[{"data":1,"prerenderedAt":403},["ShallowReactive",2],{"slug-why-and-how-to-regulate-chatgpt-like-large-language-models-in-healthcare":3},{"post":4,"relatedPosts":175,"relatedBooks":307},{"id":5,"date":6,"date_gmt":7,"guid":8,"modified":10,"modified_gmt":11,"slug":12,"status":13,"type":14,"link":15,"title":16,"content":18,"excerpt":21,"author":23,"featured_media":24,"comment_status":25,"ping_status":25,"sticky":26,"template":27,"format":28,"meta":29,"categories":30,"tags":33,"project_category":41,"contact_email_category":43,"yst_prominent_words":44,"class_list":54,"better_featured_image":71,"acf":102,"yoast_meta":119,"_links":122},51657,"2023-07-19T10:19:34","2023-07-19T08:19:34",{"rendered":9},"https:\u002F\u002Fmedicalfuturist.com\u002F?post_id=51657&_wpnonce=118047110b&status=auto-draft&type=post","2023-07-19T10:19:36","2023-07-19T08:19:36","why-and-how-to-regulate-chatgpt-like-large-language-models-in-healthcare","publish","post","https:\u002F\u002Fmedicalfuturist.com\u002Fwhy-and-how-to-regulate-chatgpt-like-large-language-models-in-healthcare",{"rendered":17},"Why And How To Regulate ChatGPT-Like Large Language Models In Healthcare?",{"rendered":19,"protected":20},"\n\u003Cp>Large Language Models (LLMs), such as ChatGPT or Bard, hold immense promise but also pose significant challenges for healthcare and medicine. To unlock their enormous benefits, we must ensure their safe application in an environment where lives are at stake.  In other words, our task is to establish a robust, ethical framework for these generative AI models – without making the boundaries so tight that it kills innovation.\u003C\u002Fp>\n\n\n\n\u003Cp>Our latest paper published in Nature’s npj Digital Medicine, “\u003Ca href=\"https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41746-023-00873-0\">The Imperative for Regulatory Oversight of Large Language Models (or Generative AI)  in Healthcare\u003C\u002Fa>” published with Dr Eric Topol analysed the challenges and the possibilities of LLM regulations to make them accessible for healthcare use. \u003C\u002Fp>\n\n\n\n\u003Cp>\u003Cstrong>We need brand new kinds of regulations\u003C\u002Fstrong>\u003C\u002Fp>\n\n\n\n\u003Cp>Regulators must cover myriad scenarios and perspectives since LLMs significantly differ from existing and regulated AI algorithms and deep learning methods.\u003C\u002Fp>\n\n\n\n\u003Cp> A range of distinct characteristics sets them apart, including their: \u003C\u002Fp>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>\u003Cstrong>scale and complexity\u003C\u002Fstrong> – LLMs utilize billions of parameters, resulting in unprecedented complexity. Tokenization, their basic “processing” method is currently not covered by healthcare regulators\u003C\u002Fli>\n\n\n\n\u003Cli>\u003Cstrong>broad applicability\u003C\u002Fstrong> – LLMs have unprecedented versatility compared to specialized deep learning models. As they can be used in various sectors, from finance to healthcare, one-size-fits-all regulations will not suffice\u003C\u002Fli>\n\n\n\n\u003Cli>\u003Cstrong>real-time adaptation \u003C\u002Fstrong>– LLMs can adapt their responses based on user inputs and the evolving context of the “conversation”, sharing similarities with adaptive algorithms\u003C\u002Fli>\n\n\n\n\u003Cli>\u003Cstrong>data privacy and security \u003C\u002Fstrong>– a multifaceted problem-complex, which spans from how to address the ownership issues of the training data – with only a slight exaggeration: the content of the internet and all libraries – to what protects user data that was the input of our prompts and to who owns the results of the human-LLM cooperation\u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Cp>\u003Cstrong>The use of LLMs in medical practices brings unprecedented challenges\u003C\u002Fstrong>\u003C\u002Fp>\n\n\n\n\u003Cp>Despite their potential, LLMs also amplify risks and challenges in healthcare. To date, no LLM has had pre-training with the corpus of medical information or with millions of patient records, images, lab data, and office visits or bedside conversations – although according to news reports, \u003Ca href=\"https:\u002F\u002Fmedicalfuturist.com\u002Fmedpalm-new-ai-medical-chatbots-will-soon-be-better-than-waiting-for-a-doctor\u002F\">Google’s MedPaLM\u003C\u002Fa> could be such an LLM, but it is \u003Ca href=\"https:\u002F\u002Fsites.research.google\u002Fmed-palm\u002F\">inaccessible for laymen\u003C\u002Fa>.\u003C\u002Fp>\n\n\n\n\u003Cp>Details about the training process for GPT-4, the most advanced publicly available LLM at the moment, remain undisclosed. Nevertheless, these models have immense potential in medicine, with use cases ranging from clinical documentation to providing personalized health plans.\u003C\u002Fp>\n\n\n\n\u003Cfigure class=\"wp-block-image size-large\">\u003Cimg loading=\"lazy\" decoding=\"async\" width=\"768\" height=\"432\" src=\"https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F07\u002FFigure-1-768x432.png\" alt=\"large language models in healthcare\" class=\"wp-image-51661\" srcset=\"https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F07\u002FFigure-1-768x432.png 768w, https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F07\u002FFigure-1-370x208.png 370w, https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F07\u002FFigure-1-1536x864.png 1536w, https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F07\u002FFigure-1-2048x1152.png 2048w\" sizes=\"auto, (max-width: 768px) 100vw, 768px\" \u002F>\u003C\u002Ffigure>\n\n\n\n\u003Cp>However, current models can also “hallucinate” results or provide misleading information, which, in medical settings could lead to incorrect diagnoses or treatment recommendations. Biases in training data can affect clinical decision-making, patient outcomes, and healthcare equity. Patients may also come to medical consultations with information from chatbots, complicating the physician-patient dynamic.