Podcast Wave

AI in the medical field is a topic of interest for high school students Olivia and Whitney. They discuss the rise of AI in their generation and how it can be beneficial but also have drawbacks. They mention Tim Miriam's article about how AI is being used by Gen Z for studying. They discuss the history of AI, starting with Arthur Samuel's development of an AI program to play checkers. They also mention OpenAI's GPT-3, which created human-like content. They explore the potential uses of AI in healthcare, such as reducing the workload of healthcare workers, assisting with studying, and improving documentation. They also mention the concerns about job loss and cost-cutting due to AI. They reference surveys showing different opinions on the replacement of doctors by AI. They believe that AI should be used to assist doctors and improve efficiency. They mention current applications of AI in healthcare, such as assisting with diagnostic imaging. They highlight Mott MD Care as a company that al Welcome to our newest episode, Artificial Intelligence and Olivia and Whitney's Perspectives. This week, our topic is on AI in the medical field. We are high school students interested in the rise of AI. In our generation, there has been a rise of this kind of technology now more than ever before. With the rise of AI, we can be able to do a lot more things. Tim Miriam wrote an article about how he has noticed the rise of AI with our generation. He said, In it, I point out that as a boomer, I do not have AI to use when in college and learn the old-fashioned ways of vectors and hundreds of hours studying in the library in my room. On the other hand, Gen Z, who are now college-age, as well as high school students, are using AI to study for them. There are many things that AI can do for us, but we don't want it to be that AI is doing everything for us. We still would like to learn. Kind of like the contrast, he calls himself a boomer. He points out that he doesn't have a lot of experience with AI, but he still is open to it. He didn't have it, but he's open to the possibilities that it might have. For example, he's talking about, I'm not exactly sure how old he is, but he might have had to go walk to the library with his own two feet, go find an encyclopedia about the topic, go home, research about it, and then write it out. If we have a question, we can just simply Google it or something like that and then write an essay about it on a computer. He's saying that there's a big rise, and with it, are they still getting the effort that they need, or are they just preloading it? It's really cool how much technology has changed in the last little bit. This week, we decided to research how AI is being used in healthcare. As with anything, there are pros and cons, and when it comes to artificial intelligence, what will be good for the world that we live in right now, and what won't be? What should we just leave behind? The real question is, will it aid in our healthcare, or will it harm us? Would you trust medical advice given by artificial intelligence? Before we explore the disadvantages and advantages, let's explore what AI really is. Artificial intelligence is known, it can be called AI for short. Computers automate tasks of humans. There are many several different types of artificial intelligence, and they all are used in different aspects and ways. Although it may seem like a relatively new thing, it's been around for a really long time. Arthur Samuel was in the 1960s, when he started one of the first artificial intelligence programs. Let's learn a little bit more about him. Samuel wanted to invent a way where he could play checkers against a computer. He spent some time with some researchers, and so what they did is that this computer would rate these checkers moves on best to worst, and it would try to figure out the best one to play. It started off not being very good. It would lose almost every single game, but over time they programmed it so that it had more experience, and it would be able to learn for itself, and that's kind of really cool. The process, which is programming it better and better over time, he called it machine learning, and it's still used today. In 1962, his program played thousands of games to develop its skill, and he called it a checker's master. By then he was satisfied with his work. He spent a long time. But the principles that Samuel used about the checkers laid the basic groundwork for artificial intelligence today. I think it's cool that Samuel wanted to be able to play chess against a computer. Sometimes, for example, when we're bored at school, we go on the computer and we're like, let's play the snake game, or let's play against the computer. I still enjoy playing face-to-face games, but also I think it's cool that we can play against the computer, and you're like, ha-ha, computer, I win, or when the computer wins, you're like, ha-ha, you beat me. But I just think it's cool that he developed these a while ago, and he still uses these kind of things today. Yeah. More recently, in 2023, OpenAI started testing GPT-3, and this was the first AI that was able to create contact that looks almost human-like. Today, AI is growing at an unbelievable rate. This AI was huge and became a mainstream phenomenon. It turned the public eye to see what is possible, what was really possible in the world today. People began to start doing things less and asking AI to do it for them. For example, hey, AI, can you remind me in two hours to start the dishwasher? Or, hey, AI, can you put this on the calendar for me? And you can ask ChatGPT to summarize your email for you, so you don't have to read it all if it's a big, long email from your boss. Just, hey, AI, can you summarize this for me? Or you can ask Snapchat AI to write a script for you to call your neighbor if you really don't know what to say to call your neighbor, but you want to get to know them. Hey, Snapchat AI, can you write me a script on how I should call my neighbor? And those are