AI in Radiology: How SimonMed's Breakthrough Is Revolutionizing Imaging

[00:00:00] Welcome to the Lessons from the Leap podcast, where we showcase visionary entrepreneurs and leaders sharing their incredible journeys and inspiring stories. Let’s dive into the show

[00:00:14] Ghazenfer Mansoor: Hello, and welcome to Lessons from the Leap. I’m your host, Ghazenfer Mansoor. On this show, I get to sit down with entrepreneurs, founders, business leaders to talk about bold decisions, pivotal moments, and innovative ideas that shape their journey. This episode is brought to you by Technology Rivers.

[00:00:30] At Technology Rivers, we bring innovation through AI and technology to solve real-world industry problems. Most of our work is on the healthcare and health tech side. If you’d like to learn more about us, head over to technologyrivers.com and tell us about your project.

[00:00:47] Today on Lessons from the Leap, we are joined by Dr. Sean Raj, CMO and Chief Innovation Officer of Simon Med Imaging. Dr. Sean, welcome to Lessons from the Leap. Can you just tell our audience your background and anything you would like people to know about you?

[00:01:06] Sean Raj: Well, thank you for having me. It’s exciting to be here to chat with you a little bit about what we’re doing at Simon Med and what I think the impact of AI and how it’s changing how we practice and how it’ll impact patients going forward.

[00:01:26] So my name is Sean Raj. I’m the Chief Medical Officer and Chief Innovation Officer at Simon Med. Simon Med is the largest exclusively outpatient radiology practice in the United States, getting close to 200 locations coast to coast, and it’s a practice founded on the beliefs of affordable, accessible, high-quality care.

[00:01:54] And I have the pleasure of leading the medical services organization and also thinking about technology and how we bring technology in at scale. And I would say that, you know, I’m a radiologist by training, a breast radiologist specifically, but over the years after my training, I came to the realization that we’re really, really good at diagnosing disease, but oftentimes that’s way downstream.

[00:02:30] And the question I always had was: is there something we can do that’s better? From prevention, and innovation, can we

[00:02:38] combine these ideas and come up with a better way to practice medicine? And so I’ll just summarize to say, at Simon Med, I’ve had this opportunity to kind of think about or maybe rethink about imaging, not just as a diagnostic tool, but also as a platform for proactive health, and that cannot be done, at least in a scalable way, without AI.

[00:03:08] And so happy to chat with you a little bit about that.

[00:03:11] Ghazenfer Mansoor: Absolutely. Thank you. So as you launch this AI-powered business, I would say, how do you convince patients and physicians to trust the algorithms with the diagnosis that isn’t even when they came in for?

[00:03:31] Sean Raj: Yeah. So, yes, you touched on it, right?

[00:03:33] So we have modified the way we’re thinking about imaging, right? We know that AI is not about replacing radiologists. It’s about embedding seamlessly into our workflows and then figuring out what can AI do to enhance the performance of our radiologists. So in many ways, it’s like saying that AI is

[00:04:04] and it’s probably been said many times before, but really the way we’re implementing it is as a force multiplier. So it’s like having a second set of expert eyes on every single case. So the radiologist uses their years and years of training and their sub-specialization and their fellowship and their decade plus of education, combines that with AI, which has been trained oftentimes on millions of cases to help raise the floor and sharpen the ceiling, right?

[00:04:38] So what that means is flagging what matters the most, and then helping our radiologists perform at the highest level for a sustained period of time. So reducing fatigue improves confidence. And so the combination of radiologist plus AI really is actually better than either silo alone.

[00:05:03] Ghazenfer Mansoor: So how is AI changing the radiology today compared to before?

[00:05:08] Sean Raj: Yeah. So I think the best way to answer that or think about that is from the lens of a patient. So AI in general, when we say this, the default is thinking about clinical AI. How does AI enhance confidence, improve precision, accuracy of the reads?

