AI Medical Assistants Have Arrived! Part 1

Thank you for reading NewHealthcare Platforms' newsletter. We focus on the massive transformative impact of exponential technologies, platform business models, and care delivery innovation on healthcare. In addition, you'll find updates on our validated products and services, designed to facilitate partnerships between startups, tech companies, and healthcare organizations in this rapidly evolving landscape. Don't forget to Subscribe and Follow!

Hello again friends and colleagues,

In previous newsletters, we explored the massive potential of Generative AI in healthcare—a traditionally conservative industry where technology adoption often takes its sweet time, measured in years if not decades. But as predicted in those prior newsletters, here we are, just a few months since OpenAI mainstreamed Generative AI with the launch of ChatGPT on November 30, 2022, and products are being rolled out with astonishing speed, leaving many healthcare insiders a little shell-shocked.

?

Giants like Amazon and Microsoft, in a groundbreaking partnership with traditionally aloof Epic, are not just running but leaping toward actualizing the tools we've discussed. In this newsletter, I'll take a brisk walk down memory lane to summarize a couple of my previous articles for context. We'll then shed light on a couple of the recent product announcements. We will conclude by peering into the crystal ball again and predicting the next few months. Attempting to project longer than a few months seems would be a fools errand. As Lenin once remarked, "There are decades where nothing happens; and there are weeks where decades happen." We're unequivocally living in those latter times! Buckle up!

?

In the December 26, 2022 newsletter: "Assistant: AI Keeps Dr. Thompson at the Bedside", we focused on the day-to-day life of Dr. Thompson. Dr. Thompson wasn't battling technology; she was harnessing it. At her side was an AI-powered "Assistant" that streamlined administrative duties like scheduling and documentation, leaving her with more quality time for patient care. The Assistant wasn't just a mute note-taker; it assisted in clinical decision-making and even responded to patients' inquiries about lab results, seamlessly integrating biometric authentication for a secure and efficient workflow. We specifically emphasized the following features that AI brought to the table:

?

AI Clinician Assistance: From handling administrative clutter to providing clinical support, AI would ease the burden on healthcare professionals.

Biometric Authentication: In a world where AI and humans co-exist in clinical settings, biometric measures would ensure a secure and quick method for clinician approval of AI actions. It also ensures safety given generative AI's tendency to occasionally "hallucinate", and regulatory compliance given the anticipated lag of regulation behind the rate of technological progress.

Effortless Evidence at the Bedside: AI's capability to quickly pull up the latest research and medical guidelines would provide clinicians swamped with data and research with up-to-date information integrating patient-specific data with the most recent research results.

Enhanced Clinician/Patient Communication: The ability of generative AI to translate technical jargon to lay language or even different languages would enhance dialogue between clinicians and patients.

This peek into Dr. Thompson's world offered an optimistic narrative, demonstrating that technology could alleviate some of the intense pressures clinicians face, rather than exacerbate them.

In my second newsletter dated May 21, 2023, "The Future of (Healthcare) Work is Hybrid Human/AI Teams!", we explored the evolving dynamics of hybrid human-AI teams in healthcare settings. I was curious then, as I am now, about how a traditionally siloed industry could bring together the power of Generative AI and the emotional intelligence of human clinicians. We discussed 4 critical aspects:

Promise of Hybrid Teams: The coupling of AI's ability to digest massive amounts of data and information with human clinical insight creates strong synergies. A prominent example was the Stanford study that highlighted the remarkable accuracy of human/AI ophthalmology teams in diagnosing Diabetic Retinopathy.

Risks on the Horizon: However, all was not roses. Concerns ranging from AI errors to over-reliance on technology underscored the challenges ahead. Here, too, we discussed the absolute necessity of ethical considerations, particularly when dealing with highly sensitive patient data.

Operational Must-Dos: Healthcare organizations need more than just a strategy; they require a blueprint. This involved extensive planning, workforce training, and most importantly, transparent communication and culture change led by the top organizational leadership.

Policy Levers: The role of public policy was emphasized with recommendations for developing ethical guidelines and revisiting reimbursement models to support the effective and responsible embedding of AI in healthcare delivery.

?

Now that we've set the stage with these ideas, let's talk about the products that Amazon and Microsoft/Epic have announced and that are being tested by pioneering organizations.

??

Amazon HealthScribe

The real pain point in clinical settings isn't necessarily the absence of innovation but the lack of time. AWS HealthScribe, announced July 26, 2023 at AWS Summit New York, promises to alleviate this bottleneck. By providing a HIPAA-eligible service that utilizes both speech recognition and generative AI, AWS HealthScribe aims to streamline clinical documentation, a process that often keeps clinicians behind screens instead of in front of patients. Here's a breakdown:

Time-saving Transcripts: HealthScribe takes physician-patient conversations and converts them into transcripts. It can automatically extract key details like medical terms and medications, making them ready for input into an Electronic Health Record (EHR) system.

