{"id":2324,"date":"2024-04-25T11:46:41","date_gmt":"2024-04-25T09:46:41","guid":{"rendered":"https:\/\/nubisoft.io\/blog\/?p=2324"},"modified":"2024-06-10T16:49:06","modified_gmt":"2024-06-10T14:49:06","slug":"ai-llms-4-healthcare-innovations","status":"publish","type":"post","link":"https:\/\/nubisoft.io\/blog\/ai-llms-4-healthcare-innovations\/","title":{"rendered":"AI &#038; LLMs: 4 Healthcare Innovations"},"content":{"rendered":"\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"507\" src=\"https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/04\/4-Healthcare-Innovations-1024x507.png\" alt=\"A graphic with the caption \u201cAI and LLMs: 4 Healthcare Innovations\u201d\" class=\"wp-image-2318\" srcset=\"https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/04\/4-Healthcare-Innovations-1024x507.png 1024w, https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/04\/4-Healthcare-Innovations-300x149.png 300w, https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/04\/4-Healthcare-Innovations-768x380.png 768w, https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/04\/4-Healthcare-Innovations-1200x594.png 1200w, https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/04\/4-Healthcare-Innovations.png 1248w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n<div class=\"lazyblock-text-ZHEIf9 wp-block-lazyblock-text\">\n  <div style=\"--size-margin-text-block: 0px;\" class=\"lazyblocks-text-wrapper wp-block-lazyblock-text\">\n    <div class=\"lazyblocks-text-columns-wrapper\">\n              <div class=\"lazyblocks-text-column\n                          lazyblocks-text-big-font-quote\n        \" style=\"--size-padding-text-block: 0px;\">\n          <p><span style=\"font-weight: 400;\">Artificial intelligence in healthcare is no longer a tale of the future &#8211; it is a reality that is happening right before our eyes. With more and more tools on the market to help doctors perform tests, make diagnoses, or develop treatment plans, we don&#8217;t think anyone doubts that artificial intelligence is a true<\/span><b> revolution in healthcare.<\/b><\/p>\n<blockquote>\n<p style=\"text-align: left;\"><span style=\"font-weight: 400;\">By 2030, the medical artificial intelligence market is expected to be worth <\/span><b>$188 billion<\/b><b>.\u00a0<\/b><\/p>\n<\/blockquote>\n<p><span style=\"font-weight: 400;\">According to experts, implementing artificial intelligence in healthcare will enable doctors to <\/span><b>increase the time they spend <\/b><b>with patients<\/b><b> by nearly 20%<\/b><span style=\"font-weight: 400;\">, thanks to reducing the time spent on administrative tasks. <\/span><span style=\"font-weight: 400;\">In addition, nurses are expected to spend approximately 8% more time with patients due to a reduction in time spent on administrative and regulatory tasks.<\/span><\/p>        <\/div>\n      \n      \n    <\/div>\n  <\/div>\n<\/div>\n\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"396\" src=\"https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/04\/Frame-30-1-1024x396.png\" alt=\"The image shows two vertical bars, one green and one blue, representing the percentage increase in time spent with patients by physicians (20%) and nurses (8%) due to AI implementation. The green bar is significantly taller than the blue bar, indicating a larger impact on physicians' time.\" class=\"wp-image-2197\" srcset=\"https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/04\/Frame-30-1-1024x396.png 1024w, https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/04\/Frame-30-1-300x116.png 300w, https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/04\/Frame-30-1-768x297.png 768w, https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/04\/Frame-30-1.png 1165w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n<div class=\"lazyblock-text-ZKEFV5 wp-block-lazyblock-text\">\n  <div style=\"--size-margin-text-block: 20px;\" class=\"lazyblocks-text-wrapper wp-block-lazyblock-text\">\n    <div class=\"lazyblocks-text-columns-wrapper\">\n              <div class=\"lazyblocks-text-column\n                          lazyblocks-text-big-font-quote\n        \" style=\"--size-padding-text-block: 0px;\">\n          <p><span style=\"font-weight: 400;\">The numbers speak for themselves: AI is having a major moment in healthcare. The hype is real, but so are the misconceptions. Some people think AI will be our healthcare superhero, solving every problem. Others worry it&#8217;ll replace doctors entirely.\u00a0 <\/span><\/p>\n<p><span style=\"font-weight: 400;\">Not quite! AI has limits, and we need to be clear on those. <\/span>That&#8217;s why we&#8217;re here to cut through the noise. This article dives into real-world examples of how large language models (LLMs) are already helping medical professionals.<\/p>\n<blockquote>By understanding how these tools work and what they&#8217;re doing today, we can build smarter and more effective solutions for tomorrow.<\/blockquote>        <\/div>\n      \n      \n    <\/div>\n  <\/div>\n<\/div>\n\n<div class=\"lazyblock-text-bUf1a wp-block-lazyblock-text\">\n  <div style=\"--size-margin-text-block: 20px;\" class=\"lazyblocks-text-wrapper wp-block-lazyblock-text\">\n    <div class=\"lazyblocks-text-columns-wrapper\">\n              <div class=\"lazyblocks-text-column\n                  lazyblocks-text-column-border\n                \" style=\"--size-padding-text-block: 70px;\">\n          <h3><b>What is the Large Language Model?<\/b><\/h3>\n<p><span style=\"font-weight: 400;\">First, let&#8217;s explain what an LLM is. A Large Language Model is a type of artificial intelligence algorithm that uses deep learning techniques and large data sets to understand and produce human-like output. They can predict the next word in a sequence based on the previous words. LLMs can answer questions, write essays, summarize text, translate languages, and even write code.