Episode 269 - Image Recognition Technology for Health with Susan Conover
Today's guest is Susan Conover, cofounder and CEO of Piction Health, which brings Image Recognition AI technology to the practice of dermatology. This Boston-based company has started to launch their product throughout New England. Susan talks to Max about how the technology could be used to help patience more effectively, and the challenges of bringing change to healthcare and being an entrepreneur.
Susan Conover
Don't get caught in a local maximum. It's easy to focus on how to optimize your current life, but important to realize your life may get a little worse in applying for jobs or vetting a startup idea on top of everything else. But that's what opens up the door to get to an even higher local maximum at your dream job or a new venture. Don't lose sight that you could be at a local maxima because likely there's more out there for you. You just need to do the work to get to the next part of the curve. - Susan Conover
Susan Conover: LinkedIn | Twitter | Facebook | Instagram
Links
Piction Health: Website | YouTube
PyTorch : An open source machine learning framework that accelerates the path from research prototyping to production deployment.
MIT: Massachusetts Institute of Technology
The Local Maximum: Email
Related Episode
Episode 258 - Sports and AI: Jason Syversen Changing the Game with SportsVisio
Transcript
Max: You're listening to the Local Maximum episode 269.
Narration: Time to expand your perspective. Welcome to the Local Maximum. Now here's your host, Max Sklar.
Max: Welcome, everyone. Welcome. You have reached another Local Maximum. Today I want to talk about this. There's been a flurry of activity in the space of AI startups recently. And this has been predated by several years, I think, from the recent interest driven by Chat-GPT.
These are real businesses that are starting up that are focused on solving real problems, which is a really good place to be in an industry. I know there's a lot of bad news with startups recently with the Silicon Valley Bank. But, you know, I think I'm going to have to get Aaron on to talk about that to you give some good analysis on that. I've started kind of soliciting ideas on the Locals.
But today, we're going to talk about the good news, the businesses that are being built up with AI that are starting and starting to sprout all over the place, or have been sprouting for many years all over the place, to be honest.
I think part of this is driven by the amazing amount of technical tools and knowledge built up over the years, helped by research institutions, some of which some of the university, some of big tech companies, lots of work by lots of people over many years, but then also, those companies themselves not being able to deliver to the market all the benefits that these technologies might be able to bring forth.
So we're going to be talking about some of these examples this year, certainly, and today, and some of the entrepreneurs and developers who are making this happen. One example is the recent interview with Jason Syversen in Episode 258 that I did on SportsVisio. In fact, you'd be interested to know that today, yes, this day, March 13 2023, I started a new job at a company called winware.ai, and I'm sure Aaron will interview me on that at some point.
But today's interview isn't about those. Today's interview is all about health and dermatology. If you have ever worked with or have used our healthcare system recently, and you listen to the show, you might be wondering whether they can take advantage of the proliferation of AI and deep learning technology that is available. So we're going to talk to the founder of a company in that space, a company that is helping dermatology customers with vision recognition, AI.
This is really interesting. This is really exciting stuff. If you think about vision recognition from the home. In fact, you'll hear in the interview that they have launched in New Hampshire, they launched in Connecticut, and most recently, I went to a launch party of theirs in Boston on Thursday night because they launched in Massachusetts. So they're growing, they're expanding states.
My next guest is the CEO and co founder of Piction Health. Susan Conover, you've reached the Local Maximum. Welcome to the show.
Susan Conover: Thank you so much for having me.
Max: And thank you so much for coming in person in the Local Maximum studio. It's always a lot of fun to to do that. And while we have the studio up here, it's always great to talk to like entrepreneurs in the in the Boston area.
So thank you so much for coming out. You found an interesting application of image recognition with Piction Health. So let's just start, just tell us about it.
Susan: Yeah, absolutely. So I was originally diagnosed with melanoma when I was 22. I tried to go see a dermatologist and was told it would take at least three months to get in. And so I went to my PCP who biopsied my mole and ended up being a stage two.
So, you know, sort of basically like, if I hadn't, if I, if I had waited, I might not be here. And so that sort of sparked my interest in dermatology and in access to care. But I'm originally trained as a mechanical engineer, worked in management, consulting, and then focus my product master's degree on on this topic at MIT.
Max: So this is very personal for you. So tell me a little bit about the product that you've created and like what people are what people are using it for currently.