\u003C\u002Fp>\n\n\n\n\u003Cp>\u003Cstrong>We need to regulate the future, not just the present\u003C\u002Fstrong>\u003C\u002Fp>\n\n\n\n\u003Cp>Typical (publicly available) advanced LLMs work with text inputs at the moment.  But we also know that these models can already analyze images, and it can be expected that they will soon deal with uploaded documents, research papers, hand-written notes, sound, and video. This means future regulations have to cover these future use cases \u003C\u002Fp>\n\n\n\n\u003Cp>Companies with approved devices that decide to implement LLMs into their services face an additional challenge. Namely, how will the FDA regulate an AI-based medical technology recently infused with LLM if the technology was already approved for medical uses?\u003C\u002Fp>\n\n\n\n\u003Cp>\u003Cstrong>Our practical expectations\u003C\u002Fstrong>\u003C\u002Fp>\n\n\n\n\u003Cp>Here is our summary of what I expect regulators will do in this regard. \u003C\u002Fp>\n\n\n\n\u003Cul class=\"wp-block-list\">\n\u003Cli>Create a new regulatory category for LLMs\u003C\u002Fli>\n\n\n\n\u003Cli>Provide regulatory guidance for companies and healthcare organizations about how they can deploy LLMs into their existing products and services\u003C\u002Fli>\n\n\n\n\u003Cli>Create a regulatory framework that not only covers text-based interactions but possible future iterations such as analyzing sound or video\u003C\u002Fli>\n\n\n\n\u003Cli>Provide a framework for making a distinction between LLMs specifically trained on medical data and LLMs trained for non-medical purposes.\u003C\u002Fli>\n\n\n\n\u003Cli>Similar to the FDA’s Digital Health Pre-Cert Program, regulate companies developing LLMs instead of regulating every single LLM iteration\u003C\u002Fli>\n\u003C\u002Ful>\n\n\n\n\u003Cp>By taking a proactive approach to regulation, it is possible to harness the potential, minimize potential harm and preserve the trust of patients and healthcare providers alike. It’s also intriguing to consider that LLMs could potentially become the first category of AI-based medical technologies regulated through the implementation of patient-focused design. \u003C\u002Fp>\n",false,{"rendered":22,"protected":20},"\u003Cp>Large Language Models (LLMs), such as ChatGPT, represent a new frontier in the use of AI in healthcare,  there’s a necessity for an entirely new kind of regulatory framework to make them safe for medical use.\u003C\u002Fp>\n",6,51665,"closed",true,"","standard",{"_acf_changed":20,"footnotes":27},[31,32],7079,504,[34,35,36,37,38,39,40],134,7671,7673,7741,7745,7809,7845,[42],950,[],[45,46,47,48,49,50,51,52,53],1831,2313,5299,5537,1683,1693,1721,1723,1789,[55,14,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70],"post-51657","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-tmf","category-artificial-intelligence","tag-ai","tag-ai-in-healthcare","tag-ai-in-medicine","tag-large-language-models","tag-medpalm","tag-chatgpt-in-healthcare","tag-bard","project_category-medical-professionals",{"id":24,"alt_text":27,"caption":27,"description":27,"media_type":72,"media_details":73,"post":5,"source_url":101},"image",{"width":74,"height":75,"file":76,"filesize":77,"sizes":78,"image_meta":98},1080,607,"2023\u002F07\u002Ftmf_article_378_720_regulate_large_language_models.png",1019402,{"medium":79,"large":86,"thumbnail":92,"medium_large":97},{"file":80,"width":81,"height":82,"mime-type":83,"filesize":84,"source_url":85},"tmf_article_378_720_regulate_large_language_models-370x208.png",370,208,"image\u002Fpng",90315,"https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F07\u002Ftmf_article_378_720_regulate_large_language_models-370x208.png",{"file":87,"width":88,"height":89,"mime-type":83,"filesize":90,"source_url":91},"tmf_article_378_720_regulate_large_language_models-768x432.png",768,432,398828,"https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F07\u002Ftmf_article_378_720_regulate_large_language_models-768x432.png",{"file":93,"width":94,"height":94,"mime-type":83,"filesize":95,"source_url":96},"tmf_article_378_720_regulate_large_language_models-150x150.png",150,28404,"https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F07\u002Ftmf_article_378_720_regulate_large_language_models-150x150.png",{"file":87,"width":88,"height":89,"mime-type":83,"filesize":90,"source_url":91},{"aperture":99,"credit":27,"camera":27,"caption":27,"created_timestamp":99,"copyright":27,"focal_length":99,"iso":99,"shutter_speed":99,"title":27,"orientation":99,"keywords":100},"0",[],"https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F07\u002Ftmf_article_378_720_regulate_large_language_models.png",{"cta_type":103,"cta_color":27,"related_books":104,"related_posts_footer":108,"related_posts":20,"subtitle":27,"key_takeaways":112},"subscribe",[105,106,107],24762,24761,24764,[109,110,111],51555,51531,51353,[113,115,117],{"title":114},"\u003Cp>Large Language Models, like ChatGPT, Bard or MedPaLM bring about unprecedented potential and challenges in healthcare and require a unique regulatory framework\u003C\u002Fp>\n",{"title":116},"\u003Cp>The capabilities of LLMs are expanding beyond just text-based interactions, to analyzing images, documents, hand-written notes, sound, and video. Thus we need to regulate the future, not just the