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Their capabilities make them applicable to a variety of healthcare areas, 4 of which we have highlighted below and selected representative case studies for each. These areas are:<\/span><\/p>\n<ul>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><a href=\"#administration\"><strong>Administration<\/strong><\/a><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><a href=\"#education\"><strong>Education<\/strong><\/a><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><a href=\"#diagnostics\"><strong>Diagnostics<\/strong><\/a><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><a href=\"#research\"><strong>Research<\/strong><\/a><\/li>\n<\/ul>        <\/div>\n      \n      \n    <\/div>\n  <\/div>\n<\/div>\n\n<div class=\"lazyblock-text-Z1gPrGn wp-block-lazyblock-text\" id=\"administration\">\n  <div style=\"--size-margin-text-block: 20px;\" class=\"lazyblocks-text-wrapper wp-block-lazyblock-text\" id=\"administration\">\n    <div class=\"lazyblocks-text-columns-wrapper\">\n              <div class=\"lazyblocks-text-column\n                  lazyblocks-text-column-border\n                          lazyblocks-text-big-font-quote\n        \" style=\"--size-padding-text-block: 70px;\">\n          <h2>LLMs in healthcare administration<\/h2>\n<p><span style=\"font-weight: 400;\">The first area of healthcare where AI is poised to make a big difference is the administrative side of healthcare. <\/span><\/p>\n<p><span style=\"font-weight: 400;\">In 2018 the American Medical Association conducted a study which was a deep dive into the time physicians spent on administrative tasks and paperwork. The report&#8217;s authors collected data from more than 20,000 physicians across nearly 30 specialties, and the results were nothing short of &#8220;mind-boggling&#8221; revealing that:<\/span><\/p>\n<ul>\n<li><span style=\"font-weight: 400;\">Nearly a third of physicians spent<\/span><b> 20 hours or more a week<\/b><span style=\"font-weight: 400;\"> on administrative tasks. <\/span><\/li>\n<li><span style=\"font-weight: 400;\">Despite the increase in administrative work, the time physicians spent seeing patients remained relatively unchanged<\/span><span style=\"font-weight: 400;\">.\u00a0<\/span><\/li>\n<\/ul>\n<p><span style=\"font-weight: 400;\"><a href=\"https:\/\/www.sabp.nhs.uk\/our-services\/mental-health\/mind-matters-surrey\/mind-matters\">Mind Matters<\/a> (Surrey and Borders Partnership NHS Foundation Trust) faced a similar problem. An increase in referrals and recruitment difficulties led to a significant administrative burden at Mind Matters. This led to inefficiencies in the referral process and negatively impacted staff wellbeing. <\/span><\/p>\n<blockquote>\n<p style=\"text-align: left;\"><span style=\"font-weight: 400;\">In response to these challenges, the Mind Matters team decided to implement an AI chatbot.\u00a0<\/span><\/p>\n<\/blockquote>\n<p>The goal was to :<\/p>\n<ul>\n<li>simplify the referral process,<\/li>\n<li>reduce administrative confusion,<\/li>\n<li>increase team efficiency,<\/li>\n<li>improve staff well-being.<\/li>\n<\/ul>\n<p><span style=\"font-weight: 400;\">The chatbot was available on the website. It was the patient&#8217;s first contact with the facility. It engaged patients in a natural conversation and guided them through the referral process. This replaced long and cumbersome registration forms and made the whole process more flexible (for example, the bot did not ask non-smokers questions about smoking). Based on the answers, there was also a process of initial redirection to more appropriate services if indicated. The system applied the appropriate notification if the patient&#8217;s condition required the fastest response.<\/span><\/p>        <\/div>\n      \n      \n    <\/div>\n  <\/div>\n<\/div>\n\n<div class=\"lazyblock-text-Z1X7S9j wp-block-lazyblock-text\">\n  <div style=\"--size-margin-text-block: 70px;\" class=\"lazyblocks-text-wrapper wp-block-lazyblock-text\">\n    <div class=\"lazyblocks-text-columns-wrapper\">\n              <div class=\"lazyblocks-text-column\n                          lazyblocks-text-big-font-quote\n        \" style=\"--size-padding-text-block: 20px;\">\n          <p><strong>Results?<\/strong><\/p>\n<ul>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Nearly all patients who provided feedback reported that Limbic was helpful (98,72%)<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">30% increase in referral handling<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Average medical worker time saved is 15 minutes per referral, which translated into 2,000 hours (as of the date of writing the case study)<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Nearly 20% of referrals were identified as ineligible and redirected to a more appropriate service, creating additional capacity for the service.<\/span><\/li>\n<\/ul>\n<blockquote>\n<p style=\"text-align: left;\"><span style=\"font-weight: 400;\">Limbic was connected to NHS Spine (the national IT infrastructure supporting health and social services in England), so it could <strong>independently<\/strong> verify the eligibility of referrals. <\/span><\/p>\n<\/blockquote>\n<p><span style=\"font-weight: 400;\">Over time, Limbic has become the most popular mental health triage chatbot within the NHS and has been adopted by 17 regions across the UK. Currently accessible to over 9 million people, Limbic has facilitated the release of over 20,000 clinical hours for IAPT services.<\/span><\/p>        <\/div>\n      \n      \n    <\/div>\n  <\/div>\n<\/div>\n\n<div class=\"lazyblock-text-xONX4 wp-block-lazyblock-text\" id=\"education\">\n  <div style=\"--size-margin-text-block: 70px;\" class=\"lazyblocks-text-wrapper wp-block-lazyblock-text\" id=\"education\">\n    <div class=\"lazyblocks-text-columns-wrapper\">\n              <div class=\"lazyblocks-text-column\n                  lazyblocks-text-column-border\n                          lazyblocks-text-big-font-quote\n        \" style=\"--size-padding-text-block: 70px;\">\n          <h2><span style=\"font-weight: 400;\">Education<\/span><\/h2>\n<p><span style=\"font-weight: 400;\">The field of medicine is inextricably linked to education. Here, LLMs emerge as transformative tools with the potential to significantly enhance the learning experience.