[00:05:30] But if we take a step back, we can think, well, can AI impact even the very beginning of the funnel, the patient journey, right? From the time an order comes in, can we engage the patient in a very precise way, so that way they feel empowered by coming to, for example, our center, Simon Med. Can we educate them on what to expect and what additional AIs are available, so that way, you know, they can opt into services?

[00:06:06] For example, last year we launched a very successful program called Mammogram Plus Heart, and what this program does, it looks for breast cancer, but AI also looks for signs of changes in the arteries of women’s breasts. So now that insight can help inform risk for developing heart disease, right?

[00:06:37] So what we’re trying to do here is help patients get more educated about their breast health and heart health using AI all at the same time in one exam, a simple mammogram, right? And that is the clinical piece, but then think about the front-end piece, right? So now we have a patient coming in, who has an order for a mammogram.

[00:06:59] We’re now using the AI to educate them, send personalized reports, like an Instagram-like reel, to the patient’s phone ahead of the appointment, where they can go through, understand about breast density, their risk for developing breast cancer, other things that we offer, for example, like looking at risk for heart disease, other things like bone health, breast MRI, whole body scan, breast ultrasound.

[00:07:29] All of these individual exams which had to come from a provider going to the doctor every single time saying, “Hey, actually, you’re due for this, you’re due for that,” and making this into a continuous health journey for the patient. So any time a patient comes to see us, the AI educates them about all of their health, their preventative scans they can participate in, and then by the time they get to the center, further education, they can opt into all of these.

[00:07:58] And then while, again, they’re at the center, they can get their mammogram, have all these additional other tests that they may opt in for, even labs we’re bringing in. And then finally, the radiologist is able to use all these AI insights to help them read the exam and report on the exam, and then finally, after the patient leaves, within seconds of the radiologist signing a report, we use our AI to gather all of this information, compile it, and then deliver it again, like in an Instagram-like reel, to the patient within seconds of the radiologist clicking Sign Report.

[00:08:37] And it shows the patient their own pictures, what the AI actually did, shows the markings, and suggests next steps. And it’s completely changed the way patients think about their health. In this case of mammogram plus heart, we’re talking about mammogram findings, which is looking for breast cancer, and then also the risk for heart disease based on their arteries, right?

[00:09:02] And so now we can connect them to other tests if they need to, like a calcium score CT. So now you can see I’ve taken a step back and said, “How is AI impacting every step of the patient journey?” It’s really providing a complete view of a patient’s health.

[00:09:21] Ghazenfer Mansoor: Yeah. And how do you make sure the AI is accurate and unbiased? At the same time, I also would add, a lot of hallucination with AI. How are you addressing that?

[00:09:36] Sean Raj: Yeah, I think that’s a risky question, right? We wanna make sure that the quality of AI stays true to form.

[00:09:46] So, in the base level of AI, which is not the future state, which is based on foundation models, which look at hundreds of pathologies at one time. The base state is AI with specific indications, right? So we know the FDA has approved the indication if it’s for looking at breast cancer nodules, for example, masses.

[00:10:20] We know that we use an AI that’s fine-tuned for looking specifically for masses on a mammogram. And so the future state is what’s where it becomes very interesting, where your question becomes very relevant, right? Foundation models at current state are those, and this is early days, mind you, are those that look for hundreds of pathologies that are fine-tuned on the database of the specific center, and use the radiologist report of that center to help shape the way the AI thinks.

[00:11:02] Now, that will be very interesting once it becomes mainstream. Like, how do we maintain the quality of that? Because there is drift over time, and so we have to be very, very on top of, like we do peer review, we have to do some sort of AI peer review. And so I think that’s some great questions that we need to be thinking about right now.

[00:11:26] Ghazenfer Mansoor: Yeah, and you talk about AI peer review. Are you talking about the human review before you share with the patient? Or are you talking about more peer review, in terms of even reviewing, I would say getting a second opinion.

[00:11:48] Sean Raj: Yeah. So I think I’m talking about more from a QC, QA standpoint.

