Full Transcript Podcast Introduction and Episode Overview 0:06 Welcome to the Doctors Making a Difference podcast, where we help physicians to be empowered with the tools they need to successful in medicine, 0:13 in finance, and in life. 0:15 Join us as we highlight doctors and other professionals around the world who are making a difference. 0:26 Today's episode is part of the Living with Metastatic Cancer series. 0:31 Longtime listeners know that I wear both hats as a physician and a patient. 0:36 So I'm excited to share this resource today. 0:38 I've been able to contact a person who is developing a large language model that can help with information that's cited and has a lot more credibility 0:48 than some of this stuff people find when they just Google something on the internet. 0:51 And also an excellent patient who's been at this for a lotta years and is really engaged with research. 0:56 The question I had asked at the beginning of this or had thought about when we developed this episode is what information resources could I turn to as 1:04 a physician where I could find out really relevant information for these rare diseases? 1:08 And if I'm a patient, how do I prepare for an interview or a discussion with my doctor so that I can actually be prepared for this? 1:16 I think you'll find this fascinating and I look forward to having you listen. 1:21 I'm excited today to have two guests with me. 1:24 And this is one of those episodes where we try to share insights, resources, and information as part of the Living with Metastatic Cancer series. 1:32 Those of you who are long-time listeners know that I get to wear both hats as both a physician and a patient with metastetic cancer. 1:40 Not really a fun road to walk, but it's a club that nobody wants to be a part of. 1:44 But here we are. 1:46 And so one of the resources that we've found to Be really helpful is to try to get information. 1:51 So when you go to your doctor, you have information and if you're the doctor when somebody comes to you, want them to looking for your clinical experience. 1:59 And so today we have two people who I consider experts on this, and I'm excited to talk with both Steve McBee. 2:05 You can go back to the August 13th, 2025 release when we recorded an episode with Steve and then Mike Wysoon. 2:13 I'd like to have both of you introduce yourself briefly. 2:15 Steve, would you mind introducing yourself to audience? 2:18 Yeah, Dr. 2:19 Crane, Steve MacB, I am a 20-year patient. 2:23 We're dealing with SFT, solitary fibrous tumor. 2:26 had gone through just about every experience possible searching for knowledge and answers to help me deal with metastatic disease. 2:39 And so here to share some of my experiences. 2:43 Excellent. 2:43 Yeah. 2:44 Mike, do you mind introducing yourself to our audience? 2:46 Yeah, thanks for having me on Dr. 2:48 Crane. 2:49 My name is Mike Wysoon. 2:50 my background is in technology. 2:52 I started a company up about 21 years ago as a tutoring marketplace called Wyzant. 2:58 And about four or five years, I left that we sold it and started up a academic literature search engine.
Why Patients Need Better Medical Information 3:07 and about 40,000 academics use it a month now. 3:11 And as a result, and from that, we've moved into using that as the core and matching it up with large language models to try to help patients ask questions 3:22 and better understand their own medical history and their health. 3:26 So through that I met Steve and we'd been working on some fun stuff together, which I think we'll get into later. 3:32 Oh, yeah. 3:33 No, this is exciting. 3:34 And so I can say this podcast has quite a few listeners who are physicians and then some who also patients because of that dual role that I get to experience. 3:43 But what I would say is this opportunity to highlight a resource and maybe an approach that patients can use and physicians can us as we approach it. 3:54 Let me just give a little bit of preface as a doctor. 3:57 I love it when patients come well prepared. 3:59 They come to the clinic visit and they say, I have specific questions about my disease and not because I think I know better than you and I'm going to 4:06 try to get a gotcha moment. 4:07 If you don't say this right, then I got you. 4:10 I like it. 4:10 When patients say I've got information and well researched. 4:13 And I'm here with a little bit of humility, seeking for information so that we can match clinical experience with good information. 4:21 And then the magic happens because you're taking a well-informed person and you are guiding them through the process. 4:26 At the end of the discussion, you make a tailored decision to that person. 4:30 This is what we're going to do based on what you know, where we are at. 4:33 You don't spend the entire visit just educating somebody about what the disease is. 4:37 you get into that next level where you say, what are we going do? 4:40 And as a patient, I love to go to my doctor's visits where they honor and respect the information that I have, that i've been given. 4:48 And, but together, you know, we're not, it's not one person talking down to another. 4:53 It's two people looking at an information set and trying to come up with a really good solution of what works for them. 4:59 and I'll give one more example before we launch into it. 5:01 I said it before, hit record. 5:03 Some of the search engines are the AI type resources can lead people to do things that are crazy. 5:11 You are not too long ago and it was late in the evening and I was working and this gentleman came in with a script that he'd gotten from chat. 5:18 GPT and chat GP T says I need to be here and here's the list of questions I'm supposed to ask you. 5:23 And basically if you don't answer all my questions, then I needed to go find another doctor. 