we know in other syndromes and other situations to be of help here and to evaluate um but again with we have to remind ourselves too that this is an accessible tolerable safe therapeutic approach and so really you know a call for research a call for evaluation um you know and that's really the point of our paper too is just for it to be considered um fitting in really in an area we don't have many other therapeutic options right now yeah it's a great point isn't it because you know drugs are our sort of go-to as clinicians and uh and i i do remember as a junior doctor myself the kind of sense of terror that it induced in me at just how common prescribing errors and and sort of you know maybe even ignorance of what the effects of drugs how much of that there was uh and and yeah they were our kind of main tool at least as as physicians for treating disease and yet it was anaphylaxis and all kinds of unpleasant side effects and so forth so they really are very dangerous things right drugs so so are there any deleterious effects i mean one would one would have thought that the power to do good would imply the power to do harm and yet it doesn't seem very easy to actually damage people with with these technologies what's what's the underlying intuition for that yeah so i i you know especially with uh transcranial electrical stimulation it comes from just a really strong record of safety and tolerability and no serious adverse events reported in clinical trials to date certainly in our high high volume of experience it's really well tolerated and so um you know you know everything has side effects as far as but they're they're usually you know at the site of the electrode you know discomfort that kind of thing that that's not something that um would prevent you from concern for for evaluating in a therapeutic context um and we've you know but uh we're learning more and more and um we definitely need more evidence for sustained continued use of these therapeutics you know a lot of the trials have been really limited to few applications and so if we look at we really believe that there's accumulative benefit so you need repeat ideally daily application for that benefit extended periods of time and and and we just don't have the the full body of evidence that we need to really guide dosing parameters and and and what we would expect for long-term use but with that there's nothing yet that would would make us think that or or or anticipate a worry about continued use so yeah yeah so we're cautiously moving forward but i do want to you know always note that there's caution and that more evidence is really needed especially for dosing guidance and parameters in this field yeah no absolutely i mean there's so while the empirical it seems to be quite a lot of encouraging empirical evidence that this stuff works and works quite well in contexts like Covid and and actually you know many different aspects of the disease right you discuss in your paper that the fact that it seems to be it can be used in the early stages when there's an acute inflammatory response it can also apparently help with uh supporting respiration through vagal nerve stimulation uh can also help with uh you know the neuralgia the the pains that are pain chronic pain syndromes that may be associated with sort of long Covid if indeed that exists um beyond a psychiatric phenomenon but um but so it's a very very powerful but um but so much ignorance too right like you said we don't really know how how this stuff is being affected i guess that's an interesting point for you Giuseppina if i may what are the sort of what are the sort of interesting problems that attract an engineer and uh someone with your your sort of skills and background to this area uh so there is a lot of technical thing in transcranial direct current stimulation and especially all the modeling of electric field so how the electric field propagates in the brain uh so and this means which is the best target area uh to reach the best um the optimal benefit from the treatment uh how how the brain activation change with current intensity so we should increase the current intensity or we should use the standard 2.1 milliamp for example how long the treatment has to be so 20 minutes or we we can increase so there is a lot of like unknown things that we are like actually investigating and also for example the real the real-time effect of tDCS on our brain with like tDCS and MRI uh protocol and so there is a lot of like interest yeah i mean i guess there's a lot of different sort of levels of analysis you can look at like you said you could you can model the the current the current and and and i guess that would be helpful for things like building systems that could uh predictively target a particular say deep nuclei in the brain or indeed a superficial um uh module of the brain but then i also guess there's the you could you can model the sort of interaction of population of neurons with that with with those currents from a sort of um electrophysiological perspective but then at a i guess at a deeper level there does seem to be this sort of question of a neural code right like somehow somehow we're re programming nervous systems which are and in turn reprogramming remodulating uh inflammatory systems or or even cardiovascular or respiratory systems um and and and i guess it comes back to um this sort of question that we had there seemed to be something of a paradox that when you sort of intervene although we have very little idea what we're actually saying in the