[00:11:55] We wanna make sure, because you know, the future state is that, and we’re starting to see this in chest X-ray, for example. There are models available today which will go through all of your chest X-rays and output a radiology report completed, right? And it’s up to the radiologist to say whether they agree with this or disagree with this.

[00:12:21] And so, if they agree, the report reading time could be fractional from what is current state, which is just using potentially AI to help the radiologist elevate performance. And that’s why I’m saying, the future of imaging is curious because these models are getting stronger, faster, more effective at doing the job of, in some aspects of what a radiologist is working on.

[00:12:57] And so, there is no current pathway today, in what is approved for that to happen, without a radiologist involved. But I think that many companies are hoping with their investment into these foundation models that they can get to a place, and if regulatory is able to, wants to allow that, there is a potential pathway, years from now.

[00:13:28] Ghazenfer Mansoor: Yeah, absolutely. No, I think you’re right. And that reminded me of one of the projects, one of our client projects that we work on, it’s called Konsuld. It’s a doctor to doctor consultation. So as different doctors are seeing some of those challenges, not specific to radiology, but that expands into other fields as well.

[00:13:46] But as they put that information, the AI gives some recommendations, but then there are other doctors that also look at it and provide the consultation on top of that.

[00:13:57] Sean Raj: Mm-hmm.

[00:13:57] Ghazenfer Mansoor: So, that’s, I would say, more than a peer review, maybe more of a group review. That would resonate. But there’s a human part that’s a key part.

[00:14:07] Sean Raj: Yeah. It’s an interesting point because, when you look at the raw numbers, only a minority of radiology practices today have adopted any sort of AI, right? And this is, to me, shocking because I’m in such an environment which is so AI forward. I mean, perhaps we’re the most advanced AI forward practice in the United States, with numerous deployments of different technology.

[00:14:34] But the question is: how do we use this technology to help as many people as possible? And I do wonder if there is a pathway to, just like you said, is there a doctor’s doctor, potential for a company, a group, to advertise those services? It would be very, very interesting.

[00:14:57] Ghazenfer Mansoor: Yeah, yeah. And how do these radiologists work with AI? Is it more like an assistant? Is it a second opinion, partner, thought partner? Like in our business, I personally use it more like a thought partner in many cases because I’m looking for something, and then AI is giving me a recommendation.

[00:15:17] But I think many times I use it more like an assistant as well because, okay, this is how I deal with my assistant. So I give those specific instructions, get the response, and give more feedback and evolve that. So yeah, how do you see it on your side?

[00:15:32] Sean Raj: We’ve seen a pretty rapid evolution over the last few years.

[00:15:36] The base case is, FDA-approved AI technology is good at identifying lesions. Well, the first even more base case is triage, right? So when you apply an AI technology, it will sort your list based on whether we think there’s a finding or not a finding, and the idea here is those that have a finding should be prioritized over those the AI thinks is negative.

[00:16:05] So triage is base level. Then the next level up is: okay, so you say there’s a finding. Can you show me the finding? And so now the AI in this group, actually looks for the lesion, or it’s called detection, right? So it’s very focused on showing you precisely, if you’re looking at something that has many slices.

[00:16:31] It goes to the exact slice and then marks the lesion for you. The next step above that is there’s risk stratification. Like how concerning is this lesion? Let’s give a score, right? Finally, I’d say the final piece of that escalation would be, well, can I help the radiologist be even more productive, right?

[00:16:53] So when the radiologist is looking at this, the case themselves, now they find a lesion, it’s circled, it’s marked, but now can you take these insights, if the radiologist agrees, and help put it into the radiologist’s report. So now the radiologist is able to quickly go through scans, find the nodules with the AI, and then have those findings put into the radiology report so they can be more efficient at saying, “I agree with all these findings,” or, “I disagree,” and so not put those findings in.