just a few examples of what you can do rather than of what ChatGPT or Snapchat AI or AI can do for you rather than doing it yourself. Yeah, I think it's really cool if you look and see how much AI has changed over time. And it is machine learning. So the more that we interact with it, the better and better it will become, and the more it's going to learn and change. And I think that's just really cool, you know, that it can just learn from its own mistakes. And I think that's kind of crazy. There are so many pros and cons to AI. This isn't in the medical school field, but due to time constraints, we're only going to be able to discuss and dive into a couple of those. One big problem with our current healthcare system is that we overwork our healthcare workers. There's not enough of them. There's a shortage of them. And so because of this, they're getting overworked, and they're working long shifts, long days on end, and that takes it back. They get tired, as they should, and they're working hard for the patients that they need to take care of. If we're able to implement AI into healthcare, it would work. It could be able to potentially work smoother because AI systems don't get tired. Yes, they work all day long, but they're a computer. They don't have feelings. They don't get tired. And so we wouldn't be overworking them. They could, for example, scan something in or something like that. It's not a real person, and therefore does not have feelings or anything like that. Medical school is expensive, and it takes a lot of time. If you're able to help AI not, like, give you the answers, but study, like, hey, how can I study, memorize these terms better, then you'll be able to hopefully and maybe graduate at the right time or memorize the things that you may need to memorize. The long hours and long, the strenuous labor is not good for anyone. By implementing AI systems in hospitals or other over time, you'll make the work less difficult and make it easier for nurses, doctors, and other healthcare workers. AI can help with simple things such as documentation. They're able to document things. Nurses and doctors spend a lot of time typing up forms and filling out patient jerks. By implementing voice detection and having AI streamline communication across many different hospitals, it's less work for someone real to do, and it will go by faster, the work. Yeah, I think it's important that we have people, you know, people are able to do people's jobs. You know, AI is able to do, you know, work that no one wants to do. The next point is about cutting costs. A lot of what we hear from AI and, like, news and stuff is, oh, AI is going to be able to cut down costs of things. And, yeah, it can reduce the number of workers, and this affects actual professionals who invest their time and money in education. The next point is about cutting costs. A lot of what we hear from AI is that it's able to cut costs because it reduces the number of workers. Well, if you're reducing the number of workers, people are going to lose their jobs. It's just if you're reducing people's jobs. And that can be kind of frustrating for these workers because they have gone to school and studied long hours for what they wanted to be when they grew up. They invested time and money in the education of healthcare, and AI has simply replaced them. With the research that we have done, one of the websites that we found is called Healthlines, and according to a recent survey of 2,000 consumers by Customer Times, almost 40% say doctors will eventually be replaced by AI. According to Healthcare Medicine, most importantly, 57% of those surveyed said AI will lead to healthcare layoffs, while 13% said that technology will help save jobs. And 10% said AI should help replace doctors in the foreseeable future, and another 28% said that will happen, but shouldn't. I think it's interesting, the difference. You know, 40% of people said they'd be replaced by AI, and 10% said they should do it, and 20% said it will happen, but it shouldn't. And it's the contrast between people's different opinions on this. It's kind of broad. But I personally think it's important to have artificial intelligence help you out. I mean, it's not meant to replace doctors. It's used to assist doctors. It can improve the efficiency of it. If we look back and see the current applications of AI, we've already had it used in health services, not even replacing doctors. Right now we have artificial intelligence. It'll help us with ultrasound or MRI machines, and this will help the radio, whatever it's called, to diagnose what's going on. And it'll just allow doctors to have a more comprehensive and know what to look onto, and I think this is going to become an important part. And to talk about overworked employees, I found a company that already helps out. It's called Mott MD Care. And what they have, and it's like a programming, and you can have, like, administrative people, they just, like, set up question and, like, response, and, you know, a doctor needs to know a certain question to assist a problem that comes up often. They can look it up, you know, in a trusted source. It also allows doctors to transcribe their notes, and it also allows them to be more efficiently sent to other, like, hospitals if they're transferring patients or to other nurses or just allow anyone to look up the information, any of the nurses and doctors to look up the information they need for the patients. And this tool, it's actually really cool. It's been recognized by the World Health Organization just for, like, the amount of work that they're doing and that they're going to change. And they keep improving year after year, and they're just, you know, they're working on a thing now where doctors can, you know, verbally dictate their notes and have it transcribed, and it can organize them and sort them so that, you know, a charge nurse or someone else is able to read the information on there and just be able to understand what's going on more fully. And