neural code it almost always seems to be either i either have no effect or to be beneficial doesn't that strike you guys is curious it strikes me as very curious indeed um so i i i guess just I was saying is that you know i i was noting that we need dosing parameters from the clinical perspective but we you know as with many you know we're not unique in this field the mechanism remains largely unknown as well and so we really need the work to understand especially when we use it to target to match what we think is a model of an illness and how that's effective yeah then we need to understand mechanistically to drive how we would use these and so again right now where it it's it's theoretical mechanism you know or built on on use and symptomatic use in other trials or rehabilitative use in other trials and and bridging that to potential use in Covid but i think that with the work and exactly what you're saying as far as as you know modulating neural code modulate you know influencing the neuroimmune response those are are largely theoretical right now and we really need imaging models and markers you know Giuseppina has done a lot of we have a simultaneous tDCS MRI study going on you know modeling the effects of tDCS application in real-time and with the effects on the brain and then looking at the um the consecutively applied results to that so to see how it changes the brain um but that all of that is so preliminary and so so i think that that it you know i just want to point out that we're in the theoretical kind of matching of disease to disease i guess you know what do they say like in theory theory and practice should be the same but in practice they're not right yeah but um yeah that's that's right so i i i guess another interesting aspect is one of the other kind of compliments i mean you mentioned yourself Leigh that you actually had a bit of a background in telehealth as well even prior to getting into uh into neuromodulation and so um there's a very there's very sort of what do you think the most sort of complementary kind of subsets of technologies are i mean i think in in your paper you discuss the possibility of doing uh you know remote uh neuro rehabilitation which is augmented by you know transcranial electrical stimulation so a very kind of like cool um sort of nexus of technologies delivering a therapy remotely you know at scale right right and that's how we actually started with tDCS because of the handheld portability of the devices and we telemedicine and we know from almost every rehabilitative therapy you need consecutive near daily application for extended periods of time and so so tDCS can be fair or tES can be therapeutic on its own for certain uses but also you compare it um to boost the learning that occurs with with a simultaneous learning activity whether it's physical or cognitive rehabilitation for instance and so with that angle especially home-based therapy is really important um to provide the ongoing daily sessions we do it remotely supervised through telehealth so it's it's very structured very guided but the technology is sent to patients at home and they're guided through it in that way um and so we had actually developed a whole system for that um eight years of development to be able to reach patients in their homes to deliver tDCS for that purpose um and i we with Covid again every you know especially that uh we suddenly were moved to telemedicine across the board patients were isolated in need of treatment so it really lends itself well to that model to both reach patients in need and also to provide that ongoing daily type of therapeutic activity to boost recovery gotcha so that kind of yes is that is that how you decided to do this working in Covid just because well it's obviously very topical at the moment uh obviously a lot of people that have uh the condition and suffering from the security so it's a very you know major topic of just because everybody's trying to think of what can we do therapeutically to help all of the patients with Covid and particularly with you know we see a lot of people who have those persisting symptoms we work with patients with neurologic disorders who then get Covid so there's just a tremendous therapeutic need across the board and so you know it was positioned in that way but also we have a um through NYU Langone here we have an innovative care uh policy that allows us so it's not approved here in the United States um but but it's considered innovative care so we have a clinical service where we provide uh tele-tDCS um to patients in their homes therapeutically or clinically and so we began to get people uh recovering from Covid and that that service as well and so that's that's kind of informed a lot of our thought trying to optimize their rehabilitation their recovery and to help them in that way and so when you're doing a sort of tele um uh session the are you why is it why do you have to why can't it be done asynchronously do you have to guide the positioning of the device that you have to guide why do you have to be yes yes you don't um you know we come from you know our primary um activity here is research and so with all the rigor of research you we you know and also from the abundance of caution again as you noted earlier you know you know so so these are patients often with with a high level of neurologic impairment who are self administering tDCS in their homes and it can be complicated there's technology involved and so we we built a system around live video connection at every point and um