Susan: Yeah, absolutely. So we have an end to end, like dermatologist experience for patients at home. So they go to our website, on their phone, answer a few questions, take a few photos of their concerning skin, hair or nail issue, and then submit that case and then within 48 hours, we have one of our dermatologists review that case diagnose that case, develop a personalized care plan for that person.
And so it's high quality care, convenient, affordable. All, because our ultimate goal is to make sure everyone can get access to high quality dermatology care.
Max: So where does the image recognition fit into this? And why are you able to get people dermatologists so fast when it takes, like you said, months to get in?
Susan: Yeah, of course. So our machine learning really comes in on our back end, there are many different points that can be inefficient in a sort of regular experience. Like, if you don't give a person any guidance on how to take a high quality photo, 50% of the photos won't be readable to a dermatologist, because they'll be focused on a table in the background, or, you know, people think they can take it closer to the skin than they can. So it'll be just too blurry, that sort of thing.
Max: So when you say a picture of like a document. I didn't realize that could happen with skin.
Susan: Oh, again, because like, you know, if you think about it, even if you're taking it close up to your skin, it may focus on the hairs and not the actual skin condition underneath. So that's one stage that we use machine learning in order to provide feedback of like, hey, take a you know, take a better photo, and we'd like to get it to the point where it's like scanning a cheque, and uploading that to your bank.
Also routing — making sure if cases can be easily managed by a nurse practitioner, then we route those more straightforward cases to them, versus if it's a more complex case, you make sure you want to have a dermatologist, assessing that case.
All of our cases are reviewed, of course by a dermatologist, but the original person evaluating it a few different other places as well that we've used machine learning in order to make sure we're delivering high quality care are sort of-
Max: So are all the images analyzed by your image classification, by your machine learning algorithms.
Susan: Yes. And that's how we know, oh, it's likely to be an easier case of acne, or it's could be hidradenitis suppurativa, which is a much more complex disease, that sort of thing.
Max: Why wouldn't I mean, it almost seems like there should be a web site where you can upload the image, and it just tells you what it is, or is that like, too dangerous to have for just average web users?
Susan: Well, that I mean, that's the origin story, right is like, a few years ago, I just thought, hey, what if I could take a photo of them all and understand what it was.
Max: Right? And just had the software myself or I had some web, you know, thing?
Susan: Yeah. But that runs into some regulatory challenges and business model challenges. And so we just, we realized that making sure every case was reviewed by an expert, but making sure our machine learning is streamlining that process. So it's a great experience for patients and providers is sort of what we've ultimately ended up making.
Max: Cool. So I want to talk more about the product, but how are your engineers like finding training data for all of this? And then also, like, what software they're using? So let's start with the training data, like where, how? Are you sitting there like labeling skin, and you know?
Susan: That's the key. That’s the big secret to our company. So we've, in the pandemic, amassed more than a million photos of skin disease cases from more than 200 dermatologists in more than 20 different countries, including South Africa, India, Tunisia, Spain, in order to create a representative dataset.
Max: So you just call them up and ask them?
Susan: Yeah, we figured out a way to connect to these dermatologists, get them excited about what we're doing and figure out partnerships. But it's certainly like, also a trickier type of data, because photos can be so highly variable distance away quality. And we've done a ton in order to clean up that data and make sure it's machine learning usable.
Max: Also the labels! I assume that dermatologists are labeling it, but is it like, Okay, I know this was the condition or is it like, yeah, I'm gonna put a label on it because I think that's what they have? Do you have like a quality issue there?
Susan: Yeah, so there's quality across many different spectrums: quality of photo, quality of diagnosis. And so we do have sort of a standardized method we've built where we know different levels of quality.
Three dermatologists all agreeing is better than one dermatologist, then having biopsy information confirming it was psoriasis or whatever, is even better than three derm confirmations. So we do have a spectrum of different quality axes. In dermatology, dermatologists are 75% top three accurate, which you may say like, what that's not 95?
Max: 75% top three accurate. Explain that to me.
Susan: Sure. So it's just a computer vision, it's more common in the computer vision world but basically like the first three guesses a dermatologist has on a case? They guessed the correct disease in that three guesses 75% of the time.
Max: Okay. Okay. So and then the correct given is when everybody agrees and, you know, like, it's almost like, does the correct answer, is the correct answer actually correct.
Susan: I know. I mean.