present\u003C\u002Fp>\n",{"title":118},"\u003Cp>The frameworks will need to make a distinction between LLMs specifically trained on medical data and LLMs trained for non-medical purposes\u003C\u002Fp>\n",{"yoast_wpseo_title":120,"yoast_wpseo_metadesc":121,"yoast_wpseo_canonical":15},"How to regulate large language models in healthcare?","To harness the enormous potential of large language models in healthcare, we need sound regulations to guarantee their safe use. This is a unique challenge.",{"self":123,"collection":129,"about":132,"author":135,"replies":138,"version-history":141,"predecessor-version":145,"wp:featuredmedia":149,"wp:attachment":152,"wp:term":155,"curies":171},[124],{"href":125,"targetHints":126},"https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Fposts\u002F51657",{"allow":127},[128],"GET",[130],{"href":131},"https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Fposts",[133],{"href":134},"https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Ftypes\u002Fpost",[136],{"embeddable":26,"href":137},"https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Fusers\u002F6",[139],{"embeddable":26,"href":140},"https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Fcomments?post=51657",[142],{"count":143,"href":144},5,"https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Fposts\u002F51657\u002Frevisions",[146],{"id":147,"href":148},51671,"https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Fposts\u002F51657\u002Frevisions\u002F51671",[150],{"embeddable":26,"href":151},"https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Fmedia\u002F51665",[153],{"href":154},"https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Fmedia?parent=51657",[156,159,162,165,168],{"taxonomy":157,"embeddable":26,"href":158},"category","https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Fcategories?post=51657",{"taxonomy":160,"embeddable":26,"href":161},"post_tag","https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Ftags?post=51657",{"taxonomy":163,"embeddable":26,"href":164},"project_category","https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Fproject_category?post=51657",{"taxonomy":166,"embeddable":26,"href":167},"contact_email_category","https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Fcontact_email_category?post=51657",{"taxonomy":169,"embeddable":26,"href":170},"yst_prominent_words","https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Fyst_prominent_words?post=51657",[172],{"name":173,"href":174,"templated":26},"wp","https:\u002F\u002Fapi.w.org\u002F{rel}",[176],{"id":111,"date":177,"date_gmt":178,"guid":179,"modified":181,"modified_gmt":182,"slug":183,"status":13,"type":14,"link":184,"title":185,"content":187,"excerpt":189,"author":23,"featured_media":191,"comment_status":25,"ping_status":25,"sticky":26,"template":27,"format":28,"meta":192,"categories":193,"tags":194,"project_category":200,"contact_email_category":201,"yst_prominent_words":202,"class_list":204,"better_featured_image":211,"acf":249,"yoast_meta":262,"_links":265},"2025-10-06T09:29:55","2025-10-06T07:29:55",{"rendered":180},"https:\u002F\u002Fmedicalfuturist.com\u002F?post_id=51353&_wpnonce=f89eb92837&status=auto-draft&type=post","2025-10-15T06:13:58","2025-10-15T04:13:58","prompt-engineering-11-tips-to-craft-great-chatgpt-prompts","https:\u002F\u002Fmedicalfuturist.com\u002Fprompt-engineering-11-tips-to-craft-great-chatgpt-prompts",{"rendered":186},"Prompt Engineering For Healthcare: 11 Tips To Craft Great ChatGPT Prompts",{"rendered":188,"protected":20},"\n\u003Cp>The article summarising basic tips for prompt engineering was one of the most popular pieces we published last year. However, as almost a year has passed, widely used generative AI models have become significantly more advanced, with capabilities we could only dream about back then. Thus it was time to give this useful piece of content a thorough update. So here is the 2024 iteration, hope you will (continue to) find it useful!\u003C\u002Fp>\n\n\n\n\u003Cp>Have you experienced the frustration of repeatedly asking Chat GPT or Gemini or Microsoft CoPilot or MidJourney (etc.) a relatively simple thing, and \u003Ca rel=\"noreferrer noopener\" href=\"https:\u002F\u002Fmedicalfuturist.com\u002Fhere-is-how-you-get-friendly-with-a-i-before-it-gets-to-the-office\u002F\" target=\"_blank\">not getting the results you were looking for\u003C\u002Fa>? You are not alone! As \u003Ca rel=\"noreferrer noopener\" href=\"https:\u002F\u002Fmedicalfuturist.com\u002Fgenerative-adversarial-networks-how-can-ai-learn-so-much-while-its-iq-remains-zero\" target=\"_blank\">Generative Artificial Intelligence\u003C\u002Fa> (AI), particularly language models like ChatGPT, have started to carve out a niche and \u003Ca rel=\"noreferrer noopener\" href=\"https:\u002F\u002Fmedicalfuturist.com\u002Fseven-reasons-to-be-optimistic-about-the-future-of-ai-in-medicine\" target=\"_blank\">offer immense potential\u003C\u002Fa>, they come with their unique challenges, one of which is how to work efficiently with them.\u003C\u002Fp>\n\n\n\n\u003Cblockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n\u003Cp>Prompt engineering is the art and science of crafting effective inputs to guide an AI model towards generating the desired output. \u003C\u002Fp>\n\u003C\u002Fblockquote>\n\n\n\n\u003Cp>In less fancy terms: prompts are our inputs, and the better we get at giving great prompts, the better the output will be. Although plugins and AI chatbots can help you, they still won’t do your job, so don’t rely solely on them. \u003C\u002Fp>\n\n\n\n\u003Cp>Using ChatGPT and similar tools indeed has its risks: it may generate its answer from hallucinated sources or may fail to provide sources altogether, and in most cases, the onus falls upon us, users, to verify the information – just to name a few. Despite its limitations, it is quickly becoming a \u003Ca rel=\"noreferrer noopener\" href=\"https:\u002F\u002Fmedicalfuturist.com\u002Ftop-ai-algorithms-healthcare\" target=\"_blank\">must-have ally\u003C\u002Fa>, which brings us to the crux of the matter – how do we ask the right questions?