<\/span><\/p>\n<blockquote>\n<p style=\"text-align: left;\"><span style=\"font-weight: 400;\">And to be clear &#8211; LLMs are not designed to facilitate shortcuts in medical education.\u00a0 <\/span><\/p>\n<\/blockquote>\n<p><span style=\"font-weight: 400;\">Rather, they offer the opportunity to equip students with more effective learning methodologies that transcend traditional rote memorization. Imagine a shift from textbook-heavy learning to a dynamic, interactive approach that fosters deeper understanding.<\/span><\/p>        <\/div>\n      \n      \n    <\/div>\n  <\/div>\n<\/div>\n\n<div class=\"lazyblock-text-2vyL1c wp-block-lazyblock-text\">\n  <div style=\"--size-margin-text-block: 70px;\" class=\"lazyblocks-text-wrapper wp-block-lazyblock-text\">\n    <div class=\"lazyblocks-text-columns-wrapper\">\n              <div class=\"lazyblocks-text-column\n                          lazyblocks-text-big-font-quote\n        \" style=\"--size-padding-text-block: 20px;\">\n          <h4>Relevance to the medical sector<\/h4>\n<p><span style=\"font-weight: 400;\">This focus on enhanced learning is particularly relevant given the pressing issue of the healthcare workforce shortage.\u00a0 One contributing factor is the notoriously demanding nature of medical education.\u00a0 LLMs, with their ability to work creatively with vast amounts of information, have the potential to democratize medical knowledge, making it more accessible and efficient to learn.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Moreover, the problem is not unique to students. <\/span><\/p>\n<blockquote>\n<p style=\"text-align: left;\"><span style=\"font-weight: 400;\">Medical education is a never-ending process &#8211; after all, practicing physicians need to stay abreast of the latest research to improve patient care.<\/span><\/p>\n<\/blockquote>        <\/div>\n      \n      \n    <\/div>\n  <\/div>\n<\/div>\n\n<div class=\"lazyblock-text-NcPwO wp-block-lazyblock-text\">\n  <div style=\"--size-margin-text-block: 70px;\" class=\"lazyblocks-text-wrapper wp-block-lazyblock-text\">\n    <div class=\"lazyblocks-text-columns-wrapper\">\n              <div class=\"lazyblocks-text-column\n                \" style=\"--size-padding-text-block: 20px;\">\n          <h4>Google MedPaLM and USMLE<\/h4>\n<p><span style=\"font-weight: 400;\">Consider the advancements made by Google. <\/span><\/p>\n<p><span style=\"font-weight: 400;\">In December 2022, their Med-PaLM model, trained on a comprehensive dataset of medical information, demonstrably achieved success on an exam simulating the USMLE (a crucial licensing exam for doctors in the United States). The model scored an <strong>impressive 67.5%<\/strong>, exceeding the passing threshold. <\/span><\/p>\n<p><span style=\"font-weight: 400;\">Notably, just a few months later in July 2023, the next generation, Med-PaLM 2, achieved an even higher score of 86.5% on the same test<\/span><span style=\"font-weight: 400;\">!<\/span><\/p>        <\/div>\n      \n      \n    <\/div>\n  <\/div>\n<\/div>\n\n<div class=\"wp-block-image is-style-default\">\n<figure class=\"aligncenter size-large is-resized\"><img decoding=\"async\" src=\"https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/04\/gifgit.gif\" alt=\"\" style=\"width:610px;height:auto\"\/><figcaption class=\"wp-element-caption\">source: <a href=\"https:\/\/sites.research.google\/med-palm\/\" target=\"_blank\" rel=\"noreferrer noopener\">https:\/\/sites.research.google\/med-palm\/<\/a><\/figcaption><\/figure>\n<\/div>\n\n<div class=\"lazyblock-text-ZiJXau wp-block-lazyblock-text\">\n  <div style=\"--size-margin-text-block: 30px;\" class=\"lazyblocks-text-wrapper wp-block-lazyblock-text\">\n    <div class=\"lazyblocks-text-columns-wrapper\">\n              <div class=\"lazyblocks-text-column\n                \" style=\"--size-padding-text-block: 20px;\">\n          <h4>How can LLMs support medical education?<\/h4>\n<p><span style=\"font-weight: 400;\">It was then realized that properly prepared chatbots could play an important role in the medical education process. Their ability to process large amounts of information, not just text, but also images, charts, and infographics, makes them a powerful tool to assist students in their studies.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Specific ways to support medical education are numerous. One of the most promising ways is to use chatbots as <\/span><b>personalized learning assistants<\/b><span style=\"font-weight: 400;\"> that explain certain complex issues in a simpler, more understandable way. Here is an example of such a conversation:<\/span><\/p>\n<p><b>Human: <\/b><i><span style=\"font-weight: 400;\">Explain to a five-year-old how the human circulatory system works?<br \/><\/span><\/i><b>GPT-4<\/b><span style=\"font-weight: 400;\">:<\/span><i><span style=\"font-weight: 400;\"> Sure, let&#8217;s imagine the human body as a big city. The circulatory system is like the city&#8217;s delivery service. <br \/><\/span><\/i><i><span style=\"font-weight: 400;\">Heart: This is like the delivery center. It&#8217;s a strong muscle that pumps blood. It works all the time without any breaks.<br \/><\/span><\/i><i><span style=\"font-weight: 400;\">Blood: This is like the delivery trucks. They carry oxygen and nutrients (the food and air cells need) to all parts of the body. They also pick up waste, like carbon dioxide, to take it away.<br \/><\/span><\/i><i><span style=\"font-weight: 400;\">Blood Vessels: These are the roads for the delivery trucks. There are big highways (arteries), smaller streets (capillaries), and return routes (veins). The arteries carry blood away from the heart to the body, capillaries connect arteries to veins in tissues and organs, and veins carry blood back to the heart.<br \/><\/span><\/i><i>So, the heart sends out the blood trucks on the vessel roads to deliver oxygen and nutrients and pick up waste. Then, they return to the heart to get ready for the next trip. This happens over and over, all day and all night, to keep our body city running smoothly!