5:27 Cause obviously chat, GPTs got it right and you might have it wrong. 5:30 Anyway, the person didn't need to be in the ER. 5:33 The diagnosis was incorrect and the questions were kind of irrelevant. 5:36 And so you have to kind be really careful how you use information to make sure that you actually use it so that it's fruitful and helpful and instructive 5:45 and not something that gets in way. 5:46 So I think as we highlight InsightfulMed, this resource, and we talk about how to approach that. 5:52 This becomes a really important conversation for both clinicians and patients. 5:57 And so I wanted to start with Steve first. 6:00 Steve, you have been an example of having a disease for a long time. 6:04 and showing up prepared with information so that when you go to a clinic visit, it's not just saying, tell me about what this disease is. 6:11 It's saying we both understand what it is, we've both looked at the research. 6:15 How does this apply to me? 6:16 I wanted to give you a chance to reflect on that experience as you've worked with clinic visits, both good and bad. 6:22 I have to tell you, I felt like I was walking into the meetings with my doctors very informed and having the right questions. 6:34 I realized early on or actually late after discovering InsipilMed that my list of questions might've been 10 deep. 6:44 And that was a little bit overwhelming when I would walk in the room and expect my physicians to answer those questions. 6:52 Reliable patient education relies on input that is correct.
Steve McBee and Mike Wysoon Introduce Their Work 7:00 and often my input was biased because I was looking for specific answers. 7:07 You know, walking into the room with even a sarcoma specialist, they're learning alongside you if they realize that a Sarcomas specialist Dr. 7:18 Brotto or Dr D'Amato were probably about 10 to 20 percent of their focus, meaning SFT is 10 to 20 percent. 7:31 They are seeing so many other rare diseases at rare sarcomas that walking into the room with the right questions can make all the difference. 7:42 So I think it's important to go into that with proper information. 7:46 And then the other factor that I kind of wish it was a little bit more transparent for patients. 7:51 People don't realize that when you show up, like when somebody comes in with, you know, they have a notepad with three pages of questions, 7:57 my heart starts to race a little bit because they don't know that I have 20 minutes and it's not because I don t want to pay close attention but I've got 8:06 three patients that are waiting and I ve got a baby that's about to be delivered and somebody from the ER checked and so-and-so is having an emergency 8:13 at home and wonders whether they should come in. 8:15 And so you want give all your attention, but you also want make sure that that time is fruitful, that you have 100% of my attention during those 20 You know, 8:24 don't plan on an hour unless you're, you know. 8:27 If you have know that, if you had that transparency, like I've booked an hours, I booked 15 minutes, You would know but in most patient interactions, 8:34 really, especially if it's a follow-up patient, not a new established care, a lot of times the system allots you 15, 20, maybe 30 minutes if your lucky. 8:41 And so you want to make sure that it is useful and show up with relevant questions and as a patient you don' want say like, 8:47 oh I'm so disappointed, but I didn't get to answer all my questions. 8:50 But you also want to kind of have a little bit of agenda setting at the beginning so you do get to those most important questions. 8:55 So, Mike, I know you talked a bit about your background with education and research and large AI models. 9:02 Obviously, this led you to this process to start InsightfulMed, which is Maybe just to like what you shared with me a little bit more than just algorithm 9:11 driven or something that leads you to the answer that you want. 9:13 We talked on before we hit record, we live in a world where you can sort of create your own reality. 9:18 You can make it so all the media you take in fits your worldview and you click on one thing and it leads to another thing that. 9:24 Leads to some similar information and pretty soon everything around you can give you the answer you want and Maybe that's not really the most scientific 9:32 inquiry method really we should be looking to wherever the data leads is wherever The science leads us and so tell us a little bit about the development 9:38 of insightful med And what makes it different than other AI or large data model search engines? 9:45 Yeah, I guess I'll start with the Second part first, which is what? 9:50 makes It different Than other large language models. 9:53 So large-language models, at their core, the way they're trained, is you put in a ton of different text and as much text as you can possibly get your hands on. 10:06 That means, in some cases, downloading all of the books ever written and digitized, legally or illegally. 10:14 But at its core it's just a probabilistic association between words. 10:20 There's no centralized idea of a fact. 10:23 Right. 10:23 If you think of it as like just a giant auto complete and literally this is how they work. 10:28 You put in your prompt, it puts out what it thinks should be the next token. 10:33 Then it put your prompts back in along with that next and spits out the nest. 10:38 It just keeps on doing it. 10:39 This is really complex auto. 10:41 Okay. 10:42 There's no facts in there. 10:44 The facts that it does spit out just happened to come from the fact that those words are associated with each other. 10:51 Interesting. 10:52 Yeah. 10:53 And then there's like training on top of that where that get people to get it to act like a chat bot. 10:59 Okay.