\u003C\u002Fp>\n\n\n\n\u003Cp>Understanding the art of ‘prompt engineering’ is becoming increasingly critical. It’s so fundamental that I would make sure it is momentarily becoming part of the medical curriculum and I recommend every medical professional to learn it. It is also part of our AI course. But while formal education can be an excellent starting point, there are practical tips and tricks to improve your prompt engineering skills in the meantime.\u003C\u002Fp>\n\n\n\n\u003Cp>Like any skill, mastery of prompt engineering comes with practice. To save you some of the time and trial-and-error, here is what I’ve learned during the past months when I spent hundreds of hours refining my own approach. I now use these daily and I’m keen to share them with you.\u003C\u002Fp>\n\n\n\n\u003Cp>Whether you’re a seasoned user of AI or just starting to dip your toes into this brave new world, these guidelines should make your journey smoother and more rewarding.\u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\">0. There is no ultimate tool – find the algorithm for the task\u003C\u002Fh2>\n\n\n\n\u003Cp>We are living in a fascinating era: even as average users we have access to a fast-growing number of generative AI solutions, gone are the days when ChatGPT was the only sheriff in town. I truly enjoy experimenting with formerly unknown (for me) algorithms and seeing their capabilities. \u003C\u002Fp>\n\n\n\n\u003Cp>Living in the EU is somewhat disadvantaged in this regard: while we still \u003Ca rel=\"noreferrer noopener\" href=\"https:\u002F\u002Fwww.anthropic.com\u002Fclaude-ai-locations\" target=\"_blank\">can’t access Claude\u003C\u002Fa>, which is quite hyped these days, we can play with others like \u003Ca rel=\"noreferrer noopener\" href=\"https:\u002F\u002Fgemini.google.com\u002F\" target=\"_blank\">Gemini\u003C\u002Fa> (which is MUCH better than good ole’ Bard was), \u003Ca rel=\"noreferrer noopener\" href=\"https:\u002F\u002Fwww.perplexity.ai\u002F\" target=\"_blank\">Perplexity\u003C\u002Fa> or \u003Ca rel=\"noreferrer noopener\" href=\"https:\u002F\u002Finflection.ai\u002Fthe-new-inflection\" target=\"_blank\">Inflection.pi\u003C\u002Fa>. And these are just “general” tools, there are literally hundreds of specialised algorithms from music creation to coding, image generation or creating meeting memos.\u003C\u002Fp>\n\n\n\n\u003Cp>The more I use these algorithms, the more I realize there is no ultimate tool, all of them are very good in certain things and weak in others. Ideally, you can use different models for different tasks. For example, ChatGPT is terrible at suggesting great titles, while Gemini is completely useless in making sense of a basic chart. But they are quite adept the other way around, so I rather use ChatGPT to support me in data analytics and Gemini to help me come up with catchy titles. \u003C\u002Fp>\n\n\n\n\u003Cp>As the generative AI landscape becomes more and more varied, experimenting with several tools and learning their strengths will save you tons of frustration. \u003C\u002Fp>\n\n\n\n\u003Ch2 class=\"wp-block-heading\">Now let’s see what to do once you have the algorithm of your choice! \u003C\u002Fh2>\n\n\n\n\u003Ch3 class=\"wp-block-heading\">1. Be as specific as possible\u003C\u002Fh3>\n\n\n\n\u003Cp>The more specific your prompt, the more accurate and focused the response is likely to be.\u003C\u002Fp>\n\n\n\n\u003Cblockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n\u003Cp>\u003Cstrong>Example:\u003C\u002Fstrong>\u003C\u002Fp>\n\u003Ccite>\u003Cstrong>Less specific\u003C\u002Fstrong>: “Tell me about heart disease.”\u003Cbr>\u003Cstrong>More specific\u003C\u002Fstrong>: “What are the most common risk factors for coronary artery disease?”\u003C\u002Fcite>\u003C\u002Fblockquote>\n\n\n\n\u003Cfigure class=\"wp-block-image size-full\">\u003Cimg loading=\"lazy\" decoding=\"async\" width=\"988\" height=\"500\" src=\"https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F06\u002FChatGPT1-side.jpg\" alt=\"\" class=\"wp-image-51359\" srcset=\"https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F06\u002FChatGPT1-side.jpg 988w, https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F06\u002FChatGPT1-side-768x389.jpg 768w\" sizes=\"auto, (max-width: 988px) 100vw, 988px\" \u002F>\u003C\u002Ffigure>\n\n\n\n\u003Ch3 class=\"wp-block-heading\">2. Describe your goal\u003C\u002Fh3>\n\n\n\n\u003Cp>Describe exactly what kind of output you are looking for. It also helps highlight what’s important for you about the output.\u003C\u002Fp>\n\n\n\n\u003Cblockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n\u003Cp>\u003Cstrong>Example:\u003C\u002Fstrong>\u003C\u002Fp>\n\u003Ccite>“I’d like to get a short list of 5 ideas for a Youtube video on the future of healthcare AI.”\u003C\u002Fcite>\u003C\u002Fblockquote>\n\n\n\n\u003Cp>When your question includes the goal (creating YouTube content), your answers will likely be more useful than if you had just asked the algorithms about the future of healthcare AI.  \u003C\u002Fp>\n\n\n\n\u003Ch3 class=\"wp-block-heading\">3. Describe your setting and provide the context\u003C\u002Fh3>\n\n\n\n\u003Cp>Consider the discussion you are having with ChatGPT as a discussion you would have with a person you just met. They might be able to answer your questions and address your challenges, but have zero background knowledge, so you need to provide your context. The better you describe your goal, your target audience, and the tone you would like to use, the more fitting the output will be.\u003C\u002Fp>\n\n\n\n\u003Cblockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n\u003Cp>\u003Cstrong>Example\u003C\u002Fstrong>:\u003C\u002Fp>\n\u003Ccite>“I’m writing an article about tips and tricks for ChatGPT prompt engineering for people working in healthcare. Can you please list a few of those tips and tricks with some specific prompt examples? My audience consists of medical professionals who are not AI experts. Please use a friendly, approachable tone.”\u003C\u002Fcite>\u003C\u002Fblockquote>\n\n\n\n\u003Cp>\u003C\u002Fp>\n\n\n\n\u003Ch3 class=\"wp-block-heading\">4. Experiment with different prompt styles\u003C\u002Fh3>\n\n\n\n\u003Cp>The style of your prompt can significantly impact the answer. Try different formats like asking to generate a list, provide a summary, or explain like I’m five (ELI5), etc.\u003C\u002Fp>\n\n\n\n\u003Cblockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n\u003Cp>\u003Cstrong>Example:\u003C\u002Fstrong>\u003C\u002Fp>\n\u003Ccite>\u003Cstrong>Direct Question\u003C\u002Fstrong>: “What are the symptoms of COVID-19?”\u003Cbr>\u003Cstrong>Request for List\u003C\u002Fstrong>: “List all the potential symptoms of COVID-19.”\u003Cbr>\u003Cstrong>Request for Summary\u003C\u002Fstrong>: “Summarize the key symptoms and progression of COVID-19.”\u003Cbr>\u003Cstrong>ELI5:\u003C\u002Fstrong> “Explain the symptoms of COVID-19 like I’m five.”\u003C\u002Fcite>\u003C\u002Fblockquote>\n\n\n\n\u003Cfigure class=\"wp-block-image size-full\">\u003Cimg loading=\"lazy\" decoding=\"async\" width=\"1483\" height=\"500\" src=\"https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F06\u002FChatGPT3-side.jpg\" alt=\"\" class=\"wp-image-51357\" srcset=\"https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F06\u002FChatGPT3-side.jpg 1483w, https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F06\u002FChatGPT3-side-768x259.jpg 768w\" sizes=\"auto, (max-width: 1483px) 100vw, 1483px\" \u002F>\u003C\u002Ffigure>\n\n\n\n\u003Cp>\u003C\u002Fp>\n\n\n\n\u003Ch3 class=\"wp-block-heading\">5. Ask it to play roles\u003C\u002Fh3>\n\n\n\n\u003Cp>This way, you can get the desired process of getting the information or input you were looking for.\u003C\u002Fp>\n\n\n\n\u003Cblockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n\u003Cp>\u003Cstrong>Example\u003C\u002Fstrong>:  \u003C\u002Fp>\n\u003Ccite>“Act as a Data Scientist and explain Prompt Engineering to a physician.”\u003Cbr>“Act as my nutritionist and give me tips about a balanced Mediterranean diet.”\u003C\u002Fcite>\u003C\u002Fblockquote>\n\n\n\n\u003Cp>\u003C\u002Fp>\n\n\n\n\u003Ch3 class=\"wp-block-heading\">6. Iterate and refine to delve deeper and\u002For get better answers\u003C\u002Fh3>\n\n\n\n\u003Cp>Unless you are simply amazing at prompt engineering, you rarely get the best response you were looking for after the first prompt. So feel free to refine your question. You are also encouraged to ask it to modify the output based on its previous response. If you’re not getting the output you want, try guiding the model with continuation prompts. This can be more effective than trying to get the perfect response in a single prompt. You can also make your prompt a conversation: ask the AI to think step by step or to debate the pros and cons before settling on an answer.\u003C\u002Fp>\n\n\n\n\u003Cp>You: “What causes diabetes?”\u003C\u002Fp>\n\n\n\n\u003Cp>AI: “Diabetes is primarily caused by the body’s inability to produce or effectively use insulin…”\u003C\u002Fp>\n\n\n\n\u003Cp>You (refining the question): “What factors lead to the body’s inability to produce insulin?”\u003C\u002Fp>\n\n\n\n\u003Cp>AI: “The body’s inability to produce insulin can be caused by autoimmune destruction of the beta cells in the pancreas, as in Type 1 diabetes…”\u003C\u002Fp>\n\n\n\n\u003Cp>If you are moving into unknown territory and looking for information in a field you are not familiar with, you can use feedback loops to get into the details. While ChatGPT at first tends to provide very generic answers, you can always use its output as input for subsequent prompts. \u003C\u002Fp>\n\n\n\n\u003Cblockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n\u003Cp>\u003Cstrong>Example: \u003C\u002Fstrong>\u003C\u002Fp>\n\u003Ccite>\u003Cstrong>You\u003C\u002Fstrong>: “What are the common symptoms of Lyme disease?”\u003Cbr>\u003Cstrong>AI:\u003C\u002Fstrong> “Common symptoms of Lyme disease can include fever, fatigue, headache, muscle and joint aches, and swollen lymph nodes.”\u003Cbr>\u003Cstrong>You:\u003C\u002Fstrong> “What are the treatment options when these symptoms are confirmed to be due to Lyme disease?”