Max: I imagine your engineers have to do a lot of statistical thinking, which is always, you know, which is, which is what we do here, it’s great.
Susan: Yeah. I mean, ultimately, the measure is, did the patient actually get better and respond to treatment? That's dream data that we're building over time. But we don't, you know, we don't have the luxury of that right now. And so we make an informed guess and then also have clinicians, like verifying the right strategy.
Max: So how are your engineers training this classifier? Are they using a particular like, you know, sort of open source software? Are they building something in-house, convolutional neural net?
Susan: Yeah. Pytorch, we're using pytorch. And then we use the Ray Tune library to select specific hyperparameters. I did ask my CTO that question earlier before this meeting. But we do try to just like, the cutting edge is always changing in computer vision and machine learning. I think our sort of secret sauce has been to amass that unique data set that no one else has, and keep supporting that and cleaning it. But that using the best in class tools that are available to do building.
Max: Yeah, it sounds like the trick here is to connect the technology available with the product. And that's I guess that's what entrepreneurs do. But that's, it's always really interesting to see how the whole thing kind of comes together. So do you have any stories about?
Maybe I don't know if you could give me your user numbers. But do you have any stories about like, this product working? Like does it encourage people to get their skin checked more often, has caught some conditions early in some patients? You know, like the one maybe you had or somebody else like, started paying off?
Susan: Yeah, absolutely. So we've just launched in New Hampshire in late December, and Connecticut, last month, and then we'll be launching within the next few weeks in Massachusetts. But we're only staying in New England; those states for now, prove our outcomes data, basically.
To the question you're asking: we have had quite a few people use our product and then get better. And that's been really great to see. And then we also do have cases where it's like a mole underneath a nail, where a dermatologist says, Hey, this really does need to be seen in person. In which case, the wait times to see a dermatologist can be three to six months or more, especially now post-pandemic. And so we're able to get them in with our partners within a week or two. And so making sure that if people do have a concerning lesion that needs to be seen in person, that they get that care that they need quickly, because they've been already assessed by an expert.
Max: Yeah, that's good. And I don't know if you have like individual testimonials, or maybe that's too hard to get, but you don't have to, like, tell me one. But you get them yourself or I mean, that must be very rewarding.
Susan: So, you know, often, like any startup, your first users may be friends or family members. And so we've had some more of those, because we're so early in our growth. But it has been really great to see how people felt.
It was really convenient that things are working, that they got to talk to a provider and, you know, ask all the questions that they had to know if it how the treatment fit into their lifestyle, that sort of thing. I don't have any specifically off the top of my head, though.
Max: Yeah, no. So that's great. I was actually wondering if you can go back like you said, you started off. You guys are based in Massachusetts, but you started off launching in New Hampshire and Connecticut? Why?
Susan: Yeah so a few different factors went into that. The first is need. Just wait times to see a dermatologist can be our you know, in our higher in these regions. The wild thing is, the number of dermatologists per 100,000 people doesn't correlate with wait time, which is like, the more dermatologists you have doesn't mean you're going to wait less, which is a shame.
But then, another factor was the regulatory environment. You know, is remote care, friendly to you know, getting paid by health insurance or are there you know, not protections against that, which you know, is more of a thing discussed in the pandemic and a few other factors like that we are close by so we can be able to support people on the ground. And then the fact that they're small as well is that we only need that many so many partners distributed throughout the state to make sure everyone's getting in person care that they need.
Max: Okay, great. Yeah. regulatory environment in New Hampshire, everything's legal. So, so and Connecticut, that's great. I, you know, as you know, I grew up in Connecticut. And so did my, my co host, Aaron here. I assume you guys are looking at expanding further after Massachusetts?
Susan: Yeah. We first want to make sure we can deliver care at scale, right, with a sort of services next, with AI and tech companies, we just need to get more iterations of that. But once we show that and we show those improved outcomes, reduced costs, improved access to care data, planning to expand, you know, across all 50 states.
Max: Yeah. So you talk a lot about the like, following up with an in-person doctor, what where do you think this tech is going? You know, maybe in the short term, and then the long term? What do you think is the role of the in person doctor once once this is kind of mainstream?
Susan: Yeah, absolutely. I mean, you know, one thing that we've seen that's made access to dermatology, even worse is the rise of Botox and cosmetic procedures, like fillers as well. Because you know, those are products that they deliver and get paid same day out of pocket, high margins. And so it can be even harder to see a dermatologist for a skin check, or concerning rash. Good luck talking to one, you know, and so.