How InsightfulMed Anchors Answers in Medical Literature 11:00 Part of training and kind of going back to what you were saying about making our own reality. 11:05 A large part of the training is making sure the user is happy. 11:10 That's true. 11:11 Right. 11:12 So happiness and what is quote unquote happy, I shouldn't even say happy engage. 11:21 And so as like you, like even the way you ask a question to a large language model can affect the, way it answers. 11:29 Right. 11:29 So like, you can essentially lead it. 11:31 It'll try to pull itself out sometimes, but like. 11:34 You can, essentially, lead into the reality that you expect. 11:37 And that's where hallucinations come in. 11:39 That's, where if the person brings in the prompt is immediately adversarial with their doctor, it'll respond as an ad, in a way that like encourages that 11:50 adversariel nature. 11:51 which is not good. 11:53 And that's why, I mean, there are some things that it is good at. 11:57 So there's been a bunch of studies out around helping with diagnosis. 12:02 Okay. 12:02 Cause you got to go back to that probabilistic nature. 12:05 You've got a. 12:06 Bunch of symptoms. 12:08 Right. 12:09 And these symptoms are all probabilistically associated with an outcome. 12:14 So it's really good at giving a list of like 10 things that might be going wrong. 12:19 But as soon as you ask it, why it thinks those 10 are there, there's no underlying why. 12:24 It's just probability. 12:25 And so it starts generating facts and it's starts hallucinating. 12:28 So then that's the big problem with large language models, just as they're built. 12:32 For us, what we do is we like anchor it in truth and fact, and that truth in fact comes from medical literature. 12:42 Every factual statement we make is cited. 12:46 And, so you can see not just like the. 12:49 Paper that we pulled the statement from, but like the paragraph from the paper we pull the same from. 12:53 So as a physician, you can verify right there that what we are saying is true. 13:00 Or I shouldn't necessarily say it's true is is anchored in medical literature because the basis of science is proving the previous person wrong. 13:08 Right. 13:09 And like improving upon knowledge. 13:12 And then no matter how much like meta analysis you do on the authority of the patient, the physician, be journal or whatever, 13:21 like our tool will never catch something like the beta amyloid crisis. 13:25 Like the reproducibility crisis, right? 13:27 It was a giant. 13:29 You guys familiar with that? 13:30 So the beta amyloid for long decades, a lot of research for Alzheimer's was focused around like beta-amyloids proteins in the brain and for decades. 13:40 And it was like, everybody was focusing on that. 13:43 Then they realized that the foundational studies that it. 13:46 All of that research was built on were fabricated. 13:50 But there's this entire body of medical literature around supporting that hypothesis. 13:57 there are systemic problems with medical literature. 13:59 So like we as a society have to deal with, but anyway, so what we try to do is make sure that every statement we make, every claim we made can be tied 14:09 back to the medical. 14:11 I think it's great. 14:13 Yeah, that's awesome. 14:14 And so when, and you can share it with your physician so the physician can actually look and like validate and verify everything. 14:21 Like you are not let astray. 14:24 via like you're in the way you phrased your prompt or something like that. 14:30 And so that's the core of what we do. 14:32 We've added other stuff onto it, clinical trials, allowing you to upload your medical records or like integrating with like the FDA APIs and like affairs 14:41 database for adverse drug reactions and stuff like. 14:44 So we can bump up all of these different like authoritative sources and provide the patient or the physician with valuable medical insights that they can 14:55 come prepared to their doctor with. 14:56 But all answers are given in the context of you preparing for your doctor's appointments, right? 15:02 It's not an adversarial thing, it's just like, how can you be as informed as possible coming into your doctors appointment so that 15 minutes that you 15:11 talked about is as high value as productive as That's excellent. 15:16 You know, I think we, we talk a lot about the information age. 15:19 We live in the Information Age, but in some ways we are transitioning into kind of the next step of that large language models, 15:26 AI type stuff. 15:28 It's here to stay. 15:29 it's going to be informing the reality for a of experiences. 15:33 And so the old model of Well, the doctor has all the information, and the patient doesn't know anything, so show up and shut your mouth and just listen 15:39 to what they say. 15:40 I think we're exiting that and going to a place where people hopefully have educated folks have access to the same data set. 15:48 The doctor, or the patients, anybody the researcher, anyone along that chain. 