Episode 61: The Autumn of AI in Biotech with Key Opinion Leaders Elaine Hamm and James Zanewicz

Want to hear 2 dumb-dumbs in physics talk about the Nobel Prizes in chemistry and physics?

And so much more. It was the autumn of AI.

It was. You're listening to Bio from the Bayou, empowering biotech leaders with the tools, assets, and expertise to improve healthcare outcomes.

Each episode provides insights and information to expand your network, advance your product line, and help your company thrive.

Serving the entire Gulf South, but recorded in the heart of New Orleans.

Welcome back to Bio from the Bayou. I'm James Zanewicz, the Chief Strategy Officer from Tulane Medicine.

And I'm Dr. Elaine Hamm, the Executive in Residence at Tulane School of Medicine. And James, I'd love to talk to you today about AI, in particular, some of the recent events that have happened in AI.

I love AI, and especially this fall, because we've had some really, really cool stuff going on. AI's come a long way this year.

Yeah.

And I guess maybe we should start about what we hit first, which was our annual meeting for the Alliance for AI in Healthcare, which you are executive director of.

I was, and well, I am. I, I was literally just thinking back to the event. It was an incredible event. We had our annual meeting for all of our members and some non-members as well, and it's really a great opportunity for people within the industry to come together and talk about some of the biggest challenges that we've been facing over the past few years and find solutions for it. So we had some great speakers. It was a little bit of a mixed bag of speakers plus panels, and we even had a special sort of breakout session where we broke into groups and talked about some of the biggest challenges we're dealing with in AI so that we can create our white paper on the AI Playbook, a resource that we're working together on to really help people that are in the industry and people that are wanting to be in the industry.

And we had your favorite thing, scientific posters, which I know you loved. You added those, you're like, we're gonna have them. And I was like, no one will love them. No one will love them. Everyone loved them. Everyone loved them. Elaine was right.

Shocker. I know, I know. And now it's it on the podcast that I was right. I love it. We had some incredible speakers too. So we kicked off the event with the CEO of Genentech Roche, Ashley McGargie, which she's fabulous. And she really helped lay out what Genentech has been doing in terms of AI and their drug discovery and development process.

And Genentech actually hosted us, which was really interesting. We were on their campus. It was very, very cool.

Yeah.

To kind of get that look. And we had really interesting panels. My favorite this year, probably because I moderated it.

Yeah, I was about to say, is it yours?

It was, but it was really interesting because it was the first time we did kind of a healthcare data access panel. And we had several different perspectives, anywhere from kind of international ethics to actual people who deal with data to a hospital system. We had Ryan Hildebrand from LCMC Health on that. And his perspective, I think, was really interesting for folks to hear because we have so many members who are focused on developing things, We haven't really had as many of the end users before. And his perspective, I think a lot of people stood up and took notice.

Yeah, no, and he gave some really good practical— when you think about sort of AI in the healthcare industry, sometimes you really start thinking about the deep sort of scientific technical aspects of it, being able to, you know, read radiological images and pathology slides and be able to sort of report and predict, you know, potential tumors and things like that. But he really got into some of the more basic aspects of utilizing AI in healthcare. And that's helping take really repetitive, boring tasks that can really bog down doctors, healthcare administrators, and turn that into a more streamlined process. And instead of thinking about the really, really deep use of AI, it was the most practical use of AI.

Yeah, and I guess there's a lot to talk about from the autumn— so autumn of AI, as I like to call it. But we mentioned Ashley, But I wanna talk about Rayyouth, Lana, and Jenny Kay.

Yeah.

The women in AI.

Yeah.

And what they have done growing from 3 women drinking wine on their porch talking about AI to 12,000 members.

Yeah, absolutely incredible. And we have a podcast episode featuring Rayyouth and hearing about her journey of creating that organization from a conversation with her friends to an incredible organization.

And that came about, she told us, because they were looking at the Time 100 people in AI and there was not one single woman on the list. And they decided to solve it. And boy, have they.

Boy, have they. Yeah. No, it's incredibly impressive. And that episode, we really do a deep dive into sort of the challenges of being a woman in AI.

Yeah, so we encourage everyone to check out their podcast, to check them out. They are partnering with us on another of my favorite autumn initiatives, the AI in Healthcare Design Lab.

Oh, yeah, that's coming up.

We just finished Bio in the Bayou. We're rolling straight into another event 2 weeks later because that sounded so smart at the time.

Sounds like a good idea at the time.

I had a dream. But they're coming in and bringing bringing several faculty members to help coach and mentor these teams. We have 30 really stellar applicants from across the globe coming in for an immersive biodesign-type weekend. Very, very excited for this, for the program, for what it's going to do for the Gulf South as far as putting us on the map in this space. Traditionally, it's been more of the richer regions, the Bay Area, Boston, San Diego, where they really lean into AI and healthcare.

Yeah.

But we have really amazing data here and a really amazing patient population who could benefit from a lot of what AI can bring to healthcare by giving access. Or we were talking at Bio on the Bayou about AI in healthcare, and is it better to have an AI solution that maybe isn't perfect but at least is something, or have nothing for healthcare? And perhaps the risk is worth the reward.

Yeah, we had another podcast that we did, and I hope that you guys will check it out, but it was specifically on Health Universe, and that was a great episode. And he's basically making it easier for you to make an app for AI. But we also talked very deeply about how AI is— people are a little bit afraid of it, especially as it relates to healthcare. But at the same time, he said, as compared to what? Is it worse, as you say, to not have healthcare or perhaps risk it with an AI-based solution? So it was an interesting conversation.

Before we start to wind this up a bit, the biggest AI event of the fall were the Nobel Prizes.

Yeah.