especially important if we're pairing it with an ongoing therapeutic activity we can just really enable that and um so it's not necessary i wouldn't say and i think there is definitely technology and models where people can self-administer i do think though that especially in the beginning it's very it lends itself really well to a daily contact therapeutic model and then both for safety and clearance checks but also just to checking in reaching a patient at home every day is very powerful and can also ensure adherence to the therapy and make sure everything's going okay so there's a lot of kind of soft therapeutic benefits around that daily connection model too um gotcha yeah okay but if you're thinking about making it sort of truly very scalable right absolutely and if the technology were sort of suitably productized i guess and made for easier the technology is for self administration or for um asynchronous administration as you said which can still be clinician guided at a lower level than live video connection um and i definitely think that that's where we're going but again we need that guided use and um um you know and always want to come from an abundance of safety precaution that you know especially where in the home in somebody's home for their use so so another interesting aspect that we often think about and i'd be interested to hear your thoughts on it because obviously one of the one of the advantages of this is it's non-invasive um but then one of the one of the interesting things that happen has happened in invasive uh neuromodulation for example in deep brain stimulation is that you can now record at the same time stimulating and there's a lot of excitement although not a lot of proof i have to say yet but one hopes a lot of excitement about being able to record simultaneously and use that signal uh as a feedback signal in order to optimize the as it were the forward stimulation if you see what i mean and so um and so what what kind of paradigms do you have in non-invasive uh this non-invasive context for getting a feedback signal to tell you maybe you know as a teaching signal is it worth to update your parameters ultimately or even tell you if you know you're having any effect i can tell you that's a very topical question that's a great question we think about that a lot do you want to you know so we won't be able to use uh uh to monitor or to monitor real-time related variability because can be used it has a bio feedback uh for like uh to see if really there is like a change and um like an answer from your body to this kind of like so that was what was the metric you were measuring the heart rate variability or right the heart rate variability right so the insert the the inter beat variability so the entropy of this sort of uh this interval is that right that's interestingly sort of the the entropy is uh is sort of correlated with health right right exactly in a sort of broad kind of way that you know and then as people get sort of more as they get closer to right death i suppose um yeah so biomarker feedback is just is really important um so far EEG has not been helpful for us to it really to inform it so that was probably the first go-to and so one of the other things that we can simultaneously acquire that would inform both the treatment response verification of treatment response help us predict individual differences and somehow mark response other than symptomatic reporting um i think it's really really important and um right now we're just not there but i think that the technology is going to enable us to be even right now when we're at home with devices we can collect a lot of data on patients um i mean so there's a lot of potential that i don't think it's it's there yet but um right but you you looked at heart rate variability did i understand correctly and but it didn't seem to really give you no it was EEG it was was our kind of the first simultaneous oh EEG sorry EEG yeah okay that didn't really give you the answers that you wanted and so now you're looking at more i guess well less neurological ones or more general kind of as a first step yeah so just things that we can collect and we can definitely collect vitals and monitor vitals in real time through the devices and so you know so but that's definitely an area of great potential because these you know with technology enabling ongoing biometric data collection there's lots of you know ways that we can and i think to your point that that's going to be a very powerful tool in the end for optimization of how we use these tools yeah so so just to sort of be clear about what you think is kind of not there yet because there are there's a lot of you know Fitbits and and various medical grade kind of uh equipment hardware that can be attached to people you know the Apple Watch now has a whole health kit kind of software development kit attached to it um so and but but are you really talking about the kind of the ability to simultaneously capture data across multiple hardware devices and also to integrate that data in a clinically usable way is that the kind of and specifically in response to the use for tES therapeutics so that that i you know across studies across uses is definitely our experience there's a lot of individual differences in responses yeah and especially bringing back to code if you look it's a Covid especially post Covid is multi-systemic right so there's a lot so what indicator can help us with the patient in real life and so that's what always drives us is the clinical challenge and what