so I just like that company. I think it's really cool, and I'm just excited to see all the innovation that's going to happen there. Well, we can use AI for a lot of things. AI, like we said before, doesn't have feelings, and AI can not be understanding that humans offer. AI doesn't have a personality as a real person does. And it's really good at analyzing data, but it doesn't replicate, like, personalized care. Like, you know, it's as simple as when you walk into an office, hi, how are you? AI doesn't really do that. And so while it can help, it cannot fully replace. Doctors should always remember that their patients are real people, not some kind of robot. And they're vulnerable beings with problems, and it's not just some robot they can do lab work on. Their life is just as precious and important as their own. Having a real personal connection with them makes a patient feel safer, and a machine won't be able to have a bond like that. Yeah. A big problem with the company I talked about earlier, My MD Care, is that it's very, it'll give, you know, word for word, but it's not individual and personalized. And that can be a big issue just for patients, because you want to have that human connection, and, you know, you want that individualized care. So I think it's hard, you know, not be able to have that. So I just don't know. On another note, while it can max the understanding that humans offer, it can also be helpful to the patient. AI has the possibility to make medical things easier for the patient. According to the National Liberty of Medicine, diabetes affects a lot of the global population, around 11.6% of the population right now. There are many possible ways for AI to be, help administer shots, and, well, not administer the shots, but, like, keep track of the shots. Because I think that, you know, like for some people with diabetes, it's kind of hard. It's not, it can be, can lift a burden off someone who has to keep track of all their bedtime and how much they administered. With AI, they can help keep track of it overall. So I think that's pretty cool. The use of AI in tracking your insulin shots and blood sugar will be able to help the world immensely. Lots of people can be forgetful and also track their shots in the wrong way. Being able to have technology keep track of it is able to be more organized, which is helping to keep you alive. I just think it's really cool that we can use, you know, AI or something to remind us and tell us how much a supplement to take, and it's going to allow, you know, them to take the math out of the equation and just make sure that, you know, they give themselves proper doses of, you know, insulin or glycogen or whatever they need. I think it's just really cool that we're able to, you know, that will make it easier for everyone while being more efficient. Now let's consider how AI might overlook social variables. Patients' needs are often beyond just simple conditions. If you were to Google something like, hey Google, my head is hurting. Like, yes, it could say, oh, you're probably, maybe you haven't had, you have a headache and you are dehydrated, so drink some water or do something like that. And so that's what AI, what Google is able to do for us. But if you were to say, oh, you know, like my head has been hurting for a while. Doctors normally, like, ask you other symptoms, and you're like, oh, okay. And then they continue to talk to you and maybe look at your history or look at your family history to see what potential things could be happening. And every person is different in how much they can do or obtain. For example, like, AI could tell you or Google or, you know, artificial intelligence can be like, hey, like, well, not hey, but, you know, it could say, maybe your head is hurting and you have a headache, drive to the doctor and get yourself an appointment or drive to Walmart and get yourself some Tylenol. But AI doesn't really know what's going on. And if you were, like, disabled or couldn't drive, it wouldn't really understand that. And so, yeah, that's that. When comparing AI to a dermatologist to diagnose skin cancer, the AI outperformed the doctors. They showed the AI many different photos of skin cancer, and over time it was able to learn from itself and it's growing more and more accurate, just like the testing that we referred to earlier, the more time that it's programmed. Now, this may seem like it is kind of useless because people can't say stuff. And it's one of the important things to remember is when we're referring to Google, it's not Google's search engine. It's Google's Gemini or BART or whatever it's called. And it's also going to include, you know, Snapchat AI, Chat QBP3. All of these different chat bots, they're all different, but they're all capable of helping and interacting with us in, like, almost human-like way. So I think it's important to remember that. So, essentially, AI can be used not only to help doctors, but from, like, an administrative perspective. It can help with accounting because there's a lot of times where hospitals are found having, like, some sort of fraudulent scandal, and so it's like, oh, this money went missing or the accountant messed up or something happened. And so by using AI with humans to check, of course, we'd be able to, you know, reduce these incidents and reduce the risk associated with people's money from a hospital standpoint because that's not good for hospitals to be mismanaging funds, you know. And so I think that's kind of cool. But I just think with this AI, we have to find a balance, you know, because we need to still allow workers, like we talked about earlier, to have jobs. You know what I mean? Mm-hmm. But we still need to have them. There's, like, opportunities in the AI, which is really helpful, and it can just allow us. It can give us a better balance of what we're doing. Yeah, yeah. I think that would be, you know, just balance it out and just help us, you know, good fun, good flow, good mix. AI, yeah, AI isn't perfect, and