15:53 And what you're looking for is a conversation just like you said. 15:56 It's informed, informative, it's useful, but you are all looking at that same information and as an educated informed Patient, 16:04 when I show up, I want to not just be talked down to or have that information just given to me like I'd never heard it before. 16:11 I wanna talk about what this study said or how's that relevant to. 16:15 What is your clinical experience? 16:17 Tell me that this is how does this apply to meet so we can come up with a plan at the end of that that I'm happy with and the doctors happy within. 16:24 Hopefully that's not adversarial, but cooperative as you mesh those to a clinical experience plus what these large language models teach us and what the 16:33 patients own experiences so that it's really kind of a tailored approach. 16:36 And the old model of how we were trained as physicians and patients is sort of changing a little bit with these, especially if you can get respected data 16:45 sources so it becomes informative. 16:47 Steve, what's your impression? 16:48 What's you're insight on this? 16:51 Well, Dr. 16:51 Crain, when I discovered Insightful Med and I talked to Mike about this, one of the first things that I did was reflected back on my 20 years of dealing
Preparing for Doctor Visits with Focused Questions 17:00 with this disease and each phase that went through, starting with discovery of a disease. 17:06 So I had a car accident and headaches and some waking seizures. 17:12 And so I took that information and just typed it in as a basic prompt into insightful meta basic question. 17:21 I just had a headache. 17:22 Had a car accident. 17:24 What should I do? 17:25 And so it directed me towards going out and seeing a doctor. 17:31 That went in and had to see T and the doctor told me I have a brain tumor. 17:38 I'm looking at the input or the output from InsightfulMed and the questions that, what I really like about it is that it will give you a basic overview. 17:49 It will get you some questions you need to be asking yourself. 17:54 Questions that you probably should be ask your doctor, three or four questions. 18:00 And when I am looking that I thought, I wish I knew that. 18:05 at that moment, when that happened, I wish I had this resource to plug in. 18:10 And I carried that through to brain surgeries and pancreatic surgery, liver surgery and ablations. 18:21 This is such a great tool for me, especially going forward now. 18:26 Any new patient that is introduced to a rare disease, I just hope that they can find this tool and utilize it in the same way that I have over the last 18:40 couple of months. 18:40 It's just been game changing for me and how I've seen my disease learned. 18:48 Even though I spent the past 20 years going to medical libraries and searching through PubMed and looking for answers, it's been the easiest, 18:59 most clear path for Yeah, I'll talk to Michael about this and I think that this is how the sharing the whole patient path from discovery of the disease 19:14 to going to a sarcoma center to different types of treatments. 19:20 And then Michael took that and ran with it and created the SFT, the insightful medicine patient guide, which was just a fabulous new tool. 19:37 It's awesome. 19:37 So Steve, when you go to a doctor's visit, again, some of the folks listening to this are going to say, I just got diagnosed with solitary fibrous tumor 19:45 and I searched up something in this podcast popped up. 19:47 Cause it, you know, just it. 19:49 I don't know anything about it but I know I'm listening. 19:51 To a guy that's had this for over 20 years, how do you prepare for a doctors visit? 19:55 Like what's your process? 19:56 And again you have a finite amount of time. 19:59 How do prepare that? 20:00 So when show up you can kind of strike this balance between well-informed and inquisitive. 20:04 Curious and inquisitive and how do you know what's your process? 20:08 Well, I've developed a there's a method to my madness and it seems to work for me I did all to playbook for me as a patient, 20:16 I preach this to my doctors and just on one page I tell my team what my expectations are in the meeting, what expectations for my life is, 20:30 my quality of life and what I want to do. 20:32 So let's help me get in for Go that direction, stay healthy enough to go that. 20:41 Make decisions so I can go in that way. 20:44 I tell my team how I communicate and some kind of my way of working. 20:50 And then I asked them to tell me their way working and then in addition to that, and again, this is really all on one page, 21:02 so it's not overwhelming. 21:04 I'll put down my 10 questions. 21:08 Overwhelming amount of questions, but I tell them. 21:11 If you are the expert. 21:16 Please reprioritize these questions of what you think is important for me. 21:20 I'm not going to walk in the room as an expert, But this is in general what I want to know. 21:25 There are probably some questions and this was where insightful Matt has been really helpful. 21:30 I can put in 10 questions and say, given my circumstances today, what are the three things that I need to ask my doctors? 