LLMs, with their remarkable conversational abilities, are uniquely positioned to help solve this crisis. But as with the dotcom boom before it, the surge of AI 'therapists' has created a turbulent landscape where separating legitimate innovation from dangerous shortcuts isn't straightforward.
The AI race heats up with a new kid on the block from China. How did it compare against a popular U.S. genAI engine on a simple healthcare industry question?
The progress in artificial intelligence (AI) is reshaping patient care, across various dimensions, from facilitating faster discharge to curating treatment plans and suggesting lifestyle changes.
Try a variety of LLM-based tools yourself. Understand the risks, limitations, and who is working on putting quality frameworks and guardrails in place. Engage your employees and partners in discussions about what might now be possible.
OpenAI’s GPT-4 was able to identify suicidal ideation with similar accuracy to clinicians, but in a much shorter amount of time, a new study from Brightside Health found.
Anthropic's Claude3-Opus performed better than GPT-4, but both fell short of humans on a test of objective medical knowledge. The study was conducted by a firm developing LLMs specifically for healthcare that claim to be incorporating peer-reviewed sources of information.
Allegheny Health Network rolled out a large language model designed exclusively for use by its staff members. You can think of the tool as the health system’s own version of Chat-GPT, said Ashis Barad, AHN’s chief digital and information officer.
Zelis CEO Amanda Eisel shares her perspective on how the company is solving the problems of a fragmented health financial system to benefit all.
Mass General Brigham researchers conducted a study to learn more about the efficacy of large language models when used to draft responses to patient messages in the EHR. The results showed that these AI tools can do a good job at reducing physicians’ workloads and improving patient education — but also that these tools have limitations that require human oversight.
With the right solutions, providers can spend more time with the patient, building the relationship, earning their trust, and focusing on the forward-looking health journey.
Researchers recently tested ChatGPT’s ability to answer patient questions about medication, finding that the AI model gave wrong or incomplete answers about 75% of the time. Providers should be wary of the fact that the model does not always give sound medical advice, given many of their patients could be turning to ChatGPT to answer health-related questions.
NYU researchers conducted a study this year in which nearly 400 people were asked to identify whether responses to patient questions were produced by human providers or ChatGPT. Participants had a limited ability to tell the source of the responses apart, so the study authors concluded that the use of LLMs like ChatGPT could be an effective way to streamline healthcare providers’ communication with patients.
When it comes to managing inbound phone calls, underperformance has devastating cost implications.
Iodine Software recently announced that it is expanding its relationship with OpenAI. By doing so, Iodine will integrate the latest GPT-4 language model into its existing AI technologies and datasets, specifically its AwareCDI product suite.
UNC Health recently launched a pilot for its in-house generative AI app called Ava, which stands for “AI virtual assistant.” The app allows staff members to ask questions about the different digital health tools available to them across the health system, and it gives them concise answers about how to use the tools effectively. The goal is to prevent staff from searching through vast training libraries.
The use of LLMs in healthcare is still incredibly nascent, so many experts are calling for a safety framework to make this uncharted territory feel safer. There are still a lot of unknowns, but avoiding clinical use cases and not inputting patients’ health information seem to be the most important guardrails providers are putting around LLMs at the moment.
There are too many flaws in AI for it to replace physicians, execs said during a Monday panel at the MedCity INVEST conference in Chicago. Instead, it should be used to make physicians' work easier.
Hippocratic AI emerged from stealth, as well as announced $50 million in seed funding through a round co-led by General Catalyst and Andreessen Horowitz. The Palo Alto-based startup bills itself as the first large language model designed specifically for healthcare.