Max: That’s so important, even though like most of the time, you're fine, it's like you want to be, you want to have peace of mind.
Susan: Yeah, exactly. And so we've just found that people have a really hard time accessing that. And so how we think about it as we can deliver end to end care, call in a prescription to their pharmacy, and, you know, monitor them if they need monitoring over time for things like psoriasis, and eczema.
But basically, like, we can serve as a form of triage, and then make sure if people need in person care, they're getting it. Things like suspicious mole biopsies, full skin checks, allergy tests, there are certain things that are sort of out of our scope at the moment. But we just want to make sure if people need care that they can access expert care, you know.
Max: Awesome. I have thought of a very provocative question. And I know, you seem like you're someone who's very careful when you're answering. But you know, you're not going to over promise or something. But like, I want to see maybe what you know, what you think of the provocative question that I thought up, which is like, do you think that this technology could kind of just beat in-person doctor diagnosis, you know, by a large margin one day and make them kind of obsolete?
Susan: So our AI today is on par with the accuracy level of an expert dermatologist, especially with outcomes data, we can, you know, with more data, get better than a dermatologist, which is, you know, a remarkable moment in time. Certainly, in radiology, we've already seen computers, and, you know, algorithms be better than radiologists at catching things, which is, you know, wild.
And so, I mean, it's a tricky question for me, because like, we work with dermatologists, we love dermatologists. They're just so in demand, we're sort of created a new model for people to connect with them. But, you know, there's never going to be a scarcity of people needing dermatologists or demand for dermatology or just trying to, you know, deliver more scale, but I know, I'm being careful on this question. It's fair for you to call me out on that.
Because I think AI can be really scary for a lot of people, including experts, you know, including experts that know everything about AI. So I think it's more important that it's designed and executed in an in an equitable and fair way to make sure all populations can get access to care.
Max: But we've also found, I think, you know, as this technology gets rolled out, the things that trick the machine are very different than the things that trick the human know, like, maybe, maybe it's something where someone has some rare thing where they have like two conditions at once or something you can kind of figure out, you know, the human might be able to figure out what's going on where the where the machine can't, so.
Susan: Yeah. It's got to have guardrails, for sure.
Max: Absolutely. At least for the time being. I think people always like to speculate, you know, one day we’ll be obsolete. I don't know, if there's anything we could say about that other than that we're not there yet. But, yeah.
Susan: I mean, one factor like, you know, there's all sorts of weird biological things in, in medicine that sort of obstruct the supercomputer being able to take over everything. And like one of those is that technically, there are 3000 different diseases in dermatology. So you can build a model that addresses the top 50 most common are critical, and get 97% of the cases, but you still have that 3%. And those 3% still need great care, you know, so right, it is just like, how do we?
Max: That's the time when you really need a doctor anyway.
Susan: It's augmentation, rather than replacement, I think, is a more helpful framing for AI in healthcare.
Max: Yeah. So I was surprised that this kind of technology has not already been deployed on a large scale. I had a very serious condition, like a year and a half ago. It was ultimately fixed by surgery, thank god, at the end of 2021. But it took them so long to figure out what was going on. And I kept asking them as a machine learning engineer, I was like, you have these ultrasounds, you have these CAT scans? Why don't you just apply a machine learning model to it? And they're like, oh, we don't have that. And I'm like, okay, so maybe I was, you know, I was in pain for like six months longer than I needed to be with a, like, a chance of getting to later stages.
Like, I don't know, it feels like, it feels like it wouldn't be that hard to- I mean, it's hard. You know, you did a lot of work. But I feel like someone should be doing this. Yeah. So why do you think it's not, you know, more widespread?
Susan: So first off, I'm sorry you had that experience. And that must have been really frustrating. And frankly, I've had that experience before as well. And it's just like, can't you can't you guys just work together? Yeah, I think. I mean, there's a lot of factors that go into why ML in healthcare isn't more mainstream.
One is business models. Real doctors get paid on a basis, you know, it's very customized per thing. But how much time they spend on something is factored into how much they get paid. And so you're not incentivized for a computer to do in a second, if your human does it in five minutes. You know, and there are some other things that, you know, impact that I think of the areas radiology definitely has the most like commercialization into the field. But even then, it's got to be some sort of tech enabled company, because it is just like a big cultural change. Right?