21:39 What is, they may be, and what needs to go in the parking lot? 21:45 And I've been really impressed at how it can re-prioritize those for me so I could be focused on the right thing when I walk into that meeting with my doctor. 21:57 That's cool. 21:57 That is a nice tool. 21:58 Like I would say, every single day in my office, I have folks who come in with a large list of medicine. 22:05 I'm not bugged by it. 22:06 What we do at the beginning of those visits is we a little bit of table setting. 22:10 Okay, well here's your questions. 22:11 And just like what you described, Steve, let's focus on these three or four because I think those are most relevant, most important to what's going on 22:18 right now. 22:18 And then we'll meet again at whatever time we will talk about the next. 22:22 I'm going to give you some resources so you can answer these others. 22:25 And it sounds like, you know, with these large language models like insightful med, You might be able to have some of the just basic nuts and bolts of 22:32 factual statements answered through that. 22:35 But it is really helpful to go. 22:37 If it was a one pager and the patients come in and say, look, I am ready to answers to these questions, but help me prioritize what I really need to know 22:46 based on your experience. 22:47 I actually like that quite a bit. 22:49 And then because it sets the idea of what is a priority for everybody. 22:54 Mike, it sounds like you've used Steve as far as kind of your guinea pig on some of this stuff for solitary fibrous tumor. 22:59 But I wondered what you know, as you develop this, you mentioned being able to upload in a private HIPAA protected way to your own information,
Using Medical Records and EHR Data with AI 23:07 your medical records as a recent development. 23:09 And I wondered about that step and then how do you envision this going forward? 23:14 Because it seems like this has a lot of potential to be really helpful for everybody in the medical decision making. 23:20 Yeah. 23:22 So let me start. 23:23 We are, for physicians out there, we respect all of the security practices of HEPA, but we are not HEEPAA certified right now. 23:33 That is not something we've gone through, but like we have all of the logging, all the verification and everything, But we don't assign BAAs or anything 23:42 right now. 23:43 We're, we're patient focused. 23:44 Just, I wanted to kind of clear that up. 23:47 But yeah, what we actually just released yesterday is the ability for you to export your information from an EHR and import it into our system. 23:59 then our agent can, as you ask questions, selectively go fetch relevant records from your past history. 24:06 So if you asked something about like, make sure Steve's 20 year history of SFT is going to come up with a lot of stuff, but it will go selectly fetch the 24:17 pieces of your record that are most relevant to answer your question. 24:21 Michael, what gave us an example? 24:23 So that is what type of records are those? 24:26 That's pathology, histology. 24:28 Yeah, so it's all text based stuff. 24:32 So like actual, like primary imaging, we don't have like foundational models to read images, but the interpretations we pull in. 24:40 It's. 24:41 All your vital records, all your notes from your doctor, All of your tests, be it blood work, he scans all like anything that is available to you. 24:52 And let's say like my chart. 24:54 When you export it is. 24:55 Available to us. 24:57 Right now we just have a one-off import and you can, you do it as much as you want. 25:03 We will in the near future allow you to essentially like do an OAuth or like a single sign-on so that we could pull your records for you. 25:13 Yeah. 25:13 But the idea there is like in. 25:16 when we're talking about like getting prepared for doctor's appointments, we can help you prepare in the context of like your historical appointments with 25:23 that doctor or the labs that that Doctor has ordered previously that have just come back that you haven't seen them since. 25:31 And yeah, the idea of being essentially patients don't necessarily always know what is the most relevant thing to give to BLM. 25:43 like social determinants of health, history of smoking, drinking, weights, that kind of stuff as well, which all oftentimes like matter, 25:50 but they are not necessarily input. 25:52 So it becomes a lot more powerful. 25:54 I could see the value in that and that as a patient, you're in the room with a doctor, it's sometimes a little bit overwhelming and you don't absorb or 26:06 hear everything that the doctor says. 26:08 And then you get the report and no offense, Dr. 26:12 Crane, but sometimes physicians write in a language that we as patients don't necessarily understand and insightful. 26:22 Then once you've uploaded that can interpret that and give you the record in the context that you understand as a patient. 26:33 And I've tested that a couple of times, and it's been a lot easier for me to look at that report from my physician or from a treatment that I received 26:45 or the scans or pathology. 