In 2023 generative artificial intelligence (gen AI) surprised, challenged and inspired the healthcare industry. We previewed and launched some tools to help healthcare organizations build solutions with gen AI that assist caregivers including MedLM, a family of foundation models fine-tuned for healthcare industry use cases, and Vertex AI Search for healthcare, designed to help find the right information and gain insights from it. In both of these cases, we’re bringing the best of Google Research to Google Cloud customers.

In 2024, we’ll see this technology move from experimenting and trial to real-world use cases, especially in areas that reduce administrative burdens, help clinicians easily find information, assist healthcare call center agents, and ultimately help organizations run more efficiently. Of course, in 2024, we’ll also continue to see gen AI experimentation for use cases that require more testing and development, like assimilating information from diverse sources such as medical images, textual clinical reports and voice. Ultimately, this technology will drive a new understanding about health and healthcare.

I’ll go into what I expect to see in all three of these areas: short-term optimization, long-term transformation, and profound learning, but bear in mind that not one is entirely separate from the others. The initial steps people take to remove near-term paperwork burdens, for example, will play a role in our overall understanding of gen AI and best practices in healthcare. That influences the big changes we’ll see over the coming decade. In other words, sweeping changes don’t arrive like a thunderbolt; they’re built incrementally. Here’s how I think that will work in 2024.

Optimizing administrative work with gen AI

Even for a system as dynamic as medicine and care delivery, the past few years have been unusually tumultuous. First, COVID-19 made stark the cost pressures, staffing shortages, fragmented technologies, and administrative complexities facing the industry on a global basis. Enter gen AI three years later, which can help relieve some of these very pressures. For example, gen AI can enable easier document creation by digesting reports and long files for faster consumption, helping ease the administrative workload for short-staffed clinicians.

Generative AI can also ease the cognitive burden on caregivers by assisting with clinical documentation, easily finding relevant information, and by helping radiologists, pathologists, and lab workers in going through large sets of results. Make no mistake, humans are more central to the process than ever, but with gen AI, they have a powerful new tool to do more satisfying work with less tedium.

According to the World Health Organization, there were about 28 million nurses in the world in 2020. Save them just five minutes a day, and that’s 266 years better focused on patient care.

The point is not to free up that much time, but to enhance the appeal of the profession. The WHO also says we’re short six million nurses. Incremental changes at scale create enormous change. To phrase it more simply: Less burnout, more satisfaction. Ultimately, restoring joy to medicine stands to benefit everyone.

Preparing for broad transformation built on gen AI

“Small differences making a big difference” is the means through which gen AI is beginning to create transformational change. To cite two very different examples: HCA Healthcare, which operates 180 hospitals and approximately 2,300 ambulatory sites of care, across the United States and the United Kingdom with tens of thousands of nurses, is working on gen AI to improve patient handoffs between nurses. Meanwhile, Mayo Clinic, which annually receives over two billion visitors to its website, is deploying gen AI-based enterprise search to improve information sharing about everything from understanding symptoms to explaining drugs and treatments.

In both cases, the efficiencies and better services driven by gen AI may seem incremental at first, but taking a longer-term perspective it becomes evident that it will pave the way for improved clinical documentation, more effective doctor-patient interactions, and ultimately, better health outcomes for patients.

While AI is providing breakthroughs in previously uncharted territories like protein folding with offerings like AlphaFold, it’s also on the way to doing transformational work to better understand what we already think we know. The most interesting areas for this may be in amassing and reconfiguring existing data in new ways. There are hundreds of millions of electronic health records, diagnostics reports and PDFs describing patient and hospital conditions that can, with the powerful computing of gen AI, generate new insights and unveil patterns. This, in turn, has the potential to enhance the patient care experience and contribute to the overall improvement of population health. MEDITECH is a leader in this space, starting with making it easier to search and summarize electronic health records and by auto-generating an initial draft of the hospital course narrative at discharge. Even more exciting is what gen AI will be able to do for a healthcare organization when it combines not only data from different sources, but also information from entirely different modes, like imaging scans, lab results, and patient interviews. Bringing these different pieces of information and data together can enable gen AI-powered solutions to more accurately and safely answer medical questions. It’s just the start of where this richer understanding is headed.

Deepening our understanding of the best ways to use gen AI in healthcare

Healthcare is a uniquely rewarding and, frankly, challenging field because it is a mixture of science, humanity, and the business of delivering care. Analytic rigor, empathy, and economic necessity all matter, and I’m proud to work in a field where so many talented people try continually to improve the balance of all three.

The greatest benefit of gen AI in 2024 and beyond will be a growing understanding of all three areas. This matters to healthcare regulators who are looking to understand and deploy new technologies. It matters to researchers and doctors who are looking to better understand patients and how to deliver better treatments. And, it matters to healthcare companies and the millions of patients who are looking for ways to have more equitable and effective care.

In the near term, we’ll start to grow this understanding. With a little luck and lots of hard work from dedicated professionals, we’ll see a long-term transformation of wisdom.