<\/i><\/p>\n<p><span style=\"font-weight: 400;\">Such an assistant can also help creatively develop certain material so that it is easier to remember. Here is an example of using GPT to create a mnemonic. <\/span><\/p>\n<p><strong>Prompt: &#8220;create a mnemonic for the treatment of pre-eclampsia&#8221;<\/strong><\/p>        <\/div>\n      \n      \n    <\/div>\n  <\/div>\n<\/div>\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-large is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"999\" src=\"https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/04\/Mnemonic-pre-eclampsia-1024x999.jpg\" alt=\"The image shows a screenshot of a chat from the ChatGPT tool. Prompt to: create a mnemonic for the treatment of pre-eclampsia. The ChatGPT response is a Mnemonic that makes it easy to remember the information - \u2018TEAM PRE ECLAMPSIA\u2019 is used as an abbreviation and expanded: T- Tocolysis (to slow down labour), E- Exlampsia prophylaxis, A- Anti-hypertensive treatment, and so on.\" class=\"wp-image-2258\" style=\"width:608px;height:auto\" srcset=\"https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/04\/Mnemonic-pre-eclampsia-1024x999.jpg 1024w, https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/04\/Mnemonic-pre-eclampsia-300x293.jpg 300w, https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/04\/Mnemonic-pre-eclampsia-768x749.jpg 768w, https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/04\/Mnemonic-pre-eclampsia-1536x1498.jpg 1536w, https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/04\/Mnemonic-pre-eclampsia-1200x1170.jpg 1200w, https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/04\/Mnemonic-pre-eclampsia.jpg 1540w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><figcaption class=\"wp-element-caption\">source: <a href=\"https:\/\/www.statnews.com\/2023\/02\/01\/promises-pitfalls-chatgpt-assisted-medicine\/\" data-type=\"link\" data-id=\"https:\/\/www.statnews.com\/2023\/02\/01\/promises-pitfalls-chatgpt-assisted-medicine\/\">Statnews<\/a><\/figcaption><\/figure>\n<\/div>\n\n<div class=\"lazyblock-text-4LLh1 wp-block-lazyblock-text\">\n  <div style=\"--size-margin-text-block: 70px;\" class=\"lazyblocks-text-wrapper wp-block-lazyblock-text\">\n    <div class=\"lazyblocks-text-columns-wrapper\">\n              <div class=\"lazyblocks-text-column\n                \" style=\"--size-padding-text-block: 20px;\">\n          <h4>What are the limitations of the LLM in medical education?<\/h4>\n<p><span style=\"font-weight: 400;\">While large language models (LLMs) offer immense potential for tool design, it&#8217;s important to recognize their limitations. In the above examples, freely available tools such as GPT-3.5 have been used. This has two major drawbacks: <\/span><\/p>\n<ul>\n<li><span style=\"font-weight: 400;\">outdated information<\/span><\/li>\n<li><span style=\"font-weight: 400;\"> tendency to generate inaccurate or misleading content (hallucination).<\/span><\/li>\n<\/ul>\n<p><span style=\"font-weight: 400;\">While paid solutions can mitigate these problems, they can&#8217;t eliminate them. Therefore, a clear understanding of these limitations is essential when designing tools or crafting prompts.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Below is an example of the potential pitfalls of LLMs in education. While the errors in the LLM&#8217;s response may seem apparent, that&#8217;s not always the case. Students who lack the skills to effectively verify the information presented by these tools are more susceptible to being misled.<\/span><\/p>        <\/div>\n      \n      \n    <\/div>\n  <\/div>\n<\/div>\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-full is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"1024\" src=\"https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/04\/frog_copilot.png\" alt=\"The image shows a frog with its internal organs highlighted, generated by artificial intelligence. It appears to be correct, but on closer inspection it is clear that almost all the organ names are made up, as is the internal structure of the frog.\" class=\"wp-image-2268\" style=\"width:589px;height:auto\" srcset=\"https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/04\/frog_copilot.png 1024w, https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/04\/frog_copilot-300x300.png 300w, https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/04\/frog_copilot-150x150.png 150w, https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/04\/frog_copilot-768x768.png 768w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><figcaption class=\"wp-element-caption\">Image generated by Copilot. <br>Prompt: \u201cCreate an infographic showing how the frog&#8217;s digestive system functions.\u201d<br><\/figcaption><\/figure>\n<\/div>\n\n<div class=\"lazyblock-text-1VnrqP wp-block-lazyblock-text\" id=\"research\">\n  <div style=\"--size-margin-text-block: 70px;\" class=\"lazyblocks-text-wrapper wp-block-lazyblock-text\" id=\"research\">\n    <div class=\"lazyblocks-text-columns-wrapper\">\n              <div class=\"lazyblocks-text-column\n                  lazyblocks-text-column-border\n                          lazyblocks-text-big-font-quote\n        \" style=\"--size-padding-text-block: 70px;\">\n          <h2><span style=\"font-weight: 400;\">How do LLMs improve medical research processes?<\/span><\/h2>\n<p><span style=\"font-weight: 400;\">Another area of modern medicine where LLMs are already being used is in research. And just as in the case of education, concurrent use cases for tools with this technology in the research process are numerous.<\/span><\/p>\n<ul>\n<li><strong>Data analysis:<\/strong> LLMs can assist in analyzing complex medical data, identifying patterns, and generating insights that can guide research.<\/li>\n<li data-sourcepos=\"3:1-3:308\"><strong>Literature summarization:<\/strong> LLMs can wade through massive amounts of medical research papers and clinical data to identify trends, synthesize information, and generate concise summaries of complex topics.\u00a0<\/li>\n<li data-sourcepos=\"4:1-4:274\"><strong>Hypothesis generation:<\/strong> LLMs can analyze existing research and patient data to identify potential connections and suggest new hypotheses for further investigation.