\u003Cbr>\u003Cstrong>AI:\u003C\u002Fstrong> “Lyme disease is typically treated with antibiotics…”\u003C\u002Fcite>\u003C\u002Fblockquote>\n\n\n\n\u003Cfigure class=\"wp-block-image size-large\">\u003Cimg loading=\"lazy\" decoding=\"async\" width=\"768\" height=\"432\" src=\"https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F06\u002Ftmf_article_365-768x432.png\" alt=\"AI medical coding doctor, hospital, clinic, man, laptop, computer\" class=\"wp-image-51069\" srcset=\"https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F06\u002Ftmf_article_365-768x432.png 768w, https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F06\u002Ftmf_article_365-370x208.png 370w, https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F06\u002Ftmf_article_365-1536x864.png 1536w, https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F06\u002Ftmf_article_365-2048x1152.png 2048w, https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F06\u002Ftmf_article_365.png 1920w\" sizes=\"auto, (max-width: 768px) 100vw, 768px\" \u002F>\u003C\u002Ffigure>\n\n\n\n\u003Cp>\u003C\u002Fp>\n\n\n\n\u003Ch3 class=\"wp-block-heading\">7. Use your previous threads \u003C\u002Fh3>\n\n\n\n\u003Cp>You can always go back to a specific discussion by clicking on the right thread in the left column. This way, you don’t have to start all over again but can just continue a discussion you already had with the AI.\u003C\u002Fp>\n\n\n\n\u003Cp>Some models are better at understanding the context in long threads, others are less capable. According to my experience, of the algorithms I use, ChatGPT is the one that performs best in building on previously used information. However, now and then it becomes “confused” when I feel like we started moving in circles. This is when I start a new thread and give it a fresh start. \u003C\u002Fp>\n\n\n\n\u003Ch3 class=\"wp-block-heading\">7\u002FB Give your important threads obvious names\u003C\u002Fh3>\n\n\n\n\u003Cp>If you use generative AI often, it is worth making sure you (re)name important, frequently used threads. This lets you quickly find them, and you don’t need to click through dozens of discussions for the one you are looking for. \u003C\u002Fp>\n\n\n\n\u003Ch3 class=\"wp-block-heading\">8. Ask open-ended questions\u003C\u002Fh3>\n\n\n\n\u003Cp>Open-ended questions often yield more comprehensive responses.\u003C\u002Fp>\n\n\n\n\u003Cblockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n\u003Cp>\u003Cstrong>Example:\u003C\u002Fstrong>\u003C\u002Fp>\n\u003Ccite>\u003Cstrong>Closed question\u003C\u002Fstrong>: “Is exercise important for patients with osteoporosis?”\u003Cbr>\u003Cstrong>Open question\u003C\u002Fstrong>: “How does regular physical activity benefit patients with osteoporosis?”\u003C\u002Fcite>\u003C\u002Fblockquote>\n\n\n\n\u003Cp>\u003C\u002Fp>\n\n\n\n\u003Ch3 class=\"wp-block-heading\">9. Request Examples\u003C\u002Fh3>\n\n\n\n\u003Cp>If there is something you are not satisfied with or don’t understand based on its response, first of all, tell the AI model that you don’t understand the answer and ask it to provide an example.\u003C\u002Fp>\n\n\n\n\u003Cp>\u003C\u002Fp>\n\n\n\n\u003Ch3 class=\"wp-block-heading\">10. Use time wisely\u003C\u002Fh3>\n\n\n\n\u003Cp>If you’re asking about a process or timeline, specify that in your prompt.\u003C\u002Fp>\n\n\n\n\u003Cblockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n\u003Cp>\u003Cstrong>Example\u003C\u002Fstrong>:\u003C\u002Fp>\n\u003Ccite>\u003Cstrong>Without time reference\u003C\u002Fstrong>: “Describe the healing process after knee surgery.”\u003Cbr>\u003Cstrong>With time reference\u003C\u002Fstrong>: “What can a patient typically expect during the first six weeks of healing after knee surgery?”\u003C\u002Fcite>\u003C\u002Fblockquote>\n\n\n\n\u003Cfigure class=\"wp-block-image size-large\">\u003Cimg loading=\"lazy\" decoding=\"async\" width=\"768\" height=\"432\" src=\"https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F06\u002FChatGPT-768x432.png\" alt=\"ChatGPT prompt\" class=\"wp-image-51361\" srcset=\"https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F06\u002FChatGPT-768x432.png 768w, https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F06\u002FChatGPT-370x208.png 370w, https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F06\u002FChatGPT-1536x864.png 1536w, https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F06\u002FChatGPT.png 1920w\" sizes=\"auto, (max-width: 768px) 100vw, 768px\" \u002F>\u003C\u002Ffigure>\n\n\n\n\u003Cp>\u003C\u002Fp>\n\n\n\n\u003Ch3 class=\"wp-block-heading\">\u003Cstrong>+1. Set realistic expectations\u003C\u002Fstrong>\u003C\u002Fh3>\n\n\n\n\u003Cp>Although GPT-4, Microsoft CoPilot or Gemini are powerful tools, they all have limitations. \u003C\u002Fp>\n\n\n\n\u003Cp>Some of these limitations are quite common: AI models hallucinate. \u003C\u002Fp>\n\n\n\n\u003Cfigure class=\"wp-block-image size-full\">\u003Cimg loading=\"lazy\" decoding=\"async\" width=\"1089\" height=\"933\" src=\"https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F06\u002FHallucination-Rate-of-Various-AI-Models.png\" alt=\"Hallucination rates of AI models\" class=\"wp-image-55295\" srcset=\"https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F06\u002FHallucination-Rate-of-Various-AI-Models.png 1089w, https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F06\u002FHallucination-Rate-of-Various-AI-Models-768x658.png 768w\" sizes=\"auto, (max-width: 1089px) 100vw, 1089px\" \u002F>\u003C\u002Ffigure>\n\n\n\n\u003Cp>Their performance is quite different in this regard, of the widely available models, ChatGPT and Gemini perform quite well, in both cases, the advanced – subscription-based – models work better.  \u003C\u002Fp>\n\n\n\n\u003Cp>But apart from hallucinations, we also have to remember that despite their impressive performances, these AI models are not intelligent in the human sense of the word. They are superadvanced autocomplete machines but \u003Ca href=\"https:\u002F\u002Fmedicalfuturist.com\u002Fgenerative-adversarial-networks-how-can-ai-learn-so-much-while-its-iq-remains-zero\" target=\"_blank\" rel=\"noreferrer noopener\">have no actual understanding\u003C\u002Fa> of what they are talking about. That is why it is crucial to make sure to always think critically about AI-generated results – as the models do not. \u003C\u002Fp>\n\n\n\n\u003Cp>In conclusion, as AI continues to grow and evolve, the importance of being adept at prompt engineering cannot be overstated. The ability to elicit useful and meaningful responses from AI can empower us to make the most of this cutting-edge technology. Remember, practice is key, and each question we ask is a step towards becoming more fluent in the language of AI. \u003C\u002Fp>\n\n\n\n\u003Cp>In general, use AI to expand your knowledge and ideas instead of solving things on behalf of you. \u003C\u002Fp>\n",{"rendered":190,"protected":20},"\u003Cp>Uncover the art of ‘prompt engineering’ with our comprehensive guide, designed to help you get the most from ChatGPT. Learn to craft effective prompts, understand ChatGPT’s capabilities, and navigate its limitations. Transform your interactions with AI today\u003C\u002Fp>\n",51365,{"_acf_changed":20,"footnotes":27},[32],[35,36,195,196,39,197,198,199],7687,7725,7833,7835,144,[42],[],[203],1833,[205,14,56,57,58,59,60,62,64,65,206,207,68,208,209,210,70],"post-51353","tag-ai-text-generator","tag-chatgpt","tag-prompt-engineering","tag-healthcare-ai","tag-artificial-intelligence",{"id":191,"alt_text":27,"caption":27,"description":27,"media_type":72,"media_details":212,"post":111,"source_url":248},{"width":213,"height":74,"file":214,"filesize":215,"sizes":216,"image_meta":247},1920,"2023\u002F06\u002Ftmf_article_372.png",181376,{"medium":217,"large":223,"thumbnail":229,"medium_large":234,"1536x1536":235,"2048x2048":241},{"file":218,"width":219,"height":220,"mime-type":83,"filesize":221,"source_url":222},"tmf_article_372-370x208.png","370","208","28583","https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F06\u002Ftmf_article_372-370x208.png",{"file":224,"width":225,"height":226,"mime-type":83,"filesize":227,"source_url":228},"tmf_article_372-768x432.png","768","432","74781","https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F06\u002Ftmf_article_372-768x432.png",{"file":230,"width":231,"height":231,"mime-type":83,"filesize":232,"source_url":233},"tmf_article_372-150x150.png","150","12501","https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F06\u002Ftmf_article_372-150x150.png",{"file":224,"width":225,"height":226,"mime-type":83,"filesize":227,"source_url":228},{"file":236,"width":237,"height":238,"mime-type":83,"filesize":239,"source_url":240},"tmf_article_372-1536x864.png","1536","864","177228","https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F06\u002Ftmf_article_372-1536x864.png",{"file":242,"width":243,"height":244,"mime-type":83,"filesize":245,"source_url":246},"tmf_article_372-2048x1152.png","2048","1152","247948","https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F06\u002Ftmf_article_372-2048x1152.png",{"aperture":99,"credit":27,"camera":27,"caption":27,"created_timestamp":99,"copyright":27,"focal_length":99,"iso":99,"shutter_speed":99,"title":27,"orientation":99},"https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2023\u002F06\u002Ftmf_article_372.png",{"cta_type":103,"cta_color":27,"related_books":250,"related_posts_footer":253,"related_posts":20,"subtitle":27,"key_takeaways":257},[106,251,252],47427,37033,[254,255,256],50987,55033,55227,[258,260],{"title":259},"\u003Cp>Prompt engineering is a key emerging skill in healthcare.\u003C\u002Fp>\n",{"title":261},"\u003Cp>With just a few tips and tricks, you can start using ChatGPT-like tools with confidence and ease.\u003C\u002Fp>\n",{"yoast_wpseo_title":263,"yoast_wpseo_metadesc":264,"yoast_wpseo_canonical":184},"Prompt Engineering: 11 tips to craft great ChatGPT prompts","Master prompt engineering with our guide, learn to craft effective prompts and optimize your interactions with ChatGPT for the best results",{"self":266,"collection":271,"about":273,"author":275,"replies":277,"version-history":280,"predecessor-version":284,"wp:featuredmedia":288,"wp:attachment":291,"wp:term":294,"curies":305},[267],{"href":268,"targetHints":269},"https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Fposts\u002F51353",{"allow":270},[128],[272],{"href":131},[274],{"href":134},[276],{"embeddable":26,"href":137},[278],{"embeddable":26,"href":279},"https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Fcomments?post=51353",[281],{"count":282,"href":283},30,"https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Fposts\u002F51353\u002Frevisions",[285],{"id":286,"href":287},59329,"https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Fposts\u002F51353\u002Frevisions\u002F59329",[289],{"embeddable":26,"href":290},"https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Fmedia\u002F51365",[292],{"href":293},"https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Fmedia?parent=51353",[295,297,299,301,303],{"taxonomy":157,"embeddable":26,"href":296},"https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Fcategories?post=51353",{"taxonomy":160,"embeddable":26,"href":298},"https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Ftags?post=51353",{"taxonomy":163,"embeddable":26,"href":300},"https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Fproject_category?post=51353",{"taxonomy":166,"embeddable":26,"href":302},"https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Fcontact_email_category?post=51353",{"taxonomy":169,"embeddable":26,"href":304},"https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Fyst_prominent_words?post=51353",[306],{"name":173,"href":174,"templated":26},[308],{"id":107,"date":309,"date_gmt":310,"guid":311,"modified":313,"modified_gmt":314,"slug":315,"status":13,"type":316,"link":317,"title":318,"content":320,"excerpt":322,"author":324,"featured_media":325,"comment_status":25,"ping_status":25,"template":27,"yst_prominent_words":326,"class_list":330,"better_featured_image":333,"acf":351,"yoast_meta":371,"_links":374},"2019-09-06T22:00:04","2019-09-06T20:00:04",{"rendered":312},"https:\u002F\u002Fapi.medicalfuturist.com\u002F?post_type=book&p=24764","2023-03-12T18:14:56","2023-03-12T17:14:56","technologies-shaping-the-future-of-pharma","book","https:\u002F\u002Fapi.medicalfuturist.com\u002Fbooks\u002Ftechnologies-shaping-the-future-of-pharma\u002F",{"rendered":319},"How