Max: Yeah. And it's such, to me, it's like, always interesting to contrast it with, like, online advertising, where you just, it's so easy to get a job. And I've done this to like, apply machine learning advanced and advanced statistical methods to, to advertising to getting eyeballs on screens to you know, okay, so we sold like, a few extra, a few extra Subway sandwiches and stuff, because we did all this math.
And I believe, you know, I've also done some, some work on like, some some useful things, but it's like, why don't they do like, 5% of the stuff that we're doing on ads in medicine, it's just, I it's very, it's very frustrating to me, but I don't. But yeah, like you said, it's the cultural shift. It's the market, there's an, I'm sure you're kind of more personally familiar with it than I have. So I don't know if you have anything to add, but that's me.
Susan: And I think the like hardest, the most difficult thing about, you know, having a tech background and being in healthcare is that, like, there's technology that was invented 40 years ago, that could be game changing toward a lot of areas that you know?
Max: Really, 40 years?
Susan: Yeah, easily because there's, I mean, we just adopted electronic records like 20 years ago, and that was based on a heavy mandate and incentives for payment. So I you know, it's a weird area, but then I think that it can be more difficult to make faster progress because of regulation limitations, and then just like, you know, conservatism, basically like, every, you know, every technology you want to check ABC XYZ, you know, and make sure everything's equitable, even if the current system isn't equitable, you know, the new technology has to be equitable.
And so, I wish I had answers for you. And we're just, you know, learning as we go and trying to, you know, make a difference for patients and providers.
Max: Yeah. So I mean, before we get to the end where I want to, like, tell people listening, how they can actually use this thing. But I'm actually the one thing we haven't covered yet. It's like, how did you get started? Because you said you were at MIT?
Susan: Yeah.
Max: What were you doing at MIT? And like, Were you looking for a product where there was a need in the marketplace? And so like, did you kind of have this idea when you were at MIT? What, how did that whole thing go down?
Susan: Yeah. So I was in a, like, entrepreneurship class in IP, which is the January session where you can basically do kind of like learn whatever you want, that's very different than what you're normally doing. And it was, it's called Nuts and Bolts of Founding New ventures.
But I just thought, like, hey, I've had melanoma and there we have this technology today, they can recognize it, why can't I just recognize it like my mole. So I don't have to go through the pomp and circumstance of trying to find a doctor who will take me and get an appointment, wait, et cetera, et cetera. And that was like, the original idea. It certainly morphed since then, in order to, I don't know, you know, address a lot of different things.
I, you know, the first time I pitched it, I actually cried. And because it's just like a very personal thing. And now I've talked about it enough that I don't anymore. But I think it resonated with enough people who have had someone who had skin cancer in their life or, you know, had someone who had a medical issue, and they were just like, you know, the story resonated. And so I, you know, kept pursuing it.
Max: What was the program at MIT that you were in? Did you get into the program to do this? Or did you join the program first, and then figure it out while you were in that class?
Susan: Joined the program and figured it out while I was in that class. Yeah. And the program I was at at MIT is system design and management. Okay, so it's kind of like a, you know, product design mixed with complex systems. Lots of stakeholders scale.
Max: Awesome. Yeah. So all right. So you're at MIT, and you've been pitching this product? Very emotional, lots of ideas. When did you start gathering the data? Like, how did you see like, when did you decide, okay, I'm really gonna get started on this?
Susan: So I spent my thesis on this talk to PCPs. And dermatologists and patients to understand the experience, built a model, realized all the different ways things can drop in healthcare. And I guess the more I learned, the more I was like, Okay, this could be improved as an experience for everyone. And I don't know, you know, just decided one day like, Hey, this is both feet fully in, let's figure it out. And then have been fortunate enough to have, like, a lot of support along the way through, you know, organizations at MIT and, and people in order to get to where we are. Awesome.
Max: What was the biggest challenge in getting this thing off the ground for you? And for your team?
Susan: Yeah. I'd say like, figuring out product market fit, which is quaint, right.
Max: I have like a million side projects that have no product market fit. I understand. Like, I can't stop working on them. So I don't know what to do. Well, I mean, this product, this podcast is actually a product where I want to find more product market fit. Maybe you could help me figure that out. But sorry, go ahead. Yeah. Tell me how did you guys figure that out?