26:47 And it kind of drives it down into some more basic language that understand. 26:54 As I like to say, it is on a 10th grade level. 26:57 I can read that. 26:58 That's great. 27:00 I'm sure you can read much higher than 10th grade level students, but I just thinking about this. 27:04 So when I went to medical school, I graduated in 2008 and it was just in that transition where things were becoming electronic health records for everything. 27:14 We wrote everything on paper at the beginning. 27:17 And then it became EHR based, which has its pros and cons. 27:20 But it's interesting how that's appreciated your comment. 27:24 And I think it is relevant for this discussion about insightful med. 27:27 Doctors have traditionally been trained that the notes were primarily to communicate to other physicians or to give a reminder to yourself of what your 27:36 treatment plan was for that patient. 27:38 Prior to the EHRS, it was never really something that was patient directed. 27:43 The notes where primarily for Other clinicians like, hey, cardiologist or other specialist, here's what's going on. 27:49 Here's the test I've ordered. 27:51 And it served as a medical record for record keeping for the patient's best interest, but not necessarily a patient facing sort of thing. 28:00 Now with the HRs, it's become very much insurances want to have all that information. 28:04 They want everything said in such a way. 28:06 and the patients have access to my chart. 28:08 And there's a lot of positive to it. 28:10 Folks can get information to their access, to the records. 28:14 One of the challenge I find as a doctor is that you're recording stuff and you doing it in real time. 28:19 You're also seeing 20 or 30 people that day. 28:22 Do you have to go home and see your family that night? 28:25 And how do you get that all in a way that's properly represented is so that it factually represents everything and now I worry about like what if I say 28:33 something if i make some comment about some word that's going to offend that patient you know because they're going look at it and you want to make sure 28:39 that its information now is kind of high stakes on those records because everybody looks at and so what I think you just described on insightful med Again, 28:48 you want to make sure that things are privacy protected and stuff like you said, but it is interesting to be able to say, 28:53 how do I take this data down to a level where I understand it and can read it? 28:58 Cause like, medicine's its own language. 29:00 I had to learn a new language when you go to medical school. 29:02 And so how did you take it from medical ease into your native language? 29:05 Yeah, totally. 29:06 That's part of like the prompting around RL is kind of bringing the technical nature of the language down. 29:14 And like when there's like pivotal concepts around like your disease or whatever might be going on or a test or something like that, 29:23 it tries to like. 29:25 use analogies and use other things to like help the patient understand what's actually happening so they have like a better understanding than just this 29:33 like really complicated word that like they've never heard before yeah so it's interesting i mean people have used this medical records feature for a few 29:41 different things uh we actually had someone recently report that are you guys familiar with liquid biopsies or like ctdna yes yeah basically for people 29:52 listening essentially There's a new procedure coming through, coming down the pipe. 29:58 It's still being actively researched, but the idea is in the case of like a tumor that has gone away, they will sequence the original tumor and then draw
Personalized Care, Rare Disease, and Better Decision-Making 30:09 blood and sequence, the DNA in your blood to see if there is evidence of that tumor's DNA circulating around. 30:15 it's very much still experimental. 30:17 I mean, there's the lot of the good evidence and it depends on like the type of tumor you have to have a cancer. 30:22 Well, like on how well it works and it's just not generally available, but we had a patient come in and upload their medical records and like our tool 30:34 started talking about liquid biopsies as a thing coming down the pipe. 30:38 And their hospital said, no, we're not going to do it. 30:40 Their insurance company said no we are not gonna do. 30:43 And they had Insightful Med write a letter to the doctor outlining all of the research about their particular tumor, how it's been well researched and 30:53 how effective it is. 30:55 So the, the doctors said okay we'll do, it and then they wrote a, letter using Insighful because you can have a draft letter for you using, 31:02 Insightsful to be insurance. 31:05 Essentially like out lining all the benefits. 31:08 And they both got approved. 