[00:17:23] And then finally, we have the reporting side AI. So everything we’ve talked about so far is about detection, right? Helping the radiologist find the findings. Now it’s about reporting. So now these findings are dropped into the radiology reporter, and the savvy AI-powered reporters assess how you would normally report this whole case, and then modify the language and the way it appears in the final report, based on the way you speak, the way you talk, the way you phrase things, and it creates a nice synthesized report with the indication, the body of the report, the contents, maybe description of the lesions, and then a final impression, which is really what clinicians use to review a radiologist report and say, “Okay, this is something I need to worry about.

[00:18:19] This is a follow-up,” et cetera.

[00:18:22] Ghazenfer Mansoor: So when you share those reports, are you getting it as the AI prediction or AI versus, oh, this is… Because then it’s Dr. Sean’s stamp on it, so it doesn’t matter how you produce it, but it’s going under your name.

[00:18:39] Sean Raj: Yeah. It’s a great question, right? So what we market and what we advertise to our referring physicians is that we are an AI-enabled or AI-powered practice, and we use AI to enhance our radiologists’ performance, right?

[00:18:57] This is not AI on autopilot. We are not there. We are using the insights to help us perform better, and the radiologist report ultimately is the radiologist report. It’s signed by the radiologist. They take the liability. And, you know, we provide the tools to help the radiologist and you can use it, you can use the technology or not use the technology, but our radiologists choose to use the technology because it enhances the performance.

[00:19:32] Ghazenfer Mansoor: Yeah. Yeah. So we talk a lot about garbage in, garbage out in our tech world. In your case obviously it is a huge set of data, because AI is relying on that data for training. So how do you make sure you have a clean, massive set of data, imaging data from all those 170 centers to make them AI ready?

[00:19:55] Sean Raj: Yeah. Well, I think that is certainly a problem, and, I’ll just tell you anecdotally it’s interesting because, you go to these conferences and now the AI part of our biggest trade show in radiology, there’s hundreds and hundreds of companies. And everyone says, “Oh yeah, we have the best technology that does CT scans,” and, “We have the best technology that does brain scans,” this, that.

[00:20:28] And so the question is: can everyone not… I mean, the statement would be that everyone cannot have the best technology. And so what we prefer to do, and unfortunately I don’t think many companies, or more practices can do that just because of limited bandwidth. But what we like to do is make a stress test of sorts.

[00:20:49] And so we make, when we go to talk about breast cancer, right? So we collect all of our really, really hard-to-detect breast cancers that we know, confirmed breast cancer. And then we’ll bring in different vendors who say, “Oh yeah, we have the best breast cancer AI product.” And then we will test our challenge set with all of these different AIs and see straight up who performed the best.

[00:21:15] And so we prefer to do in-house testing, and not take the advertised claims at face value. And, I can’t tell you how many times in different parts of the body where we have chosen a vendor because it was the best at the time, and then pivot it. We don’t get stuck with a vendor.

[00:21:37] We pivot fast. We’re always looking for the edge. We’re always looking for what is the best technology at the moment, and we will move accordingly. So it’s been an interesting journey because so many companies have come and gone in the AI space. And the final piece I’ll say is that the pace of innovation is so fast right now.

[00:21:59] Like the best company today may be in last place at the end of the year. And so you gotta be ready to keep up.

[00:22:09] Ghazenfer Mansoor: Okay. No, that’s cool, and I think now while we’re on it, so what is the 10-year vision for SimonMed that most people haven’t seen yet?

[00:22:18] Sean Raj: Yeah. Well, I think the 10-year journey, year one is in the books I’d say of that 10-year journey.

[00:22:26] We are very, very much focused on being able to deliver actionable insights to our patients at scale. So what does that mean? Well, we know that there’s all this amazing technology which we can use to gather additional insights and then share it to the patient. So that way they can be healthier, they can learn about their health, and then they can take the next best step for their health.

[00:22:58] And so last year we launched our mammogram plus, mammogram plus heart services. And we realized very early on, we realized that patients appreciate this added insight, but they still did not change their behavioral psychology. They learned, but it wasn’t moving the needle as far as action.