we can do so that was uh you know so we're trying to think of ways you know and you're right that there's a lot of technology in some ways though there's no clear path to exactly what technology is going to be useful sometimes the complexity and the availability of the technology makes it harder to move forward so we think about that a lot and i think that soon that will be really part of the use of these kind of therapeutics for sure yeah yeah i see so the the the people that are developing the devices are generally not kind of developing these additional tools for the kind of if we can call it kind of there's the obviously this kind of device data where it sort of i think the next step i think that there is i think there's a lot of interesting device designs coming down and i also think that the devices may be able to also deliver digital health therapeutics so carrying the stimulation you know so we believe very much that that uh ongoing therapeutic activity training rehab um with the stimulation the stimulation boost that that benefit um potentiates the benefit and so if the device can deliver the digital therapeutics i think there's a lot of interest early development in pairing the digital health therapeutics with the stimulation therapies as well yeah i see what you mean yeah um because there's a sort of synergy between them and if you can do that one in one fell swoop that would be obvious and Giuseppina from from from your view from your perspective what are the one of the main technical challenges for the next few years say five to five to ten years as a biomedical engineer as a biomedical engineer um for sure like uh being like able to more this like more specific in the stimulation of the target of the brain target so because of course we are using some like electrode montage that are standard but like on standard measure but of course like a lot of difference between um between patients and so i think the next step will be reduce this like variability in between subjects yeah and this maybe you will get like a better like result in terms of like uniformity of response to them so what we i mean what we really need ultimately right is a kind of digital twin a digital double of the patient right in silico uh sort of experiments on and work out what the best what the best sort of location and and and protocol would be for that individual right but that's no no mean feat but i guess are these are these primarily are they primarily software innovations then we're talking about rather than hardware innovations or the is there as much work to be done on the hardware as there is with the with the software I think we need to work more on the software because then the output would be just an adaptation it's just like our like understanding or also in like matching different like uh speciality life like uh um and like um activation of the brain and so what is like all the world of like functional MRI with like all this like trying to find the best target so you mean it's kind of like uh okay so it's more it's almost more like a scientific question how are these how does the brain we need to understand kind of how the brain works if we want to deliver we'll definitely have it solved within the next three hundred years i was gonna say if we have to wait for that so that's that's great but i think you know the you know so individualization of you know so i think in the end i would agree with you in that we'll have technology that'll drive individualization of the therapeutic use and also combined across different stimulation therapies um so uh so you know yeah invasive and non-invasive and i you know we're very early and a lot of exciting advances on the forefront with all of this for sure for sure and yeah i mean also i guess yeah there's so many different um so many different targets and you know as you say there's why why wouldn't two targets be better than one you know maybe there's and so the complexity just becomes bewildering doesn't it when you think of the number of places that you could kind of relocate your stimulator and the number of different parameter settings i had a chat with a Boston Scientific engineer the other day about their their basically their their their their vanilla DBS system if i can call it that and even that has 250 million different possible settings you know so that the chances of finding you know the optimal ones for any given patient scene right in settings matter i think you know we always just call attention to dosing parameters because if you look at reviews of studies uh you know some studies use one session some use five everybody uses different intensities and all of that probably matters i mean if you made an analogy to medicine if somebody took one pill somebody took five pills somebody took one dose something to and then you try to analyze the fact of that one drug across all these studies so you know the dosing parameters matter tremendously just even with our standard first entry so we can understand yeah well a lot of interesting work to be done in the future and no doubt you guys will be instrumental in doing much of that work so we'll look forward to uh to seeing uh and following your your work in in the future um so yeah i'm afraid we've reached the end of our allotted 30 minutes but it's been a pleasure Giuseppina and Leigh to speak to you both and just for for just bringing attention to all these great topics that you're covering so um we really appreciate the opportunity to speak with you brilliant thank you very much
2021-09-02