 Hello everybody, I'm Jan Willen-Wassman and I'm going to share some ideas about computational audiology. I've chosen a playful approach and I hope you will appreciate it and stay in the game. Here is the group of authors and collaborators with whom I've been developing ideas about computational audiology. And our mission is to help people with hearing loss with the best available treatment and tools. And there's more than 1.5 billion people worldwide with some degree of hearing loss. And with conventional audiological care we cannot address the global burden of hearing loss. We do not have enough capacity due to lack of professionals, equipment and efficiency. And here we might benefit from distributed services and distributed expertise and collaboration. And anyways, we will need new tools and ways to provide patient-centric care. And there's many developments going on. For instance, deep learning on medical images, big tech companies entering the medical market, wearable technologies, and all these developments are transforming healthcare and are also changing audiology. So what's computational audiology? You could say it's a model-based approach to audiology using modern data collection tools, including remote or virtual tools. And as part of the virtual computational audiology conferences last year and also the upcoming conference, we have collected many examples of promising research projects or resources and innovations to illustrate computational audiology or to share knowledge and, I hope, to foster collaboration. And here are some examples of technologies that could improve access to hearing health services. For instance, automated audiometry or speech-to-text apps or consumer electronics that is nearing the functionality of hearing aids. And what's interesting is that access to mobile devices is not an issue anymore. The coverage is more or less the same in developed countries and developing countries. So that means that if something works locally, you can apply it globally. Well, of course, you have to think about local restrictions. And there's also many examples of computational audiology that lead to improved precision or may lead to improved precision. For instance, active learning based machine learning audiometry could be used to assess what's the most informative next stimulus when performing audiometry, which increase precision or reduces time commitments. Or you could build better patient profiles. For instance, based on genomic profiles. And here's an example of a study by Wells et al. who collected genome-wide sets of genetic variants from a large population, 250,000 individuals, used to find the origins for self-reported hearing difficulties. And other examples of building patient profiles are the examples given by Ro Sanchez Lopez in his talk. Or another way of improving precision is using deep neural networks, for instance, to simulate the auditory system. So I would like to tell this story or put this story into a game. And many human activities can be regarded as a game. And here are two types of games, finite games and infinite games. And finite games have clear defined rules and boundaries, and all players know what the objectives are and what they need to do to win. An example is the game of chess. Infinite games, on the other hand, have no clear rules and there's no winning or losing. The only objective is to keep playing. And healthcare is an example of an infinite game since we will never be able to cure all people and end the game. So let's adopt an infinite mindset and play the game of computational audiology. And since I cover a lot of virtual elements, let's choose a video game as example and turn an existing game into the game of computational audiology. If you have the will and resources to play, press enter. We are in the game now. And anyone can join this game. It's an infinite multiplayer game. Anyone can join who has the will and resources to play the game. And sometimes players quit the game. For instance, a clinician who retires or new players enter. And you can choose a role, be a scientist or a clinician, developer and so on. Of course, patients are also in this game. And what I hope to demonstrate is that our roles develop and our training continues and our rituals change over time. We cannot choose the rules, but collectively we might change the rules. And can we unlock new superpowers and build new characters? Here's an example of a superhero called Microface who was created in 1942. It's a human sound amplifier. He has a built-in microphone in his mask and a voice amplifier. And isn't that a kind of super audiologist? What I find a fun fact is that he can attach telephone wires to his mask to make phone calls. It was the state of the art in 1942, this technology. But Microface was updated this year. And I think the lesson we can draw is that characters develop even when in a strong position. And our field is innovative and keeps inventing. And this means we as players need to evolve too. And here's another fictitious player. This is Alan, a virtual audiologist that uses computational powers to help patients. He uses algorithms and data collection tools and shares his knowledge with patients and care providers. What I find a fun fact is that he has wireless capabilities, as you can notice this big antenna on his head. And he wears the audio public domain mark, which is a symbol to indicate that the sounds he produces are free of known copyright restrictions and therefore in the public domain. And the number of possible players in this game is infinite. Here's another random example for let's say a virtual ENT doctor. And yeah, who knows how these virtual players will look like in the future? And will they show human-like responses? Are we able to distinguish them from humans? Or would they be equivalent in intelligence, which could be tested with the Turing test? And this artificial photo here that was created by a deep neural network, it's not an existing person. You can make those pictures yourself by the website stated here. This person does not exist.com. And anyway, we face a number of risks we cannot solve on our own. The entire hearing community, including scientists, clinicians, developers, companies, policymakers and patients will need to work together on this to avoid unethical use, reduce bias and protect privacy. We need to make clear who is responsible and can we deal with vested interests from players that benefit from the status quo? Or do we know how to deal with professional resistance to change? And here I think comes in the role that EVA's member societies could play. First of all, we should anticipate what developments mean for our practice and it will probably also depend from country to country. But be aware that the game of audiology has no boundaries. So what players do in Europe affects players in Australia and so on. So again, collaborative efforts are key. Are we able to critically appraise new solutions, share best practices, develop user requirements so that clinicians and patients can safely use AI? And can we define standards so that we can more easily share tools and data? And what could virtual EVA's members do? And this is of course just a thought experiment by giving human traits to machines to provoke thoughts on how to assess virtual care or hybrid care. Here I mean a mix of virtual and physical care or AI based audiological services. So imagine Alan and his friends, his virtual colleagues, want to become licensed virtual audiologists. How are we going to regulate this so that we can safely use AI, protect against unethical or unfair competition and advocate equal pay? And here I mean a fair pace of the cake for everybody. And as part of the hearing community, we as researchers, developers and clinicians can strengthen our collaboration so that we can responsibly unlock new powers for our patients. We can share best practices, improve cooperation and build a community that uses similar tools and data sharing pipelines. So for this we are looking for experts to review and rate resources, nice examples, so that we can learn and share the best online tools with the community. We collect them on the website. And for examples from other fields, have a look at OneMind.org. They have for instance an app that's providing healthcare workers an easy way to track their mental health. And the app also provides links to access for immediate support and mental health resources for issues like sleep and stress. So that caregivers stay healthy. But based on recommendations by peers. And that's I think how we can help each other in reaching a common goal. Thank you for your attention. We've run out of time and credits. You can of course insert resources to continue. Enjoy the game and thanks again for your attention.