[00:23:24] And so very decisively, we looked for a company that we could work with to build out engaging reports, and I’ve referred to now twice an Instagram-like reel. So, we found a company called Cascade Health, and this company has created these almost like three-dimensional reports for us.

[00:23:51] So it takes the patient’s images marked up by the AI. It takes our results, makes it easy to understand, and then combines this into this reel where the patient’s able to read the report, understand the report, and learn what is the next best step so I can change my breast cancer risk, or I have a family history of this cancer.

[00:24:17] What should I do next? That final page of this reel, it helps schedule the patient for the next step, it engages the referring provider, and we’re able to see substantial increased engagement with the patient, engagement of the referring providers, and most importantly, the next step is actually completed.

[00:24:42] It does not fall on deaf ears. So we did this well in women’s health. We’re now in, we have now successfully launched this in cardiac health. As you know, one in three women will be affected by heart disease. The number is equally staggering in men. We’re focused on bone health, metabolic health, body composition, looking at the endemic that we have of fatty liver disease, so metabolism, and just expanding this type of programming to the whole body.

[00:25:16] For whatever exam a patient comes for, there should be a way that we can give additional insights to the patient. I mean, we recognize that CMS and the payers are not interested in necessarily paying for all these AIs today in their current state, right? Only like a fraction of a fraction of all these different AI products available have some sort of pathway to reimbursement today.

[00:25:40] And so we wanna give these AIs to the patient at an affordable cost, and we know that the patient is happy to pay. We have tremendous buy-in. And so when you ask what’s the future of Simon Med, we are gonna be the frontrunner, the leader in the space of being able to give patients healthy medical advice at scale coast to coast so they can be better custodians of their health.

[00:26:14] Ghazenfer Mansoor: Cool, cool.

[00:26:16] In one sentence, what is the biggest leap the US healthcare system needs to take by 2030?

[00:26:26] Sean Raj: Oh, I would like to see… Man, that’s a tough one. But look, I would like to see a lot of different things. But if I have to say one thing, I think, look, we know that imaging today provides tremendous insights, but it’s about converting insights into action, and I really, really hope that we’re able to change our attitude.

[00:26:55] Like, if we’re talking about preventative health, for example, right now the whole mentality of screening is one size fits all. But with the technology we have today, we can truly move to one size fits one. Personalization of all imaging, of all screening. We take the insights and we empower the patients with personalized recommendations so that they can change their trajectory of their lives so they can live the healthiest lives.

[00:27:30] That’s what I wanna see.

[00:27:33] Ghazenfer Mansoor: And, in your opinion, what is more important in the next five years, the predictive AI diagnostic or maybe generative AI chatbots? Where do you see AIs moving, and which one is more important?

[00:27:49] Sean Raj: Well, I would say there’s a lot of different areas, man. I think there’s so many areas for attack that are needed today, and there’s a lot of amazing companies working in the spaces to address those problems.

[00:28:02] But look, I can tell you from a macro lens, and you can make your conclusion. The number of cancers is set to double by 2040 compared to 2024. The population continues to age, and the scans, the volumes, the volumes of scans are supposed to substantially increase to keep up with an aging population.

[00:28:32] And guess what? A lot of this imaging that is set to explode is advanced imaging, CT, MRI, PET/CT, right? It’s not, it’s, you know… And so I guess my point here is we’re not doubling the radiology workforce, especially not overnight. We have this macro trend in imaging that less and less medical students want to go into radiology. The volumes, as they continue to increase, we have less and less relative workforce, and so something has got to give. And this is where I find that AI is going to have to make the biggest impact, otherwise the field’s in trouble. It’s just a numbers game, right? You can only work so hard.

[00:29:22] And radiologists have been compensated to date. As you can see, if you look at the average volumes of how many studies are read per day by radiologists, that number has increased decade over decade over decade. And so, something’s gotta give. The only solution I see, in my opinion, is the advent of AI.

[00:29:42] And so we have to figure out, as a society, how do we continue to scale AI in a safe framework where AI continues to help the radiologist perform at the highest level.

