LLM’s and Healthcare: The Changing Landscape

LLM’s and Healthcare: The Top Players and Comparative Solutions

Rohit Mahajan

In this article, we take a look at what’s trending in Large Language Models (LLMs) for healthcare and examine some of the top players and their solutions.

In recent years, large language models (LLMs) powered by Generative arti?cial intelligence (GAI) have emerged as a groundbreaking technology with the potential to revolutionize the healthcare landscape. These sophisticated models, such as OpenAI's GPT-4, have the ability to process and generate human-like text, making them invaluable tools for transforming various aspects of the healthcare industry.

LLMs provide a quantum leap in capabilities for AI to understand and process medical language and context - from supporting clinical decision-making to summarizing doctor’s notes. The importance of?medical LLMs?in healthcare is multifaceted and far-reaching.

LLMs for healthcare are currently in use and are already providing for:

·????? Enhanced Clinical Decision Support - One of the most promising applications of LLMs in health care is supporting clinical decision-making. These models can analyze vast amounts of medical literature, research papers, and patient data to provide evidence-based recommendations for diagnosis, treatment plans, and personalized medicine.

·????? Accelerated Medical Research – LLMs have a remarkable ability to process and analyze extensive repositories of scientific data in all manner of text, from detailed published clinical papers to doctors’ notes and anecdotal patient reports. With their innate ability to uncover patterns and correlations from massive amounts of unstructured data, LLMs aid researchers in accelerating medical discoveries, identifying potential treatments, and advancing our understanding of diseases and their management.

·????? Improved Patient Engagement – Physicians and other healthcare professionals often struggle to communicate complex medical information to patients in a clear and understandable manner. LLMs can bridge this gap by generating patient-friendly explanations of medical conditions, treatment options, and medication instructions. By simplifying medical jargon, LLMs can empower patients to actively participate in their own care, leading to improved adherence to treatment plans and better overall health outcomes.

·????? Streamlined Health Data Management? - The healthcare industry involves numerous administrative tasks such as claims processing, appointment scheduling, and insurance documentation. LLMs can automate and streamline these processes, reducing the administrative burden on healthcare providers and improving overall e?ciency. For example, chatbots powered by LLMs can handle basic inquiries, schedule appointments, and provide information about insurance coverage, freeing up sta? to focus on more complex tasks.

GAI and LLMs for healthcare are only in their infancy, but they are poised to expand exponentially over the next decade. The GAI market for healthcare is valued at more than $1 billion as of 2022, and it is projected to swell to nearly $22 billion by 2032. Technology giants, venture capitalists, and private equity firms will play a significant role in that market, acting as investors, partners, and innovators.

The Top Players in LLMs for Healthcare

With LLMs poised to revolutionize healthcare, it should come as no surprise that some of the biggest names in tech are taking notice and investing billions in the development of LLM and GAI solutions for healthcare. Let’s look at some of the top players and their current solutions and ventures.

Google

Google is partnering with health systems like HCA Healthcare to use generative AI technology to improve workflows on time-consuming tasks, such as clinical documentation. Google has also expanded its generative AI model Med-PaLM, which is specifically trained on medical information, to more health customers.

Of all the tech big guns currently entering the Medical LLM market, Google arguably has the most potential to shake things up. The tech giant recently introduced a healthcare-themed version of their PaLM 2, which is not surprisingly called MedPaLM2.

Google's MedPaLM2 is designed to answer medical questions and, according to Google, was the first AI system to obtain a passing score on USMLE questions from the MedQA dataset, with an accuracy of 85.4%, matching expert test takers.?The model is based on Google’s PaLM with 54B parameters.?

Amazon Web Services

In July of 2023, Amazon Web Services (AWS) made headlines when it announced HealthScribe, its own clinical documentation tool that uses generative artificial intelligence to summarize doctor-patient visits. With the launch, AWS joined a number of other tech companies, including Microsoft, in the space, jockeying for a share of the market as providers look for ways to cut the documentation burden on physicians.

A few months later, in September, Amazon announced it would invest up to $4 billion in artificial intelligence company Anthropic as the AI arms race heats up.

The two companies are forming a?strategic collaboration to advance generative AI, and the startup selected Amazon Web Services as its primary cloud provider. Along with the hefty investment, Amazon also took minority ownership in the two-year-old startup.

IBM

To help healthcare organizations multiply the impact of AI, IBM offers watsonx,?their proprietary “enterprise-ready” GAI and data platform. According to the company, "Beyond conversational search, watsonx Assistant continues to collaborate with IBM Research and watsonx to develop customized watsonx LLMs that specialize in classification, reasoning, information extraction, summarization, and other conversational use cases. Watsonx Assistant has already achieved major advancements in its ability to understand customers with less effort using large language models.”

Specifically, IBM watsonx Assistant AI healthcare chatbots can help providers do two things: keep their time focused where it needs to be and empower patients who call in to get quick answers to simple questions.