31:10 But it's just, again, the power of integrating your records with our cool. 31:15 I got really excited when I heard that one. 31:17 Well, we live in a new time, like I said. 31:19 And I've had to shift, say, to 2008, 2007 model of writing notes on paper to now it is an outward patient-focused sort of record that everybody, 31:29 from the insurance to the patient, everybody is going to be seeing that and having ready access to it. 31:34 So now where we say we're entering this large language model, time where AI is going to be here and it's important to know how to use it properly because 31:42 there's lots of improper uses where, like you said, it just gives you another prompt and an autocorrect and people are led to things that are foolish and crazy. 31:49 But we need to have this really based on good published data and then as clinicians and researchers, you need to go wherever the data leads you. 31:57 But if you have access to all the date at once, instead of having to spend five hours in a library combing through it or in PubMed, 32:04 it gives you tremendous power. 32:05 And so you'll have that at your fingertips. 32:07 The hope that I have, and I know Steve has, as well, I'm sure Mike, is that with this power, that you lead to solutions that are not just theoretical or 32:16 this worked for somebody or it might work, but to actually say, hey, this is the Peter Crane solution. 32:21 This might actually work for your tumor, your subtype. 32:24 And then you go from in the world of cancer, instead of saying like, you have this cancer that you kind of get to the place where everybody has a rare 32:31 cancer because it's the Peter Crane cancer. 32:34 You know what I mean? 32:35 Everybody has the specific cancer we know, what your genetic mutation is. 32:38 We know the treatment is that works for your cancer and then instead saying, oh, You have X, Y and Z cancer you really focus it down and say, 32:46 it is really specific to you and the treatments can become much more tailored. 32:50 So I'm pretty excited about where this can go. 32:52 it's just a matter of trying to do it right no totally i mean as like a personal story so the medicine inherently has been kind of like especially guidelines 33:03 have been built around like at a population level not an individual level and a family member went in got an x-ray found something weird on the lung There's 33:13 unrelated incidental findings, but they did a CT scan, found a nodule with some features that was like 17 centimeters, I believe. 33:21 And they're like, OK, we're going to wait six months. 33:24 Let's do follow up imaging. 33:25 The guidelines say we should do that. 33:27 I hope that I was millimeters, not centimeters. 33:30 Seventeen centimeters is quite large. 33:33 That would pick up most of the airway. 33:36 Anyways, there's some feature on it that were clearly pointed to like doing something sooner. 33:42 And like, or through medical literature, we decided, oh, well, this is probably not something you'll sleep six months on. 33:48 You should go against the guidelines in this scenario. 33:50 So she got a second opinion with someone who knew what was going on and they did a biopsy, lobectomy and started her on like a specific treatment. 34:01 She's still here and healthy with actually no signs of the tumor, which is great, right? 34:06 But again, like that's where like the population level recommendations fall down and not all doctors have the ability to spend a ton of time doing deep 34:16 research on every patient. 34:17 It's just that It's not the doctor's fault, it's just how the American medical system is structured. 34:25 We're going to get to a system where it is more personalized, and I think AI can be a big part of that. 34:30 As it pertains to solitary fibrous tumor, this discussion gives me a lot of hope, personally, people in our community, but just across the whole spectrum 34:40 of what we consider rare diseases. 34:42 And there are things that are more common, But I think that your particular iteration of it, each individual does have specific things and we have the 34:50 ability to tailor personalized medications and treatment plans a lot more with this kind of technology. 34:56 As we kind get a little bit closer to wrapping up our time here, I wanted to give each of you a chance to kind some final thoughts. 35:02 And maybe if you don't mind, share with our audience how they can communicate with you or reach out to you, or follow your work if have a website or a 35:09 LinkedIn profile or something like that. 35:11 Steve, do you mind starting? 35:13 Yeah, you can reach up to me on LinkedIn. 35:17 I have Soldier Virus Tumor-focused newsletter on Substack. 35:24 Which is really good, by the way. 35:26 Thank you. 35:27 It's an attempt for me to share like all of my institutional knowledge and experiences with other SFT patients. 35:38 And so that's kind of where I hang out. 35:44 Thank you, Steve. 35:44 Any final words, any thoughts that you would leave with our audience as you kind of finalize your time here? 35:49 You know, I think that using a tool like InsightfulMed has been very helpful for me, and that comes from having 20 years' experience of researching and 35:59 thinking that I was on the cutting edge of knowing about our disease.
