AI's Unprecedented Rate of Progress & Adoption in Healthcare

Thank you for reading NewHealthcare Platforms' newsletter. We focus on the massive transformative impact of value-based medical technologies, platform business models, and care delivery innovation on healthcare. I will occasionally share updates on our unique services for companies building next-generation care solutions. You can request an exploratory consultation at https://scheduler.zoom.us/sambasta-newhealthcareplatforms/introduction-call

DISCLAIMER: This newsletter contains opinions and speculations and is based solely on public information. It should not be considered medical, business or investment advice. The banner and other images included in this newsletter are AI-generated and created for illustrative purposes only. All brand names, logos, and trademarks are the property of their respective owners. At the time of publication of this newsletter, the author has no business relationships, affiliations, or conflicts of interest with any of the companies mentioned except as noted. OPINIONS ARE PERSONAL AND NOT THOSE OF ANY AFFILIATED ORGANIZATIONS!

Hello again friends and colleagues,

We are witnessing an unprecedented rate of progress in the field of artificial intelligence, particularly in the development of large language models, or LLMs. These advanced AI systems, capable of understanding and generating human-like text, are revolutionizing various industries, and healthcare is no exception. Today, we will discuss how the rapid adoption of LLMs in healthcare, driven by their ability to improve productivity and reduce provider burnout, is paving the way for a transformative shift in the industry.

The Unprecedented Rate of LLM Progress

Research is shedding light on the remarkable pace at which LLMs are advancing. A recent study [Algorithmic Progress in Language Models by Ho, Besiroglu et al posted on arXiv 3/9/24] revealed that the effective computational capability (aka Compute) available to train LLMs has increased A TRILLION FOLD over the last 10 years! This was accomplished through a improvement in the hardware used to train those models, an increased efficiency of the algorithms that run on those computers, and a massive increase in investment in data centers, and recruitment of scientists from other fields like physics, engineering and mathematics to work on creating successive generations of these models.

The above has led to new AI-powered products being introduced at a rapid pace, with breakthroughs being announced every week. Two notable examples among these innovations are OpenAI's Sora and Cognition Labs' Devin.

Sora is a groundbreaking text-to-video generative AI model developed by OpenAI. This advanced model can create realistic and imaginative scenes from text instructions, generating videos up to a minute long. Sora focuses on understanding and simulating the physical world in motion, translating text prompts into visually compelling videos with vibrant characters, dynamic camera movements, and detailed scenes. Despite some current limitations in accurately simulating complex physics or specific cause-and-effect instances, Sora's deep understanding of language enables it to interpret prompts effectively and generate engaging content. OpenAI is preparing for Sora's deployment with safety measures being implemented, including adversarial testing by domain experts to ensure responsible usage. Sora's potential applications span various sectors, promising a significant advancement in generative video technology.

The worlds depicted in the two videos below and their inhabitants never existed. They were totally generated it by Sora based on the prompts that precede them. And this is version 1 of Sora!

Prompt: A stylish woman walks down a Tokyo street filled with warm glowing neon and animated city signage. She wears a black leather jacket, a long red dress, and black boots, and carries a black purse. She wears sunglasses and red lipstick. She walks confidently and casually. The street is damp and reflective, creating a mirror effect of the colorful lights. Many pedestrians walk about.

Prompt: A litter of golden retriever puppies playing in the snow. Their heads pop out of the snow, covered in.

Another breakthrough released last week is Devin, developed by Cognition Labs. This innovative AI software engineer can autonomously find and fix bugs in computer code, assist in maintaining and debugging computer programs, and train and fine-tune its own AI models. Devin's capabilities extend to tasks like setting up context gathering and editing files without human assistance. Devin promises to revolutionize the field of software engineering by offering advanced automation and problem-solving abilities. Devin represents a new type of LLM interface called Autonomous Agent. Agents can take on complex tasks, plan how to accomplish them, and work independently to achieve the goals. It is able to review and test its own work, and correct mistakes till the goal is accomplished. The Agent model will expand to other industries!

These two examples, Sora and Devin, represent just a fraction of the numerous breakthroughs in AI technology being released every week. As the field of AI continues to advance at an unprecedented pace, we can expect to see a wide range of innovative products and services that harness the power of AI to solve complex problems and transform various industries.

AI and LLMs in Healthcare

The healthcare industry is witnessing an unprecedented rate of adoption of LLMs, driven by their potential to improve productivity and reduce provider burnout. One area where LLMs are making a significant impact is ambient clinical documentation, which allows doctors to consensually record their visits with patients and automatically transform the conversations into clinical notes and summaries using AI. LLMs are solving this problem.

Companies like Microsoft's Nuance Communications, Abridge, and Suki have developed solutions that leverage LLMs for ambient clinical documentation. These tools help doctors focus on meaningful connections with patients by reducing the administrative workload associated with writing notes, placing orders, and summarizing patient information. Surveys have shown that more than 90% of physicians report feeling burned out on a regular basis, largely due to the paperwork they are expected to complete. LLMs are solving this problem.

The adoption of ambient clinical documentation solutions is rapidly expanding. Microsoft's Nuance announced that more than 200 organizations are now using their DAX Copilot technology, with Stanford Health Care deploying it across its entire enterprise. Abridge's solution is live across 55 specialties and 14 languages, automating more than 92% of the clerical work it focuses on. Suki is used by more than 30 specialties in around 250 health organizations nationwide, with new customers going live with the technology every week.

The success of LLMs in ambient clinical documentation is expected to accelerate the adoption of other applications in healthcare, such as clinical decision making, personalized treatment, and patient engagement and education. As doctors experience the benefits of reduced administrative workload and increased time for meaningful patient interactions, they will be more open to exploring additional ways in which LLMs can enhance their practice.

The Future of Healthcare with LLMs

The rapid advancement of LLMs and their growing adoption in healthcare promise a future where AI and human expertise work hand in hand to improve patient outcomes and transform the delivery of care. As LLMs continue to evolve, their potential applications in healthcare will expand, enabling more personalized, efficient, and accessible care.

In the realm of clinical decision making, LLMs can analyze vast amounts of medical literature, patient data, and clinical guidelines to provide evidence-based recommendations to healthcare providers. This can help improve diagnostic accuracy, treatment selection, and overall patient outcomes. Furthermore, LLMs can enable personalized treatment planning by leveraging patient-specific data and medical knowledge to generate tailored treatment plans that consider individual needs, preferences, and risk factors.

LLMs also have the potential to revolutionize patient engagement and education. By powering intelligent chatbots and virtual assistants, LLMs can provide patients with personalized health information, answer common questions, and offer guidance on self-management strategies. This can empower patients to take a more active role in their health and improve adherence to treatment plans.

I expect these predictions to be faced with the same skepticism as the predictions about ambient medical assistants did when LLMs first came out in late 2022. There is no doubt that much work is needed to resolve the challenges of maintaining data privacy, security, and address ethical concerns, and potential biases. Ensuring responsible development and deployment of LLMs in healthcare will require collaboration among stakeholders, including policymakers, healthcare organizations, and AI experts. But I believe society and the healthcare industry will find ways to address all these concerns. The human and economic benefits enabled by the unprecedented technological progress will be a massive driving force!

A healthcare revolution is upon us with AI and human expertise working together to create a healthier, more personalized, and more accessible care for all.

?

If you enjoyed today's newsletter, please Like, Comment, and Share.

See you next week,

Sam