[00:29:58] Ghazenfer Mansoor: No, thanks. Thanks, that’s a good insight. So, if a founder is building a medical AI tool today, what is one intentional leap they need to take to be enterprise ready for someone like you?

[00:30:17] Sean Raj: Yeah, I think there’s a really big decision that needs to be made. First off, if you’re talking about clinical AI, I could argue easily that one of the big, big problems in radiology, if you look at national numbers, is actually probably an even bigger problem than clinical AI, so I’m gonna actually use this as the main example here.

[00:30:42] If you look at the national numbers, one out of three orders that come into practice will not convert to a scheduled exam. So what that means is the patient goes to the provider, the provider puts an order in, patients only follow through two-thirds of the time with a scheduled order. And then the number of exams from the time of scheduling to the time of actual exam performance is also a drop-off, right?

[00:31:13] And so we’re talking about a huge, huge number when you talk about the whole of the United States. You’re talking about a huge number of patients not getting the care they need. So how do we, as imaging practices or as a founder of an AI startup, how can we focus on opening up the tap, getting patients the care they need, and helping them potentially connect them to the right care at the right time?

[00:31:46] We know that if you don’t engage the patient the same day as their doctor visit, there is an exponential drop-off whether a patient’s actually gonna follow through on the recommended next steps. It’s a huge opportunity, right? Because you can talk about AIs that move the needle downstream, but your downstream is now, like, this type of variance.

[00:32:11] But think about if you go upstream and you can use technology to help improve that front-end piece. You’re now impacting millions and millions of more lives than you would with the downstream clinical AI. So it’s just like, which sandbox do you wanna play in? But there’s opportunity in every aspect, even operations, the RCM side, of course, the clinical side as we spent much of the conversation on.

[00:32:35] There’s a ton of opportunity in imaging, in medicine in general for AI to help you create the next billion-dollar company.

[00:32:45] Ghazenfer Mansoor: So while we’re on it, other than obviously on the radiology side, we talk a lot on AI. So are there any other innovative technologies you’re using, not necessarily in your business, but in general, have you seen some other good ones that you think people should be aware of? Doesn’t have to be relevant to your radiology.

[00:33:11] Sean Raj: Yeah. You know, thanks… yeah, I mean, there’s numerous companies. It’s just, it’s always great to see companies that are solving problems with out-of-the-box solutions. Like, for example, I’ll just come back to radiology, right?

[00:33:30] It’s crazy. We’re in 2026 today, right? And yet 90% of the orders that come in today to our organization are still done by fax machine. I mean, we’re like-

[00:33:47] Ghazenfer Mansoor: Wow.

[00:33:48] Sean Raj: Yeah. It’s mind-blowing, right? Technology from the ’90s or whatever, decades-old technology is still the main way that orders come in and out of the outpatient imaging world.

[00:33:59] Of course, we don’t have any hospital relationships or any joint ventures where the technology would be a little bit better, like more electronically connected. And so, someone created a company created a technology to scrape the… better than OCR technology. I mean, OCR had a gazillion problems, right?

[00:34:17] But created a technology to use the AI to ingest the fax at a very, very high level and set the patient up, create a patient folder, create a patient jacket, and then set the patient up for prior authorization. Simple concept, but substantially impactful for an outpatient practice, right?

[00:34:41] We’re talking about attacking a problem from the 1990s, but yet has millions of dollars of impact in the outpatient world. So it’s just combining different ideas. You don’t necessarily have to be in imaging, have imaging expertise to really move the needle, when you look at all the problems in healthcare.

[00:35:00] Ghazenfer Mansoor: Well, no, and that’s a good reminder about fax. And I was at a HIMSS and WiV conference, and there were a couple of booths about fax software. I don’t remember the name off the top of my head, but that was interesting.

[00:35:13] Sean Raj: Sure.

[00:35:14] Ghazenfer Mansoor: Cool. So yeah, a couple more. Lastly, from your seat, what is the missing link in today’s health tech stack that would actually make radiologists’ life 10 times easier?