\u00a0<\/li>\n<li data-sourcepos=\"5:1-5:218\"><strong>Drug discovery and development:<\/strong>\u00a0LLMs can be used to analyze vast datasets of molecular structures and biological pathways to help identify potential drug targets and accelerate the development of new treatments.<\/li>\n<li><strong>Paper Writing:<\/strong> LLMs can assist in writing research papers, ensuring they are well-structured and adhere to the relevant style guides.<\/li>\n<\/ul>\n<h4>LLMs in the drug discovery process<\/h4>\n<p><span style=\"font-weight: 400;\">The discovery of new drugs is a complex, time-consuming, and costly process. It often involves years of meticulous research, rigorous testing, and substantial financial investment. <\/span><\/p>\n<blockquote>\n<p style=\"text-align: left;\"><span style=\"font-weight: 400;\">The traditional drug discovery process can take up to 15 years and cost over $2.6 billion, according to a study by the Tufts Center for the Study of Drug Development.<\/span><\/p>\n<\/blockquote>\n<p><span style=\"font-weight: 400;\">In this context, LLMs have emerged as a transformative tool in the medical research field, particularly in drug discovery. Its capabilities allow it to analyze vast amounts of the biomedical literature, patents, and clinical trial data in a fraction of the time it would take a human researcher. This can significantly speed up the initial stages of drug discovery, which involve identifying potential drug targets and candidate molecules.<\/span><\/p>        <\/div>\n      \n      \n    <\/div>\n  <\/div>\n<\/div>\n\n<div class=\"lazyblock-text-Z18mQKu wp-block-lazyblock-text\">\n  <div style=\"--size-margin-text-block: 70px;\" class=\"lazyblocks-text-wrapper wp-block-lazyblock-text\">\n    <div class=\"lazyblocks-text-columns-wrapper\">\n              <div class=\"lazyblocks-text-column\n                          lazyblocks-text-big-font-quote\n        \" style=\"--size-padding-text-block: 20px;\">\n          <h4>BenevolentAI case study<\/h4>\n<p>A positive example of the use of such technology is BenevolentAI, a company based in the UK. They used an LLM they developed to identify a potential treatment for COVID-19. The company&#8217;s AI-based hypothesis was that &#8220;baricitinib&#8221;, a drug approved for the treatment of rheumatoid arthritis and other diseases, might be effective in treating COVID-19.<\/p>\n<p><span style=\"font-weight: 400;\">This hypothesis was validated through a large-scale randomized clinical trial sponsored by the National Institute of Allergy and Infectious Diseases (NIAID). The trial included more than 1,000 patients and began on May 8 to assess the efficacy and safety of &#8220;baricitinib&#8221; plus remdesivir versus remdesivir in hospitalized patients with COVID-19.<\/span><\/p>\n<blockquote>\n<p style=\"text-align: left;\"><span style=\"font-weight: 400;\"> The results showed that \u201cbaricitinib\u201d in combination with remdesivir achieved the primary endpoint of the study, which was a statistically significant reduction in the time to recovery for patients in comparison with remdesivir alone.<\/span><\/p>\n<\/blockquote>\n<p><span style=\"font-weight: 400;\">Furthermore, data from Eli Lilly\u2019s COV-BARRIER trial showed that &#8220;baricitinib&#8221; reduced deaths in hospitalized COVID-19 patients by 38%. This is the largest clinical effect reported to date for a reduction in mortality in this patient population<\/span><span style=\"font-weight: 400;\">.<\/span><\/p>        <\/div>\n      \n      \n    <\/div>\n  <\/div>\n<\/div>\n\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"300\" src=\"https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/04\/Baricitinib-1024x300.png\" alt=\"The image shows a light blue robot and text. The text says &quot;baricitinib&quot; is a drug developed by AI and researchers that reduced deaths in hospitalized COVID-19 patients by 38%.\" class=\"wp-image-2276\" srcset=\"https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/04\/Baricitinib-1024x300.png 1024w, https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/04\/Baricitinib-300x88.png 300w, https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/04\/Baricitinib-768x225.png 768w, https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/04\/Baricitinib.png 1168w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n<div class=\"lazyblock-text-Z1d6xcO wp-block-lazyblock-text\">\n  <div style=\"--size-margin-text-block: 0px;\" class=\"lazyblocks-text-wrapper wp-block-lazyblock-text\">\n    <div class=\"lazyblocks-text-columns-wrapper\">\n              <div class=\"lazyblocks-text-column\n                \" style=\"--size-padding-text-block: 0px;\">\n          <p><span style=\"font-weight: 400;\">In summary, the implementation of this solution has yielded promising results. The use of AI by BenevolentAI in research for COVID-19 treatment led to the identification of a potential treatment, which was validated through clinical trials and showed promising results in reducing recovery time and mortality in hospitalized COVID-19 patients. The LLMs suggest novel biological targets that could evolve into a drug discovery program if experimentally validated, deemed safe, and able to provide an opportunity to differentiate.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">However, building trust in the methodology and preventing algorithmic biases remain key barriers to AI deployment.<\/span><\/p>        <\/div>\n      \n      \n    <\/div>\n  <\/div>\n<\/div>\n\n<div class=\"lazyblock-text-qvtAB wp-block-lazyblock-text\" id=\"diagnostics\">\n  <div style=\"--size-margin-text-block: 70px;\" class=\"lazyblocks-text-wrapper wp-block-lazyblock-text\" id=\"diagnostics\">\n    <div class=\"lazyblocks-text-columns-wrapper\">\n              <div class=\"lazyblocks-text-column\n                  lazyblocks-text-column-border\n                          lazyblocks-text-big-font-quote\n        \" style=\"--size-padding-text-block: 70px;\">\n          <h2><span style=\"font-weight: 400;\">How LLMs help diagnose?