On this episode of Lifespan News, Emmett Short discusses how GPT-4 is making waves in the medical field.
Script
Hands down, the most exciting technology in the world right now is GPT-4 and large language models in general, which are being given new skills at a wild pace. And although projects like ChaosGPT, which was built to destroy humanity, tend to grab a lot of headlines, for every one of those, there are hundreds more being used to advance humanity.
This is especially true in the longevity and medical field in general. While GPT-5 and beyond are on the horizon, and coming closer every day, we haven’t even scratched the surface of what GPT-4 is capable of. It is already, today, streamlining doctor’s notes and paperwork, synthesizing new chemicals that have never existed, and providing faster diagnostics.
In this video, I’ll cover some amazing ways GPT-4 is shattering the medical landscape making healthcare GFX: better, faster and cheaper and talk about AI’s potential for radically changing longevity in the not so distant future.
First up, drug discovery.
The traditional process of drug discovery has felt like, “well, if we burn enough money… eventually a drug shows up, right?” But sometimes you burn the money for nothing, money gone, no drug. Reminds me of every trip I’ve had to Mexico.
Using a “zero-shot” AI model, Absci was able to design de novo therapeutic antibodies that bind to specific targets without relying on any training data from antibodies known to bind to those targets. This approach generates antibody designs that are unlike those found in existing antibody databases, resulting in novel compounds with potentially higher efficacy.
Another company called Generate:Biomedicines recently partnered with The University of Texas MD Anderson Cancer Center to co-develop protein therapeutics for advanced cancers.
Here is the CEO Mike Nally talking about their tech.
The promise of generative biology goes beyond existing proteins found in nature and can create novel proteins purpose-built for your specific needs. Trial-and-error drug discovery is like a horse and buggy compared to generative AI drug discovery. GPT-4 and other generative AI technologies are simply faster, more effective and affordable ways to find treatments.
Next up, diagnostics. In today’s healthcare landscape, traditional diagnostic methods can feel like playing a high-stakes game of whack-a-mole, with human error, potential misdiagnosis, and unnecessary testing rearing their ugly heads. But with the advent of GPT-4, diagnostics in healthcare are on the cusp of a revolution.
Generative AI’s unparalleled pattern recognition and natural language understanding capabilities hold the promise of improving diagnostic accuracy, ultimately making healthcare better, faster, and cheaper.
Dr. Isaac Kohane, a computer scientist at Harvard and a physician, recently put GPT-4 to the test, discovering it was able to answer US medical exam licensing questions correctly more than 90% of the time. According to Dr. Kohane, GPT-4 even demonstrated better clinical judgment than “many doctors” he has observed. In one stunning example, GPT-4 correctly diagnosed a rare 1 in 100,000 condition, using only a few key details provided by Dr. Kohane.
Despite its impressive capabilities, GPT-4 is not without its limitations. The AI model can still make mistakes and may occasionally produce incorrect information.
Dr. Kohane writes in his book, “The AI revolution in medicine,” that he was “both impressed and horrified.” “On the one hand, I was having a sophisticated medical conversation with a computational process,” he wrote, “on the other hand, just as mind blowing was the anxious realization that millions of families would soon have access to this impressive medical expertise, and I could not figure out how we could guarantee or certify that GPT-4’s advice would be safe or effective.”
Growing pains aside, the implications are incredibly exciting. With faster and more accurate diagnoses, GPT-4 has the potential to significantly reduce misdiagnosis rates and unnecessary testing, leading to better resource allocation and cost savings in the healthcare industry, but it’s like Tesla’s full self-driving software, it still needs a human in the driver’s seat taking responsibility and intervening when necessary.
In addition to its diagnostic prowess, GPT-4 can also assist doctors in communicating with patients more effectively, translating medical jargon into easily understandable language, and even providing suggestions for compassionate bedside manners. And when it comes to cracking jokes, it’s already less awkward than when your gynecologist does it.
This leads us to my final example, paperwork. The staggering weight of excessive documentation and administrative tasks has long plagued healthcare professionals. In fact, one study revealed that doctors spend two hours on paperwork for every hour spent with a patient. This burden contributes to physician burnout, and negatively effects the overall efficiency of the healthcare system.
But now, a groundbreaking solution has emerged: Microsoft and Nuance Communications have created DAX Express, a fully automated clinical documentation application that seamlessly combines ambient AI with GPT-4’s advanced reasoning and natural language capabilities.
These notes are instantly available for the clinician to review, and the application is tightly integrated with electronic medical records software. Clinicians retain the final “sign-off” on the clinical note, ensuring accuracy and quality.
The recent integration of GPT-4 into DAX Express has been a game-changer. By reducing doctors’ workloads and administrative tasks, DAX Express empowers doctors to focus more on their patients, reduces administrative costs, and improves the accuracy of clinical notes, ultimately leading to better patient care. Besides helping physicians, DAX Express can also help with diagnosis because a patient’s history is kept in constant context.
To sum up, GPT-4 is revolutionizing healthcare, opening the door to a future filled with potential. Imagine the possibilities in the next ten, or even five years, specifically in the realm of curing aging. GPT-4 could help decode the intricate biological mechanisms of aging, and generate original compounds and treatments that slow or even reverse the process. Personalized longevity planning that adapts to every choice you make so you always have a clear path to optimal health and vitality.
Of course, there are risks, and it’s too early to take our hands off the wheel. Despite its advanced capabilities, GPT-4 is still prone to errors and hallucinations, which could lead to misdiagnosis, inappropriate treatments, or even harmful interventions if not carefully managed. AI systems like GPT-4 are trained on vast amounts of data that may contain inherent biases. If these biases are not addressed, they could perpetuate and amplify existing disparities in healthcare, further marginalizing vulnerable populations.
As we stand in awe of GPT-4’s transformative impact, we must confront these challenges head-on, ensuring a future where the incredible potential of AI in healthcare benefits everyone, while minimizing its risks and pitfalls. But if we do, the future looks bright.
We’d love to hear your thoughts and experiences regarding GPT-4 and its impact on healthcare. Share your insights, opinions, and any personal stories you may have in the comments section below.
Lastly, if you found this content valuable and want to stay updated on the latest breakthroughs and discussions, please remember to like, share, and subscribe to our channel. By doing so, you’ll not only be supporting our mission to share knowledge, but you’ll also ensure you never miss an exciting update in the world of AI and healthcare. Speaking of that, check out the next video right here.
