Kris Pahuja Podcast Transcript
Kris Pahuja joins host Brian Thomas on The Digital Executive Podcast.
Welcome to Coruzant Technologies, Home of The Digital Executive Podcast.
Brian Thomas: Welcome to The Digital Executive. Today’s guest is Kris Pahuja. Kris Pahuja is the Co-founder and Chief Product Officer of Piramidal with a proven track record in AI and healthcare from significant roles at Google and Spotify.
His mission is to revolutionize medical diagnostics by teaching AI to read brain waves, enhancing efficiency and accuracy in healthcare.
Well, good afternoon, Kris. Welcome to the show!
Kris Pahuja: Thank you. Thank you for having me.
Brian Thomas: Absolutely. I appreciate you making the time. I’ve got some great things to talk about today around your company and you being the Chief Product Officer.
So, I’m going to jump right to your first question. Kris, could you share the journey that led to the founding of Piramidal and your vision for revolutionizing medical diagnostics using AI and brainwave data?
Kris Pahuja: Absolutely. So, we founded Piramidal, my co-founder and I last year in July. But we’ve known each other for a little bit longer.
So, my co-founder Dimitri, he did his PhD at King’s college in neurology and AI. And he always had this vision that why are we looking at brainwaves in the hospital visually without any sort of analysis on it. And so, he made it his PhD project and developed some really cool tools, made some publications.
And decided to commercialize it down the line. And then that’s when he and I met and I started advising him for a little bit on, you know, what are some of the possibilities to do that and decided that maybe we should partner up, start a company and build a product around it. So that’s how we started a little bit.
We decided to apply for funding, apply for Y Combinator, got into Y Combinator in January of this year, and did the program. And that’s how we started the company. In terms of our vision for revolutionizing understanding the brain, the idea is to build the largest model to have ever existed around the brain.
It’s a model that is trained on brainwave data. I’m very similar to how chat GPT is trained on text, which is human language. For example, Our goal is to build a model that is trained on brain language. So, we understand everything that’s happening in the brain. So, all these things we think about our brain that are currently subjective, like, you know, if I have pain or if I have, you know, if I’m focused or not, it’s all subjective thoughts that we have.
Can we quantify those things? And that’s the long-term vision of really understanding what’s happening in our brains.
Brian Thomas: Thank you. And I really love that. I’ve been in healthcare a long time myself on the tech side, and I always thought working around monitors that are monitoring, whether it’s the pulse rate or the EEGs, those sorts of things I thought was amazing and fascinating, and I’m glad that you’re in this space trying to make the world a better place and helping patients.
So, I appreciate that, Kris. And Kris, beyond epilepsy diagnosis, what other neurological conditions can Piramidal AI’s model potentially diagnose? And how soon can we expect these applications to be available?
Kris Pahuja: So that’s a really good question. And in reality, anything that uses EEG currently, we can diagnose.
with Piramidal in an automated way without any human intervention. Now that’s the first side of the equation. So that doesn’t mean that we cannot identify new biomarkers, right? EEG is widely used currently, but it does change like, you know, thousands of times a second and our human eye can only detect certain patterns.
So, the more longer-term vision is how do we detect for things that the human eye cannot currently see. And what are those biomarkers that we can identify? So, in terms of timing, you know, we have a research effort ongoing with a bunch of universities and PhD labs. And, you know, as the model matures more and has more generative properties, we’ll start to see those biomarkers for other disorders that currently don’t exist.
For example, Alzheimer’s, Parkinson’s. Mental health disorders, for example. So, the idea is to really expand into quantifying the brain for all sorts of disorders, starting with epilepsy.
Brian Thomas: Thank you. I like how you highlighted some of the things that the human eye cannot detect. Right? So, I’m glad it’s a, I know it’s a partnership here between machine and human.
And we talk a lot about this in other areas around AI, but I really appreciate you highlighting that. And Kris, the shortage of neurologists is a significant issue. How does Piramidal aim to alleviate this problem? And what role does AI play in bridging that gap?
Kris Pahuja: It’s definitely, it’s a huge problem. And you know, it’s not only neurologists, it’s also EEG technicians.
It’s ICU technicians, every hospital we talk to, there’s an extreme shortage. And a lot of this work is currently getting outsourced to companies where they hire a bunch of people looking at the stuff because the hospitals don’t have enough people. And the way we can alleviate that problem is just by automating the process.
So instead of a doctor, you know, taking an hour to look at a 24-hour EEG or multiple hours to look at a week-long EEG, we can automate that entire process and have it happen in seconds. So now a doctor can go through multiple patients’ data in a matter of seconds. Few minutes. That’s what’s the cool thing about AI, right?
Is it’s assisting the doctor. It’s not calling any diagnosis so it can pick out what the doctor needs to see in an easy show it to a doctor. The doctor is like, okay, that makes sense. That’s an epileptic seizure. That’s status epilepticus, for example, and then check, check, you know, create the report onto the next patient.
And the doctor can go through a lot more patients quickly. So, you need a lot less neurologist to actually cover the number of patients that are out there because of that.
Brian Thomas: Thank you. And again, we are introducing a lot of this new technology and machine learning and AI to help augment the work that humans do.
And again, another point that I’d like to highlight there. And Kris, last question of the day. Can you discuss how Bromadil’s technology can be integrated with devices like VR headsets, you know, the Apple Vision Pro or Medi Quest to enhance real time brain activity monitoring and interaction?
Kris Pahuja: Yes, absolutely. It’s a great question. I’m glad you asked that. And I don’t know if you noticed as well, but Apple also patented the Air Pods to have to record EEG as well. So, it does look like the signal in the market is, you know, we’ve tackled EMG through Apple Watch and others. Can we move on to doing the same with EEG with headsets and quantify what’s happening in our brain in some way?
So, the way we, you know, in the long-term we envision this is having a model that empowers all these analysis around the brain. So, if a VR headset or the vision pro is recording EEG and has EEG sensors, anyone should be able to build an app on top of it to leverage by the model. The same way people are leveraging LLMs to build all sorts of applications around the brain.
So, an example could be, let’s say, someone who wants to build a meditation app. But then you also want to see how you responded to that meditation. What happened to your brain during the meditation? And can we give a score of how much it helped you? That could be built on top of wearing a Vision Pro on top of our model, where you actually know the benefits of that meditation versus potentially a placebo sometimes of feeling better right after.
So, you know, the opportunities are endless for us to really track what’s happening inside us with these consumer devices.
Brian Thomas: That’s awesome. And I really want something like that to come out and go mainstream. And I, I hope it’s Piramidal. I hope it’s you guys who get that out there. I think that would be awesome for actual patients, you know, real people to be able to, to see some of that stuff, because I know things like meditations are very helpful and it’s, you can actually, when you see something, visualize something, I think that people will embrace meditation more.
So I appreciate that. And Kris, it was such a pleasure having you on today. And I look forward to speaking with you real soon.
Kris Pahuja: Absolutely. Thank you for having me.
Brian Thomas: Bye for now.
Kris Pahuja Podcast Transcript. Listen to the audio on the guest’s podcast page.