And I know you, having taken graduate-level courses on this, are dying to talk about the chemistry one. Even though I have a degree in chemistry, she's stealing it from me.

I know. Well, I want to crack my knuckles before we get into this. I'm going to talk all about protein prediction. But really, so the Nobel Prize in Chemistry, with one half of it went to David Baker at the University of Washington, at Howard Hughes Medical Institute as well, for computational protein design. And then the other half jointly went to Demis Hassabis and John Jumper at Google DeepMind on their work for protein structure prediction. I mean, I think back to, I mean, you mentioned the graduate-level courses, and man, I wish that young, naive graduate student could know that one day you could use these types of technologies to do what took us, you know, one, it may not even have been possible for us to do, but being able to predict protein structures is a huge, important piece of being able to design new drugs. And here we were trying to do it, you know, based on amino acids, and it's just an incredibly difficult and challenging problem, a 50-year-old problem of being able to predict protein, you know, proteins are very complex structures. These discoveries really hold enormous potential as it relates to a number of different aspects to medicine and other things.

And then from chemistry to physics, I mean, physics was also an AI win this year, and that was John Hopfield and Geoffrey Hinton. And they did really groundbreaking work in the field of artificial neural networks, specifically foundational discoveries and interventions that allow ML or machine learning to use those neural networks to do things faster, better, store data. So Hopfield specifically developed a structure for storing and reconstructing information. Well, Hinton invented a method to identify patterns within data, so almost like sub-patterns that aren't normally seen, which laid groundwork for advanced AI systems in this space and was just really, really cool. And I always love talking about physics since it's the one thing that almost kept me from graduating college.

Me too. Ironic, me too.

A wonderful branch. Yes, I remember being in and begging my professors like, I have worked so hard, please give me a C.

Please.

Did great in my English classes. Wasn't my major. No, but I mean, that was what was really exciting to me is that happened right as we were at the AAIH annual meeting.

Yes.

We were able to talk about like, hey, the first one came out, then the second one happened, I think while we were there.

Yeah.

And we were like, this is the autumn of AI.

Yeah.

The end.

Yeah. Hence the name of this podcast episode.

Exactly.

So, well, let's talk a little bit about the AI tools that we use.

Yeah. Well, I use Jasper. I love it.

You love Jasper?

Jasper.

Yeah.

'Cause I use Jasper for marketing. It is an AI tool specifically for marketing communications professionals. So that half of my job, the non-BD half, It actually can produce a press release. It can help you write in people's voices. You can upload samples of writing and create that as a method and a tone that it always writes things in. So you say, hey, I want to write like Dr. Elaine Hamm today, and tomorrow I want to write like Ian McLaughlin, and today I want to write like myself. And it does that really well. Yeah, I mean, I could, I could be sending emails from you and you don't know.

Which James is in the room with us right now?

Deepfake James. Deepfake post bio in the bio James. But that, I mean, I use ChatGPT as well. I know folks on the team love and use Claude.

I love Claude.

I know you do.

I just started using it. So yeah, ChatGPT, first of all, the paid version is so much better than the unpaid version. And unfortunately, you have to kick in a little money to get that better version. But I recently started using Claude, which is produced by Anthropic. Anthropic also gave a talk at our annual meeting as well. They're doing a lot of work with Genentech in addition to a number of other people and Amazon. But anyway, Claude is really interesting from a scientific perspective. It was recommended to me from a colleague from Elanco, Karthik. He was telling me, he's like, no, I like Claude so much better. It's a little bit more scientifically minded in a lot of ways. And so I've been kind of going through and comparing them, you know, results of looking and summarizing research articles that I've found. It's a great use of how you can use a ChatGPT-like function is being able to put in a research article and say, summarize this, what are the main talking points and doing a comparison between the two. And it's really interesting.

I know there are a myriad more out there that people use. I know like Mark Sadam at NYU Langone is using AI to do his entire tech transfer back office. Like they're analyzing information disclosures, they're using it to streamline negotiations, to draft agreements. Like he is so far ahead of everything. He's like, my staff have time to do a lot more than they used to. Yeah.

Yeah. You think of the repetitive task. Again, it's not always the really hard things that you're going to use AI for. It's like, what are the things that really drag you down? What are the bottlenecks? Is it doing the same sort of repetitive task? That's the perfect use for AI.

It's not going to replace people, but people who use it will replace people who don't.

100%. Yep.

That's a pretty good wrap-up. That's a pretty—

yeah, I, I like that. Shall we leave it here?

I think so.

I mean, there's much more to come. There'll be the spring of AI, I'm sure.

Of course.

And there'll be new all kinds of exciting inventions coming out, and we'll see, you know, again, going back to the Nobel Prizes, I mean, some amazing things can come from these types of discoveries, and Being in healthcare is an incredible opportunity to bring new drugs to the market, devices, and things that can really help patients. So it's an exciting time to be in healthcare and AI.

So anyone who's interested in that should reach out to Elaine through the AIH and perhaps become a member and have their voice heard.

Yeah, no, absolutely. And as always, we'll have that information in our show notes. We'll put links to the AI website. You can see some of our posts on LinkedIn. We've done some snippets of the videos from the conference, including Matthew Lundgren from Microsoft. He's the Chief Data Science Officer for Microsoft Health. Really great talks that we have on a variety of topics as it relates to AI and healthcare.

And I think that's it.

You good?

It's a wrap.

That's a wrap.

I mean, I don't know if I'm good, but I think the podcast is good.

There we go. Well, thank you all for joining us. Thanks for listening, and check out those show notes.

Thanks for joining us for Bio from the Bayou, and we hope you'll join us again.

To never miss the latest updates in biotech, hit the subscribe button now.

We'll catch you on our next episode of Bio from the Bayou.