over time, it can get better. It can learn, just, like, with the test. Mm-hmm. And, like, everything in life, AI will make mistakes. For example, if a teenage girl were to go to the hospital, AI is just, like, super helpful here. We can help out the doctors by just, like, reducing, like, the stuff that they have to do, you know, like their shoulders and things like that, love, and other things. Like, everything in life, AI will still make mistakes. We still make mistakes every single day, and that's just part of life. If a teenage girl, for example, if a teenage girl were to go to the hospital complaining about her high heart rate, so she was complaining about a high heart rate, and she was good at the hospital, and her blood gets drawn, everything comes back normal, and it was analyzed by AI, it comes back normal, and so she goes home, and then six weeks later, she dies due to cardiac arrest. Hypothetically, in this story, if she dies six weeks later due to cardiac arrest because AI analyzed her data, it would come back to her. That's not really, like, oh, yeah, like, we should not have that happen. So, with real doctors, they would have probably asked questions about their family history, done some more research, how long have they noticed their heart rate, and stuff like that. Mistakes can also become biased outcomes, and so if AI keeps doing things like that, we have to just learn to have balance in what we're doing with AI. And this can also harm patients and their desire to come to the hospital to stay home if AI is going to do it for them and do the doctor's job. And this decreases their desire and their awareness to be able to counter the untreated effects. This can also have unequal treatment, misdiagnosed, or under-diagnosis of certain demographic groups and all people. Yeah, add to that, I just, I don't know, like, human doctors, you know, we have, you know, humans feel compassion for each other, and we just don't know if AI is able to feel this similar way towards us. And there still is biases that can be put into this AI, and it will be used for an entire time. And when doctors, you know, are, they can be charged with, you know, malpractice, but what happens if the AI is doing it wrong? You know, would they stop using it? Would they change it? It's just a lot of questions surrounding, you know, should this be used? And, you know, how, what is happening? So although you can still do a lot with AI in healthcare, human surveillance is still essential. One example of this is that surgery robots operate logically as opposed to empathy. AI has been around for a few decades and continues to mature. As this area advances, there is more interaction between healthcare professionals and tech experts. According to Drexel University, Christopher C. Nang, who has a Ph.D. in Clinical Neurology, explains, AI requires human input and review to be leveraged effectively. AI still requires input. We can't trust everything we say and need to be checked. It's still learning and growing just like a human would. As AI develops, use of education are required for medical professors to operate in their fields. So should we trust AI to do something like that if someone has to have lots of execution background? AI is generally dependent on data networks. AI systems are susceptible to security risks. As AI uses data to make systems smarter and more accurate, cyber takes will incorporate AI to become smarter, which takes successes and failures, making it harder for you to be able to predict it. Yeah, just like adding on to that, a lot of, like, data breaches happen online. Like over, you know, on a piece of paper, you know, there's a less likely chance for, you know, that stuff to be stolen and taken. But on, you know, over the computer, I mean, everyone's data is susceptible. And, you know, we have to check AI, and it makes them redundant, you know. If you think somebody has the answer, if they're the answer, people are going to get lazy or, you know, just not really want to do it. But it's just important to, you know, know that we have to do it. This is part of our due diligence. And over time, it'll get better and better. But I just think for the foreseeable future, we will have to check what AI does. The best thing that we can start doing now is educating ourselves and being open and honest about the use of AI. Yeah, it's important for us to, you know, to cite our sources and know where our information is coming from. And we're coming to a day and age where it's hard to know, you know, what we can trust. And so it's really important, you know, for us to realize. It's good to be educated about the topic. Also, if you're a doctor, it's good to be open with your patients so that they can be aware of it as well. We hope to hear from you again next week. If you're hearing this message, that means you have made it to almost the end of our podcast. This episode, we talked about the pros and cons about AI. More specifically, we talked about how AI still has much more work to do before it can be fully implemented to the medical field. It needs to be checked by professionals. And it's not developed enough to be able to help humans. And there's too much of a risk right now for AI to be completely in charge of the medical field. We hope to hear from you again next week. The knowledge of AI in healthcare is very important. What we decide to do with our health will undoubtedly affect us in one way or another. For now, it is best that we avoid AI in healthcare. We hope that you enjoyed. And if you did, please leave us a review on Apple Podcasts or Spotify. Or wherever you see in this podcast. Please share this with others who you think this might be beneficial to. We hope to hear from you in the weeks ahead. Next week, what are we going to talk about? Next week, we're going to talk about technology and the healthcare industry. And not artificial technology, but modern technology that we have added to us. Thank you guys for listening us. We hope to see you next week. Peace out.