Host Natalie Watson
Hello and welcome to the BLA Connections: A Clear Voice, Series 6 – this is the finale! We’ll be closing with a collection of five podcasts, which we recorded back at Cutting Edge Laryngology, 2024 in London, UK. We have three amazing keynote speakers, Marc Remacle, talking about his 30 years; ‘What did I learn?’ Anais Rameau from New York, talking about AI and the future of laryngology. And Nupur Nerurkar on vocal folds, cysts, sulci and mucosal bridges.
We also listened to Martin Birchall and Marc Remacle talk on hard and soft robotics. Martin was one of our previous presidents. And we finally have the remarkable, amazing Professor David Howard, our first president of the BLA, talking about his career in laryngology, and also what he’s doing now and what he is planning to do in the future.
With regards to Cutting Edge 2024, we’re very pleased to announce that all the talks and presentations are live on talking slides, which you can access if you’re a BLA member or delegate of the conference. We hope that you enjoy our podcast. Please do have a listen.
Host Natalie Watson
Hello and welcome to the BLA Connections: A Clear Voice. I’m your host, Natalie Watson, and we’re here live at the Cutting Edge Laryngology Conference 2024. I’m delighted to introduce Anais Rameau, who is working in the Weill Cornell in New York City. She’s here as one of our amazing keynote speakers, and today delighted us with all the information we needed to understand AI, artificial intelligence, in laryngology, welcome!
Dr. Anaïs Rameau
Oh, well, pleasure to be here!
Host Natalie Watson
It’s a real delight to have you here today. So tell us, you spoke very eloquently about AI in laryngology. For all our podcast listeners, what is AI in laryngology?
Dr. Anaïs Rameau
AI is a method. It’s not really a new science within laryngology. It will allow us to process a lot of existing data, and new data, in ways that we were not capable of doing before. The hope is that this will allow us to create some tools to improve access to care in laryngology, especially areas that are underserved in terms of voice, airway and swallowing care, but it is not really a new branch of science.
Host Natalie Watson
Yeah, and I love the fact that you and your colleagues have coined a new word ‘Audiomics’ – tell us more about that.
Dr. Anaïs Rameau
Sure. So we basically had observed that there were many more papers that were coming through that were focused on the potential of voice as a biomarker of health, and we thought that this needed some standards, and some community discussion about the protocols for data collection, and how to best process the data, how to best create standards for sharing data, and we felt that creating an umbrella name would allow us to really kind of create a home for that community. So that was our goal. And this was published earlier this year, and we’re seeing some people starting to adopt the word, and we’re hoping that will also allow us, whenever we’re doing literature search, that we could use ‘Audiomics’ to find some of the papers that are relevant.
Host Natalie Watson
Amazing. It’s a real interdisciplinary term as well. You’re working with respiratory physicians for cough, and you’re mentioning the importance of these in dysphagia as well. And I think it’s a really exciting field. And you heard it here, it’s ‘Audiomics’, and keep an eye out for it. The other thing you were interested in was ‘Videomics’ – so tell us about that?
Dr. Anaïs Rameau
Yeah, so ‘Videomics’ was a term that was created by another team, Alberto Paderno and Chris Holsinger at Stanford. And so ‘Videomics’ I think, is the most tangible application of AI within laryngology. It’s basically the application of AI, or computer vision, to our video laryngoscopies. And as I mentioned during the talk, one of the issues that I see in laryngology, and I feel like I’m still early in my career, I’ve been in practice for seven years. Is that we don’t have common templates for describing stroboscopies. We each have our own ways of doing that. I don’t know if you have a standard in the UK, but I see a lot of variability, and we don’t, it’s all qualitative assessments, and I think that AI has the potential of bringing that objectivity, and standard, and allowing us to really keep better track of outcomes, and also having a standard for diagnosis. So I think that’s really where I see the most potential, right now.
Host Natalie Watson
Yeah, so we’ve got Audiomics, we’ve got Videomics, and now in general it is encompassed by AI, but as you were saying in the lecture, AI can be used for good, but also there’s some reservations about AI, and some considerations that we need to be aware of as laryngologists about interpreting data from AI, and also how we roll it out to the to the global community. Can you expand a little bit more about the concerns that laryngologists should have with AI in laryngology?
Dr. Anaïs Rameau
Yeah, if I may, I’m just really thankful for the time I spent in the UK. Previously in my education, I spent time at the LSE studying health policy, I spent some time at the Nuffield Council on Bioethics, and I went to Cambridge to do a degree in Philosophy of Science. So I love thinking about ethical issues, epistemological issues, and certainly there are so many when it comes to voice.