Susan: I'm talking to a ton of people, be it potential customers, talking to experts in their area, understanding deeply how things are done today. And the variation on how things are done today in different areas and regions, the way it’s different for rural versus urban. All sorts of, Boston's unique medical environment versus other places. And figuring out how that could match.
Max: One of the best places in the world to do medical.
Susan: Absolutely. And yeah, just figuring out how that matched with like, what we can solve and how we can solve it.
Max: Wow, yeah, that's a that's that's a long road congrats on taking that road. No, it's, I know, it's really hard. It's probably very, it could get it could even get demoralizing at times, particularly if you're like invested in the project to have to hear, I'm sure you've had to hear bad news or people not liking the idea or you know, And then people kind of, you know, not getting back to you or not being very interested. How do you deal with that? Just as an entrepreneur and a person with a project, because I feel like a lot of people deal with that issue a lot.
Susan: I mean, I think it's, it's still hard. I still take nos personally, which isn't great. And I'm working on it. But I don't know. I mean, it's, it's sort of like with the infirm. Okay. That person said, No, here's the reason. With the information we have today. Do we think we should still keep going? Yes. Okay, great.
So like, every opportunity could be a learning experience. Some you just need to throw out the window. But, but basically, like, is this important enough to keep going? Like, because like the downside? You know, I think also people don't realize the downsides. The downside of pursuing a theme versus compared to the upside? Like, if you really take those evenly and equally. The upside is always worth pursuing.
Max: Yeah, for sure. I find it's always helpful to like, you know, have some other people in your life, whether it's your co-workers or something, to talk it over with, like a friendly group, and to try to disassociate the just the facts, conversation. Yeah. But I'm saying like, that's my idea. I don't necessarily always succeed in that.
So let's get to like, this is what a lot of people are trying to thinking right now. They're like, I gotta get something on my skin checked out. I gotta use this. Oh, what would you tell them? Where should they go?
Susan: Yeah. So I think it's more straightforward if you're on your phone, but just opening your browser, going to pictionhealth.com, like Pictionary, but pictionhealth.com. And clicking on get Get Care Now. And then just getting started.
Max: All right, awesome. Susan, thank you so much for coming on the show. Do you have any last thoughts about our conversation today? And you already said where people can go and use the product, but where can they go to find out more? Where could they get in touch with you? You never know who's listening.
Susan: Yeah, sure, absolutely. So an easy way to get in touch with us is just emailing support@pictionhealth.com. We also have thing, a phone number listed on the website that you can call and talk to my co-founder and my CTO, Pranav. And basically, yeah, I mean, I just really appreciate the opportunity to talk about what we're doing.
I think what we're doing is definitely on the nerdy side, but like nerdy in order to make sure care, you know, the dermatologist can manage 10 to 20 times the number of patients they can now and so that's just what we're trying to do. But if you're a patient in New Hampshire and Connecticut, or soon, Massachusetts, we'd love to help you out.
Max: Awesome. Yeah, nerdy side. I love that, my audience loves that. Susan, thank you so much for coming on the program today.
Susan: Oh my gosh, thank you so much. Thanks.
Max: All right. So big congrats to pitch unhealth on their latest launch in Massachusetts. I know that Susan wanted to make a pun or a story based around the idea of the local maximum. That's something that I highly encourage here on the program.
So she sent me the following quote, which she gave to a, I believe someone in the media, when asked if she had advice, had advice for women in data science, she wrote, “Don't get caught in a Local Maximum. It's easy to focus on how to optimize your current life. But important to realize your life may get a little worse when applying for jobs or vetting a startup idea on top of everything else, but that's what opens up the door to get to an even higher local maximum at your dream job or at your new venture. Don't lose sight that you could be at a local maximum, because there's likely more out there for you. You just need to do the work to get to the next part of the curve.”
Well, as you can imagine, I like that. I think that's advice we could all use. And I love a Local Maximum reference. That's what the show is all about. All right. Have a great week, everyone.
That's the show. To support a Local Maximum. Sign up for exclusive content and the online community at maximum.locals.com. A Local Maximum is available wherever podcasts are found. If you want to keep up, remember to subscribe on your podcast app. Also, check out the website with show notes and additional materials at localmaxradio.com. If you want to contact me, the host, send an email to localmaxradio@gmail.com. Have a great week.