AI-powered Summaries: Building upon the simple transcript, HealthScribe employs generative AI to summarize the conversation for clinical records. It even cites the sources of each line in the transcript for the sake of accuracy and transparency.

Specialty-specific: Initially, AWS HealthScribe is focusing on two specialties—general medicine and orthopedics. This is a smart move as it allows for fine-tuning and specialization, which are key in medical settings.

Security and Compliance: Crucial for any healthcare application, HealthScribe encrypts data in transit and at rest, giving healthcare providers control over where their data is stored.

AWS HealthScribe is one of many tools that are ushering in a new era of AI-driven clinical documentation. AWS HealthScribe presents itself as the Swiss Army knife for healthcare providers eager to incorporate generative AI into their clinical applications. It's a plug-and-play solution that seemingly tackles the administrative time-drain that exhausts today’s clinicians, aligning remarkably well with what I envisioned for our imaginary Dr. Thompson.

?

Microsoft/EPIC

Epic's dominance in the U.S. EHR (Electronic Health Record) market is nothing short of remarkable, serving as a linchpin in the healthcare data ecosystem. With over 305 million patient records—an increase of 50 million from the previous year—Epic has expanded its footprint substantially, adding 83 hospitals and 14,330 beds in 2022 alone. This gain stands in stark contrast to its competitors, none of which added beds and only Oracle Cerner adding hospitals. Epic now controls 35.9% of the U.S. acute care hospital market share and 47.6% of hospital beds. This upward trajectory, marked by a 13% revenue spike to $3.8 billion in 2021, has been consistent since 2017, vastly outpacing its nearest rivals. With upcoming plans to be used by all hospitals on the U.S. News & World Report's Best Hospitals Honor Roll and key collaborations with tech giants like Microsoft, Amazon and Apple, few companies have the impact that EPIC has in the care that the huge majority of Americans receive.

In a blog post on August 22, 2023, Eric Boyd, CVP, AI Platform at Microsoft, made the following announcement:

“We combined Microsoft’s large-scale cloud and AI technologies with Epic’s deep understanding of the healthcare industry and clinical workflows to address many current issues affecting clinicians. Today, we are announcing the expansion of our strategic initiative to bring AI to healthcare at scale, integrating conversational, ambient and generative AI technologies across the Epic electronic health record (EHR) ecosystem. Intended to speed development of solutions for healthcare’s most critical needs, the initiative will expand secure access to AI-powered clinical insights and administrative tools within a wide range of Epic modules to enhance patient care, increase operational efficiency, improve healthcare experiences, and support the financial integrity of health systems globally.

We are working together to rapidly deploy dozens of copilot solutions that securely unlock the potential value that the Microsoft Cloud and our AI technologies enable as health systems strive to overcome the urgent staffing, financial and clinical access challenges they face today. Epic will showcase many of these new capabilities that build on our Azure OpenAI Service and Nuance DAX Express solutions at its annual Users Group Meeting today, including:

Enhancing clinician productivity with note summarization: Building on the previously announced AI-assisted Epic In Basket, the new solutions are targeted at increasing clinical efficiency for physicians and nurses, helping them become more productive in their daily clinical workflow. The solutions will help support faster documentation through suggested text and rapid review with in-context summaries.

Enhancing clinician productivity with embedded ambient clinical documentation: Leveraging Nuance’s Dragon Ambient eXperience (DAX) technology, which is already deployed with hundreds of Epic customers and currently supporting thousands of physicians, Epic will showcase this DAX Express AI technology embedded into the native Epic Hyperdrive platform and Haiku mobile application, further enhancing a seamless workflow experience for users. In addition, Nuance has been named by Epic as one of the first Partners in Epic’s Partner and Pals third-party vendor program.

Driving administrative efficiencies through reduction in manual, labor intensive processes: Revenue cycle management is one of many areas where generative AI can meaningfully improve efficiency. For example, Epic will demonstrate an AI-powered solution that provides medical coding staff with suggestions based on clinical documentation in the EHR to improve accuracy and streamline the entire coding and billing processes.

Advancing medicine for better patient outcomes: By using Azure OpenAI Service, Epic is now delivering generative AI exploration for an initial set of users via SlicerDicer to fill gaps in clinical evidence using real-world data and to study rare diseases and more.

Our work to integrate Azure OpenAI Service and Nuance ambient technologies within the Epic ecosystem shows that broader strategic collaborations can rapidly accelerate the availability of actionable AI-driven solutions for healthcare organizations and the patients they serve.”

Few partnerships have the potential to impact healthcare as Microsoft/EPIC/ OpenAI can!

?

I'm out of time and space! So projections for the next few months will have to wait till next week.

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

See you next week,

Sam