Primarily, voice is a biometric identifier, and that’s very strongly imposed by our IRBs in the US. So it’s very hard to share voice data. And voice is also part of our social persona, and not everybody wants health data to be associated with that. So that’s one major issue. The other big issue that I see, especially with Audiomics, is it’s great to screen for disease, but if you don’t have a treatment, or if you don’t have access to treatment, then is that really worth it? Because I think that it can also be very negative to just have a diagnosis and not have a management plan.
Host Natalie Watson
Yeah, yeah. We know from studies, cough, from our previous podcast, is associated with genetic conditions that can predict neurological conditions 20 years in the future, and the same with voice, we see in our clinics day to day neurological conditions that start with voice problems. And as you say, if there isn’t a very clear treatment protocol to help with those, it’s a minefield at the moment. So we’re blessed with all this knowledge, then what do we do with it? And it’s a bit like in Gattaca, you know, getting your genetic results and figuring out what your life has in store for you. You know it’s a coming reality.
Dr. Anaïs Rameau
That’s right, but, I think, where the value could be with voice as a biomarker of health – so let’s say there is a new drug that comes and you want to monitor its impact on your health. Then you can do that at home, in your own setting, and you could do that on a regular basis, where you don’t have to go to see the doctor, like once every few months. You can actually monitor yourself on a daily basis.
Host Natalie Watson
And what is great about that is, it’s not like the white coat syndrome, where you’ve got the blood pressure high when you see a doctor, in normal life, normal experiences, you know? And so often our patients come and say, Oh, it was really bad a week ago, but obviously not now. So to have that ability to have the control of being able to display when voice is not entirely normal, then that’s really beneficial.
Dr. Anaïs Rameau
Absolutely. Yeah.
Host Natalie Watson
The other things that I really loved from your presentation were the kind of, out of the box ways, to improve health with AI and laryngology. And you mentioned a few really interesting things and the EMG based silent speech device. Have you any experience of that yourself?
Dr. Anaïs Rameau
Yeah. So we actually have a patent on this approach, and we have one publication, and right now I have a postdoc who’s doing a two years training program to become the CEO of that company. So we’re very excited. We’re not the only one in this space of brain computer interface, but basically, we’re hoping that we can find alternatives to either the electrolarynx or the TP for patients who don’t have a voice from laryngectomy, or for other conditions.
Host Natalie Watson
Amazing. What an exciting time we have integrating the computer scientists, peer physiologists and clinicians, and patients, and advocates, all together, to public policy writers as well, just to come together to really forward medicine, and to forward public health.
Dr. Anaïs Rameau
Sure, yeah. No, it is an exciting time, absolutely. And I want to encourage everybody to educate themselves about AI, because I’m not a computer scientist. I learned this by self education, and it’s really important, because we’re going to be depending on those tools more and more moving forward, and we want to be critical and be good advocates for our patients, and I think it’s very important to not just leave the technologies to do this and trust them. I in fact, really think that clinicians need to be involved along with patients.
Host Natalie Watson
Yeah. So how do we? Are there any projects, or anything that you would like to promote, to say that we can get involved with, or anyone who’s listening can get involved with, and we can write in the show notes?
Dr. Anaïs Rameau
Sure. First of all, I always struggle as an associate editor for machine learning and AI for the Laryngoscope to find reviewers. So I think that if I send you a paper, it would be really great if you really try to look at the paper critically, because oftentimes we actually need the clinicians to look at what data was used, how it was processed and whether that makes sense. You don’t have to worry too much about the technical aspects of the AI. So that’s one way to get involved.
The other way to get involved, as I mentioned earlier, is to be advocates for your patients. Oftentimes, we’ll have patients who come and say, my automated speech recognition system, Siri or whatever, Alexa is not understanding me, and making notes of these things, understanding your patient’s experience, and seeing how you can be an advocate for them, so that technology can be also sensitive to these limitations.
Host Natalie Watson
Yeah. Brilliant. I mean, I love the fact that you have this advocacy role, and you share this with all of us, and remind us of our role too. So yeah, so thank you so much.
Dr. Anaïs Rameau
Yeah, you too.
Host Natalie Watson
Thank you so much for all that you’re doing, helping with the interface of clinicians and AI and the patients, and coordinating, and presenting it today.
Dr. Anaïs Rameau
Yeah, thank you so much, it’s a pleasure to be here. Thank you.