Final Thoughts and Where to Find the Guests 36:04 This tool has game-changing for me. 36:08 And it's so easy to sign up and use. 36:10 I don't have any vested interest in this, except that I think for our community, for all the questions that we see that are repeated over and over in the confusion, 36:22 especially in Facebook groups, which is really the only place that have. 36:28 Our time with our physicians is very limited. 36:31 Physicians don t come into our This is a tool that can be very helpful and guiding you to better understand what you're facing. 36:45 That's excellent. 36:46 Thank you very much. 36:47 Mike, what are your thoughts as we conclude here at Contact? 36:51 Generally, I'm somewhat apprehensive. 36:54 This is gonna sound weird for someone who's doing what they're doing about where AI is going to take us as a society. 37:01 There are really amazing uses of it, and there are terrible uses. 37:07 And having people understand what a good use is really important. 37:11 That's kind of what we're trying to do here. 37:14 How do we, in this case, rectify like an incredible information asymmetry between like doctors and patients and there's tools like ours out there like 37:25 open evidence that I know a ton of doctors use but patients can't use it and they purposely keep patients away by requiring an NPI number to sign up And 37:37 like all of their funding comes from like pharmaceutical companies, conflict of interest. 37:43 And so we're just trying to build something to help people and patients to the extent that like we can help. 37:50 That's our sole goal here. 37:52 Yeah. 37:53 So I mean, for people who are interested, especially in this community, Steve alluded to it. 37:58 It's the solitary fibrous tumor patient guide. 38:01 You can find it under insightfulmed.com, INCITEFULMED.COM, slash, resources. 38:11 And it'll pop up there. 38:12 Uh, our goal with that is to essentially create a list of resources, especially starting with the rare stuff that can help patients kind of anchor themselves 38:23 and understand and get up to speed. 38:25 And there's even things, there was a feature on there where you can type in your address or your zip code and it will pull up the researchers who are closest 38:34 to you who're studying the thing that you're suffering from. 38:38 part of like what we want to do is empower patients. 38:40 And that's like making sure your meetings with your doctor are good, but also your meeting with the right person, which is especially important in a rare disease. 38:49 Next. 38:50 Cause your average PCP is not going to understand anything about SFD. 38:54 Right. 38:54 No, like I, I'm a primary care physician. 38:57 I don't think I'd ever even heard of solitary fibrous tumor before I became a patient of it. 39:00 You hear a lot of sarcomas, but there's so many weird, rare sarcombas and you never ever see them. 39:05 So no, really helpful and appreciate what you're doing, Mike. 39:08 Thank you for taking the time to come on and Steve, you as well. 39:11 thank you. 39:12 Taking the to do this. 39:13 And I think this episode will be really informative to those who listen. 39:16 It's been informative for me and I hope you both stay in touch. 39:23 Thanks for tuning in to the Doctors Making a Difference podcast. 39:26 And thank you for what you do to help your patients and your community. 39:29 Your work truly helps so many people. 39:31 We produce this content to have the tools you need to stay in medicine and to highlight the amazing work being done by physicians around the world. 39:38 Please note that while I am a physician and many of the guests on this program are also physicians or other professionals, 39:44 the discussions on the podcast do not represent my employer or any professional organizations to which I belong. 39:51 This podcast is for your information and entertainment only, and should not be taken as professional advice. 39:56 You should seek appropriate professional advise pertaining to your own situation. 40:01 Please check out more of our content on our website, DoctorsMakingADifference.com. 40:05 Also, please follow, like, or subscribe on your podcast player or YouTube channel, and give us a five-star rating. 40:11 It really helps to spread the message. 40:13 Finally, if you'd like to be a guest on the podcast, Or if we'd to nominate someone else to the guest, Please visit the website or email admin at Doctors 40:23 Making A Difference dot com. 40:24 See you next time.
Comments