<\/span><\/h2>\n<p><span style=\"font-weight: 400;\">Many people might think using AI in medicine is all about diagnosing illnesses. We can easily imagine a scenario where a chatbot asks a patient a few questions, analyzes their answers along with other information like treatment history, and then &#8211; poof! &#8211;\u00a0 the AI delivers a diagnosis and treatment plan.<\/span><\/p>\n<p><strong>But the reality is far more nuanced. Diagnosing a patient is a complex puzzle that requires a doctor&#8217;s knowledge, in-depth data analysis, and even strong interpersonal skills to understand the patient&#8217;s experience.\u00a0 <\/strong><\/p>\n<blockquote>\n<p style=\"text-align: left;\"><span style=\"font-weight: 400;\">Recreating that process entirely with a tool is extremely difficult, if not impossible, at this stage.<\/span><\/p>\n<\/blockquote>\n<p><span style=\"font-weight: 400;\">It is easier, although still a significant challenge, to create an AI tool that supports the work of a human, thus saving his time and energy. Based on real-life examples, LLMs can:<\/span><\/p>\n<ul>\n<li><span style=\"font-weight: 400;\">improve clinical workflow efficiency, <\/span><\/li>\n<li><span style=\"font-weight: 400;\">reduce diagnostic errors,<\/span><\/li>\n<li><span style=\"font-weight: 400;\">assist healthcare professionals in making timely and accurate diagnoses.<\/span><\/li>\n<\/ul>\n<h4><strong>Can LLMs assist with Differential Diagnosis?<\/strong>\u00a0<\/h4>\n<p><span style=\"font-weight: 400;\">A real-life example of the use of LLMs in diagnostics is the test application of such a tool to support differential diagnosis. The study was conducted, again, by Google. They used a model with a transformative architecture &#8211; PaLM 2, fine-tuned with data from the medical domain.\u00a0\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">The objective was to introduce an LLM optimized for diagnostic reasoning and evaluate its ability to generate a DDx alone or as an aid to clinicians. DDX is a process where a doctor considers multiple possible conditions that could be causing a patient&#8217;s symptoms.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">The study involved 20 clinicians who evaluated 302 challenging, real-world medical cases from the New England Journal of Medicine (NEJM) case reports. Each case report was read by two clinicians who were randomized to one of two support conditions: <\/span><\/p>\n<ul>\n<li><span style=\"font-weight: 400;\">From search engines and standard medical resources,<\/span><\/li>\n<li><span style=\"font-weight: 400;\"> or LLM support in addition to these tools.<\/span><\/li>\n<\/ul>\n<p><span style=\"font-weight: 400;\">All clinicians performed a baseline, unassisted DDx before using each of the assistive tools.<\/span><\/p>\n<h4>The results were promising.<\/h4>\n<p><span style=\"font-weight: 400;\">The LLM tool was very good at suggesting possible diagnoses (differential diagnoses). In fact, it did better than doctors on its own!<\/span><\/p>\n<ul>\n<li><span style=\"font-weight: 400;\">Almost 60% accuracy in the top 10 diagnoses (vs. only 33% for doctors). <\/span><\/li>\n<li><span style=\"font-weight: 400;\">Doctors who used the LLM to help them also came up with better diagnoses than doctors who didn&#8217;t.<\/span><\/li>\n<\/ul>\n<blockquote>\n<p style=\"text-align: left;\"><span style=\"font-weight: 400;\">In conclusion, this study suggests that LLMs hold promise for improving clinicians&#8217; diagnostic reasoning and accuracy, especially in complex cases. <\/span><\/p>\n<\/blockquote>\n<p><span style=\"font-weight: 400;\">Further real-world evaluation is warranted to assess their ability to empower physicians and expand patient access to specialist-level expertise.\u00a0<\/span><\/p>        <\/div>\n      \n      \n    <\/div>\n  <\/div>\n<\/div>\n\n<div class=\"lazyblock-text-1cfewu wp-block-lazyblock-text\">\n  <div style=\"--size-margin-text-block: 70px;\" class=\"lazyblocks-text-wrapper wp-block-lazyblock-text\">\n    <div class=\"lazyblocks-text-columns-wrapper\">\n              <div class=\"lazyblocks-text-column\n                          lazyblocks-text-big-font-quote\n        \" style=\"--size-padding-text-block: 20px;\">\n          <h2><span style=\"font-weight: 400;\">Summary<\/span><\/h2>\n<p><span style=\"font-weight: 400;\">The AI-powered tech revolution is just starting, and its reach is expanding across industries. However, healthcare presents a unique challenge. Here, the stakes couldn&#8217;t be higher, leading to a web of regulations and a natural conservatism when it comes to adopting new ideas. <\/span><\/p>\n<blockquote>\n<p style=\"text-align: left;\"><span style=\"font-weight: 400;\">This means some promising solutions are stuck in testing, delaying their real-world impact.<\/span><\/p>\n<\/blockquote>\n<p><span style=\"font-weight: 400;\">Despite this hurdle, these advancements offer a glimpse into the future of healthcare, hinting at where the market is headed. The companies that can thoughtfully and swiftly implement these solutions are the ones to big win.<\/span><\/p>        <\/div>\n      \n      \n    <\/div>\n  <\/div>\n<\/div>\n\n<div class=\"lazyblock-text-ZMcf0d wp-block-lazyblock-text\">\n  <div style=\"--size-margin-text-block: 70px;\" class=\"lazyblocks-text-wrapper wp-block-lazyblock-text\">\n    <div class=\"lazyblocks-text-columns-wrapper\">\n              <div class=\"lazyblocks-text-column\n                  lazyblocks-text-column-border\n                \" style=\"--size-padding-text-block: 70px;\">\n          <h3><span style=\"font-weight: 400;\">Who we are?<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">At NubiSoft, we specialize in working with the healthcare sector on dedicated solutions using AI. <\/span><\/p>\n<p><span style=\"font-weight: 400;\"><strong>We believe that properly designed solutions can make a real difference in the work of medical professionals.<\/strong> <\/span><\/p>\n<p style=\"text-align: left;\"><span style=\"font-weight: 400;\">We are ready to help your company join the AI revolution safely and sustainably. Contact us when you think you&#8217;re ready for it too.