[00:35:30] Sean Raj: Yeah. In imaging, we have what’s called silos, these imaging… like, these imaging silos, per se, right? So the tech stack historically of imaging has been bandaged together, especially in the outpatient space.

[00:35:48] So what does that mean? So that means most imaging practices have something called a PACS, okay? And I’m just gonna say in layman’s terms, it is the software which shows the images, and you have what’s called the RIS, otherwise known as a radiology information system. And this holds, it’s like the nucleus of the imaging practice.

[00:36:10] It holds the patient records, almost like an EMR, right? And these two pieces of technology really work together to be the workhorse of imaging practices. And so every additional piece usually flows through the RIS somehow, right? So you’re talking about your scheduling module, your RCM piece, right?

[00:36:33] And other components of practice. So, I think that over the last decade or so, there’s been a real push to kind of have a unified platform. It’s still though, still 20, 30 years later after the advent of all this technology, I’d say the majority of practices still have these silos.

[00:36:56] And you’re sitting there having to choose who do I want for my PACS, who do I want for my RIS, who do I want for my RCM vendor, and who plays best with each other, you know? So the big change, the big impact would be, a light AI technology, that kind of just as a layer that sits on top of all of this, and it really orchestrates all of this high volumes of information to make it easier for all parties involved, the radiologist to have access to the great information.

[00:37:25] And then, oftentimes put this into the report if necessary. Like, what’s the indication? ‘Cause that helps guess what? With downstream getting paid for the exam, right? The coding for the exam needs to be appropriate so that RCM and the radiology report all kind of agree with what’s happening, again, to help get paid for the exam.

[00:37:42] Like, a technology that can come sit in a complicated environment and just connect the flow of information rather than having to put down HL7 interfaces between these things and hope that nothing gets broken. So, yeah, I think there’s a tremendous potential for that because, while there’s a unification process happening at the level of RIS vendors and PACS vendors, many practices don’t have the money to spend on expensive new technology upgrades.

[00:38:16] So a light layer, using today’s tools that are available, could really change that, and attack that problem.

[00:38:27] Ghazenfer Mansoor: No, cool. Thanks, Dr. Raj. This is very insightful. Thanks for sharing all your expertise and what SimonMed is doing. Tell our audience, where can people find you if somebody wants to connect with you, brainstorm or anything.

[00:38:42] Is there a place, any LinkedIn, website, email? Whatever you want to share, we’ll add it to our podcast notes.

[00:38:50] Sean Raj: Yeah. I mean, happy to chat. It’s interesting, many of the great innovations we have today at SimonMed have been from reach-outs. I welcome founders of startups reaching out with ideas.

[00:39:07] Some of our great technology today has been because of just a random LinkedIn reach-out, so I’m on LinkedIn, just my name, Sean Raj. Or through email is fine, sraj@simonmed.com. And you can always go to our website www.simonmed.com.

[00:39:26] Ghazenfer Mansoor: Absolutely. We’ll add that. So one last question before we end up.

[00:39:29] What excites you most about the future of AI and preventive healthcare?

[00:39:37] Sean Raj: Yeah. So I think this goes right back to the major point, is that today, the current state, we use AI to help find lesions earlier when they’re more subtle, and it’s great at diagnostics. It’s a great tool to help us boost our diagnostic performance.

[00:40:04] But we want more than that, right? We want to turn insights into action at scale, right? And so really, this is a combined effort of our radiologists, our providers, the AI that powers all of this, all tools that we can use together to really impact the way our patients learn about their health and digest information, so that way they can make the best decisions for themselves going forward.

[00:40:39] Ghazenfer Mansoor: Cool. Thanks again, and thanks for joining the podcast Lessons from the Leap. Thanks everyone.

[00:40:46] Sean Raj: Thank you.

[00:40:47] Thank you for tuning in to the Lessons from the Leap podcast. Don’t forget to hit subscribe so you never miss an episode. We’ll catch you next time.