Microsoft

At the HLTH 2023 conference in October, Microsoft introduced new data and GAI solutions with LLM capabilities that will help healthcare organizations stay focused on improving patient and clinician experiences while delivering quality care more efficiently and at a lower cost. According to the company, “Together, these new solutions offer healthcare organizations a unified, safe, and responsible approach to their data and AI strategy and enable them to take advantage of the breadth and scale of Microsoft Cloud for Healthcare.”

By leveraging Microsoft Fabric -- the company’s AI-driven data analytics platform -- the Microsoft Cloud for Healthcare gives healthcare organizations the ability to combine data from previously siloed sources across their operation, such as electronic health records (EHRs), Picture Archiving and Communication Systems (PACS), labs systems, claims systems and medical devices. The solution brings structured, unstructured imaging and medical device data into the Fabric data lake with open data standards using FHIR, DICOM, and MedTech services, providing customers with one common architecture. Additionally, connectors and converters make it easier to transform FHIR, DICOM, and MedTech data from one format to another or build pipelines for specific use cases.

NVIDIA

NVIDIA recently announced the successful creation of GatorTronGPT, a medical LLM solution they developed in conjunction with researchers from the University of Florida. According to the University, the GAI program can generate doctors’ notes so well that physicians couldn’t tell the difference in a study published by both partners.

In this proof-of-concept study, physicians reviewed patient notes — some written by actual medical doctors while others were created by the new GAI program — and the physicians identified the correct author only 49% of the time.

The GatorTronGPT effort is the latest result of an ambitious collaboration announced in 2020 when the University of Florida and NVIDIA unveiled plans to erect the world's fastest AI supercomputer in academia.

John Snow Labs??

John Snow Labs has long been a leader in natural language processing (NLP)? tools and algorithms for healthcare use cases. In addition to data labeling and extraction, they have tools to de-identify clinical notes and healthcare data.?

JSL recently announced an LLM based on BioGPT (an older, smaller LLM trained on medical information) with fine-tuning based on JSL data and NLP tools. The model may perform better than ChatGPT in areas such as patient de-identification, entity resolution such as extracting procedure codes and healthcare terminology, and accuracy in clinical summarization.

Hippocratic AI

In May of 2023 healthcare large language model Hippocratic AI?landed?$50 million in seed funding led by General Catalyst and Andreessen Horowitz. Hippocratic AI's large language model has passed various healthcare certifications and licensure exams and will be developed alongside medical professionals, including physicians, nurses, and genetic counselors. Its founders hope the technology will help solve staffing shortages and burnout and augment providers to give them more time at the patient's bedside.

How BigRio Helps Bring LLM and Advanced AI Solutions to Healthcare

Take one look at the comparative solutions currently being offered by the biggest names in tech, and there should be no doubt that LLMs have emerged as a groundbreaking technology with the potential to revolutionize the healthcare landscape.

However, when it comes to leveraging LLMs, there are two primary approaches: building your own model or utilizing existing models developed by organizations like OpenAI.

Of course, it is much easier to use an off-the-shelf LLM solution; however, while these “open sources” LLMs have gained signi?cant attention and utility across various ?elds, including healthcare, they are limited by their necessary non-specificity.

What if you could build an LLM model for your healthcare organization’s unique targets and needs? You can, with BigRio’s Help!

Creating a large language model from scratch requires extensive resources, the expertise of AI developers and data scientists, the MLOps team, and computational power. It involves training the model on massive datasets, ?ne-tuning it through multiple iterations, and optimizing its performance. This process demands substantial time, expertise, and computational resources, including high-performance hardware and storage systems.

However, custom build models are substantially more powerful for your use cases than using "o? the shelf" models. The good news is that the BigRio team has extensive knowledge about Reinforcement Learning (RL) and Natural language processing (NLP), which are the basis of LLMs.

BigRio has been a facilitator and incubator in leveraging AI to improve healthcare delivery, originally in the field of diagnostics and research. We have recently been focusing our efforts on supporting startups and developing our own solutions that use LLMs and GAI to improve those areas of healthcare as well as in direct patient interactions and drug discovery.

?

You can read much more about how AI is redefining healthcare delivery and drug discovery in my new book Quantum Care: A Deep Dive into AI for Health Delivery and Research. It’s a comprehensive look at how AI and machine learning are being used to improve healthcare delivery at every touchpoint.

?

Rohit Mahajan is a Managing Partner with BigRio. He has a particular expertise in the development and design of innovative solutions for clients in Healthcare, Financial Services, Retail, Automotive, Manufacturing, and other industry segments.

BigRio is a technology consulting firm?empowering data to drive innovation and advanced AI. We specialize in cutting-edge Big Data, Machine Learning, and Custom Software strategy, analysis, architecture, and implementation solutions.?If you would like to benefit from our expertise in these areas or if you have further questions on the content of this article, please do not hesitate to?contact us.