Technologies are Shaping the Future of Pharma",{"rendered":321,"protected":20},"\n\u003Cp>We designed this e-book to serve as a collection of relevant examples, best practices and exciting ideas that can help any pharmaceutical company prepare for change. Many pharma companies have been trying to hop on the “digital train”. \u003Ca href=\"https:\u002F\u002Fleanpub.com\u002Fthefutureofpharma\" target=\"_blank\" rel=\"noreferrer noopener\">This e-book\u003C\u002Fa> was meant to prove that instead of a train of innovation, stakeholders should think in terms of spaceships and while there is still time to embrace digital health and patient empowerment, those that do it faster will get exponentially ahead of their competitors.\u003C\u002Fp>\n",{"rendered":323,"protected":20},"\u003Cp>We designed this e-book to serve as a collection of relevant examples, best practices and exciting ideas that can help any pharmaceutical company prepare for […]\u003C\u002Fp>\n",10,49977,[327,328,329],1681,2349,4977,[331,316,332,57,59,60],"post-24764","type-book",{"id":325,"alt_text":27,"caption":27,"description":27,"media_type":72,"media_details":334,"post":107,"source_url":350},{"width":335,"height":336,"file":337,"filesize":338,"sizes":339,"image_meta":349},320,415,"2019\u002F09\u002Ffuture-of-pharma.png",23355,{"medium":340,"thumbnail":345},{"file":341,"width":342,"height":220,"mime-type":83,"filesize":343,"source_url":344},"future-of-pharma-320x208.png","320","49756","https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2019\u002F09\u002Ffuture-of-pharma-320x208.png",{"file":346,"width":231,"height":231,"mime-type":83,"filesize":347,"source_url":348},"future-of-pharma-150x150.png","21008","https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2019\u002F09\u002Ffuture-of-pharma-150x150.png",{"aperture":99,"credit":27,"camera":27,"caption":27,"created_timestamp":99,"copyright":27,"focal_length":99,"iso":99,"shutter_speed":99,"title":27,"orientation":99},"https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2019\u002F09\u002Ffuture-of-pharma.png",{"leanpub_url":352,"buy_button_text":353,"preview":354},"https:\u002F\u002Fleanpub.com\u002Fthefutureofpharma?utm_source=books&utm_medium=referral","Get it on Leanpub",[355,357,359,361,363,365,367,369],{"image":356},"https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2019\u002F09\u002F1114_pharma_ebook_2022-cover-copy.png",{"image":358},"https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2019\u002F09\u002FTechnologies-Shaping-The-Future-of-Pharma-UPDATED-03.png",{"image":360},"https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2019\u002F09\u002FTechnologies-Shaping-The-Future-of-Pharma-UPDATED-04.png",{"image":362},"https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2019\u002F09\u002FTechnologies-Shaping-The-Future-of-Pharma-UPDATED-05.png",{"image":364},"https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2019\u002F09\u002FTechnologies-Shaping-The-Future-of-Pharma-UPDATED-06.png",{"image":366},"https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2019\u002F09\u002FTechnologies-Shaping-The-Future-of-Pharma-UPDATED-07.png",{"image":368},"https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2019\u002F09\u002FTechnologies-Shaping-The-Future-of-Pharma-UPDATED-08.png",{"image":370},"https:\u002F\u002Fcdn.medicalfuturist.com\u002Fwp-content\u002Fuploads\u002F2019\u002F09\u002FTechnologies-Shaping-The-Future-of-Pharma-UPDATED-09.png",{"yoast_wpseo_title":372,"yoast_wpseo_metadesc":373,"yoast_wpseo_canonical":317},"How Technologies are Shaping the Future of Pharma - The Medical Futurist","This e-book serves as a collection of relevant examples, best practices and exciting ideas that can help pharmaceuticcal companies prepare for change.",{"self":375,"collection":380,"about":383,"author":386,"replies":389,"wp:featuredmedia":392,"wp:attachment":395,"wp:term":398,"curies":401},[376],{"href":377,"targetHints":378},"https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Fbook\u002F24764",{"allow":379},[128],[381],{"href":382},"https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Fbook",[384],{"href":385},"https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Ftypes\u002Fbook",[387],{"embeddable":26,"href":388},"https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Fusers\u002F10",[390],{"embeddable":26,"href":391},"https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Fcomments?post=24764",[393],{"embeddable":26,"href":394},"https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Fmedia\u002F49977",[396],{"href":397},"https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Fmedia?parent=24764",[399],{"taxonomy":169,"embeddable":26,"href":400},"https:\u002F\u002Fapi.medicalfuturist.com\u002Fwp-json\u002Fwp\u002Fv2\u002Fyst_prominent_words?post=24764",[402],{"name":173,"href":174,"templated":26},1788949155113]