<\/span><\/p>        <\/div>\n      \n      \n    <\/div>\n  <\/div>\n<\/div>\n\n<div class=\"lazyblock-text-6bMUU wp-block-lazyblock-text\">\n  <div style=\"--size-margin-text-block: 70px;\" class=\"lazyblocks-text-wrapper wp-block-lazyblock-text\">\n    <div class=\"lazyblocks-text-columns-wrapper\">\n              <div class=\"lazyblocks-text-column\n                  lazyblocks-text-column-border\n                \" style=\"--size-padding-text-block: 70px;\">\n          <h3><span style=\"font-weight: 400;\">Sources:<\/span><\/h3>\n<ul>\n<li><a href=\"https:\/\/www.statista.com\/statistics\/1334826\/ai-in-healthcare-market-size-worldwide\/\">https:\/\/www.statista.com\/statistics\/1334826\/ai-in-healthcare-market-size-worldwide\/<\/a><\/li>\n<li><a href=\"https:\/\/radixweb.com\/blog\/ai-in-healthcare-statistics\">https:\/\/radixweb.com\/blog\/ai-in-healthcare-statistics<\/a><\/li>\n<li><a href=\"https:\/\/www.ama-assn.org\/practice-management\/sustainability\/do-you-spend-more-time-administrative-tasks-your-peers\">https:\/\/www.ama-assn.org\/practice-management\/sustainability\/do-you-spend-more-time-administrative-tasks-your-peers<\/a><\/li>\n<li><a href=\"https:\/\/www.statnews.com\/2023\/02\/01\/promises-pitfalls-chatgpt-assisted-medicine\/\">https:\/\/www.statnews.com\/2023\/02\/01\/promises-pitfalls-chatgpt-assisted-medicine\/<\/a><\/li>\n<li><a href=\"https:\/\/limbic.ai\/nhs-talking-therapies\">https:\/\/limbic.ai\/nhs-talking-therapies<\/a><\/li>\n<li><a href=\"https:\/\/transform.england.nhs.uk\/key-tools-and-info\/digital-playbooks\/workforce-digital-playbook\/using-an-ai-chatbot-to-streamline-mental-health-referrals\/\">https:\/\/transform.england.nhs.uk\/key-tools-and-info\/digital-playbooks\/workforce-digital-playbook\/using-an-ai-chatbot-to-streamline-mental-health-referrals\/<\/a><\/li>\n<li><a href=\"https:\/\/sites.research.google\/med-palm\/\">https:\/\/sites.research.google\/med-palm\/<\/a><\/li>\n<li><a href=\"https:\/\/mededu.jmir.org\/2023\/1\/e46885\">https:\/\/mededu.jmir.org\/2023\/1\/e46885<\/a><\/li>\n<li><a href=\"https:\/\/www.ncbi.nlm.nih.gov\/pmc\/articles\/PMC10463084\/\">https:\/\/www.ncbi.nlm.nih.gov\/pmc\/articles\/PMC10463084\/<\/a><\/li>\n<li><a href=\"https:\/\/lifesciences.n-side.com\/blog\/what-is-the-average-time-to-bring-a-drug-to-market-in-2022\">https:\/\/lifesciences.n-side.com\/blog\/what-is-the-average-time-to-bring-a-drug-to-market-in-2022<\/a><\/li>\n<li><a href=\"https:\/\/www.techuk.org\/resource\/aiweek2023-benevolentai-mon.html\">https:\/\/www.techuk.org\/resource\/aiweek2023-benevolentai-mon.html<\/a><\/li>\n<li><a href=\"https:\/\/www.prnewswire.com\/news-releases\/benevolentai-data-from-eli-lillys-cov-barrier-trial-shows-baricitinib-reduced-deaths-in-hospitalised-covid-19-patients-by-38-301265356.html\">https:\/\/www.prnewswire.com\/news-releases\/benevolentai-data-from-eli-lillys-cov-barrier-trial-shows-baricitinib-reduced-deaths-in-hospitalised-covid-19-patients-by-38-301265356.html<\/a><\/li>\n<li><a href=\"https:\/\/www.mobihealthnews.com\/news\/study-google-reveals-llm-helps-accurately-diagnosis-complex-cases\">https:\/\/www.mobihealthnews.com\/news\/study-google-reveals-llm-helps-accurately-diagnosis-complex-cases<\/a><\/li>\n<\/ul>        <\/div>\n      \n      \n    <\/div>\n  <\/div>\n<\/div>\n\n<div class=\"lazyblock-posts-2dpIF8 wp-block-lazyblock-posts\">\n\n\n<div style=\"--size-margin-posts-block: 70px;\" class=\"lazyblocks-posts-wrapper wp-block-lazyblock-posts\">\n                                <h3 class=\"lazyblocks-posts-header\">People who read this article also read:<\/h3>                     \n\n        <div class=\"lazyblocks-posts-container\">\n\n                            <div class=\"lazyblocks-posts-single-wrapper\">\n                                        <a class=\"lazyblocks-posts-image-wrapper\" style=\"--post-thumbail-url: url('https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/01\/Blog-templateTelemedicine-150x150.png');\" href=\"https:\/\/nubisoft.io\/blog\/telemedicine-in-elderly-care\/\">\n\n\n                                                    <img loading=\"lazy\" decoding=\"async\" width=\"580\" height=\"287\" src=\"https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/01\/Blog-templateTelemedicine-1024x507.png\" class=\"lazyblocks-posts-image wp-post-image\" alt=\"\" srcset=\"https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/01\/Blog-templateTelemedicine-1024x507.png 1024w, https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/01\/Blog-templateTelemedicine-300x149.png 300w, https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/01\/Blog-templateTelemedicine-768x380.png 768w, https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/01\/Blog-templateTelemedicine-1200x594.png 1200w, https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/01\/Blog-templateTelemedicine.png 1248w\" sizes=\"auto, (max-width: 580px) 100vw, 580px\" \/>                                            <\/a>\n\n                    <div class=\"lazyblocks-posts-info-wrapper\">\n                                                    <div class=\"lazyblocks-posts-category-wrapper\">\n                                <a href=\"https:\/\/nubisoft.io\/blog\/category\/healthcare\/\" rel=\"category tag\">Healthcare<\/a>                            <\/div>\n                        \n                        <span class=\"lazyblocks-posts-read-time\">1 min read<\/span>                    <\/div>\n\n\n                    <div class=\"lazyblocks-posts-title-wrapper\">\n                        <h2 class=\"lazyblocks-posts-title\"><a href=\"https:\/\/nubisoft.io\/blog\/telemedicine-in-elderly-care\/\">Telemedicine in Elderly Care<\/a><\/h2>\n                        <a class=\"lazyblocks-posts-read-more\" href=\"https:\/\/nubisoft.io\/blog\/telemedicine-in-elderly-care\/\">Read more<\/a>\n                    <\/div>\n                <\/div>\n                            <div class=\"lazyblocks-posts-single-wrapper\">\n                                        <a class=\"lazyblocks-posts-image-wrapper\" style=\"--post-thumbail-url: url('https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/01\/Blog-templateE-services-150x150.png');\" href=\"https:\/\/nubisoft.io\/blog\/e-services-in-the-polish-healthcare-sector\/\">\n\n\n                                                    <img loading=\"lazy\" decoding=\"async\" width=\"580\" height=\"287\" src=\"https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/01\/Blog-templateE-services-1024x507.png\" class=\"lazyblocks-posts-image wp-post-image\" alt=\"\" srcset=\"https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/01\/Blog-templateE-services-1024x507.png 1024w, https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/01\/Blog-templateE-services-300x149.png 300w, https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/01\/Blog-templateE-services-768x380.png 768w, https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/01\/Blog-templateE-services-1200x594.png 1200w, https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2024\/01\/Blog-templateE-services.png 1248w\" sizes=\"auto, (max-width: 580px) 100vw, 580px\" \/>                                            <\/a>\n\n                    <div class=\"lazyblocks-posts-info-wrapper\">\n                                                    <div class=\"lazyblocks-posts-category-wrapper\">\n                                <a href=\"https:\/\/nubisoft.io\/blog\/category\/healthcare\/\" rel=\"category tag\">Healthcare<\/a>                            <\/div>\n                        \n                        <span class=\"lazyblocks-posts-read-time\">8 min read<\/span>                    <\/div>\n\n\n                    <div class=\"lazyblocks-posts-title-wrapper\">\n                        <h2 class=\"lazyblocks-posts-title\"><a href=\"https:\/\/nubisoft.io\/blog\/e-services-in-the-polish-healthcare-sector\/\">E-Services in the Polish Healthcare Sector<\/a><\/h2>\n                        <a class=\"lazyblocks-posts-read-more\" href=\"https:\/\/nubisoft.io\/blog\/e-services-in-the-polish-healthcare-sector\/\">Read more<\/a>\n                    <\/div>\n                <\/div>\n                            <div class=\"lazyblocks-posts-single-wrapper\">\n                                        <a class=\"lazyblocks-posts-image-wrapper\" style=\"--post-thumbail-url: url('https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2023\/03\/Frame-3-2-150x150.png');\" href=\"https:\/\/nubisoft.io\/blog\/e-prescription-how-does-it-actually-work\/\">\n\n\n                                                    <img loading=\"lazy\" decoding=\"async\" width=\"580\" height=\"435\" src=\"https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2023\/03\/Frame-3-2-1024x768.png\" class=\"lazyblocks-posts-image wp-post-image\" alt=\"\" srcset=\"https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2023\/03\/Frame-3-2-1024x768.png 1024w, https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2023\/03\/Frame-3-2-300x225.png 300w, https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2023\/03\/Frame-3-2-768x576.png 768w, https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2023\/03\/Frame-3-2-1200x900.png 1200w, https:\/\/nubisoft.io\/blog\/wp-content\/uploads\/2023\/03\/Frame-3-2.png 1280w\" sizes=\"auto, (max-width: 580px) 100vw, 580px\" \/>                                            <\/a>\n\n                    <div class=\"lazyblocks-posts-info-wrapper\">\n                                                    <div class=\"lazyblocks-posts-category-wrapper\">\n                                <a href=\"https:\/\/nubisoft.io\/blog\/category\/healthcare\/\" rel=\"category tag\">Healthcare<\/a>                            <\/div>\n                        \n                        <span class=\"lazyblocks-posts-read-time\">5 min read<\/span>                    <\/div>\n\n\n                    <div class=\"lazyblocks-posts-title-wrapper\">\n                        <h2 class=\"lazyblocks-posts-title\"><a href=\"https:\/\/nubisoft.io\/blog\/e-prescription-how-does-it-actually-work\/\">E-prescription &#8211; how does it actually work?<\/a><\/h2>\n                        <a class=\"lazyblocks-posts-read-more\" href=\"https:\/\/nubisoft.io\/blog\/e-prescription-how-does-it-actually-work\/\">Read more<\/a>\n                    <\/div>\n                <\/div>\n                    <\/div>\n    <\/div>\n<\/div>","protected":false},"excerpt":{"rendered":"<p>AI has limits, and we need to be clear on those. This article dives into real-world examples of how large language models (LLMs) are already helping medical professionals.<\/p>\n","protected":false},"author":13,"featured_media":2318,"comment_status":"open","ping_status":"open","sticky":false,"template":"post-2024.php","format":"standard","meta":{"_case_study_excerpt":"","footnotes":""},"categories":[221],"tags":[488,493,496,491],"class_list":["post-2324","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-healthcare","tag-ai","tag-case-studies","tag-hospitals","tag-llm"],"_links":{"self":[{"href":"https:\/\/nubisoft.io\/blog\/wp-json\/wp\/v2\/posts\/2324","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/nubisoft.io\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/nubisoft.io\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/nubisoft.io\/blog\/wp-json\/wp\/v2\/users\/13"}],"replies":[{"embeddable":true,"href":"https:\/\/nubisoft.io\/blog\/wp-json\/wp\/v2\/comments?post=2324"}],"version-history":[{"count":9,"href":"https:\/\/nubisoft.io\/blog\/wp-json\/wp\/v2\/posts\/2324\/revisions"}],"predecessor-version":[{"id":2755,"href":"https:\/\/nubisoft.io\/blog\/wp-json\/wp\/v2\/posts\/2324\/revisions\/2755"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/nubisoft.io\/blog\/wp-json\/wp\/v2\/media\/2318"}],"wp:attachment":[{"href":"https:\/\/nubisoft.io\/blog\/wp-json\/wp\/v2\/media?parent=2324"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/nubisoft.io\/blog\/wp-json\/wp\/v2\/categories?post=2324"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/nubisoft.io\/blog\/wp-json\/wp\/v2\/tags?post=2324"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}