Imagine if you could talk to your data directly – in plain English – and ask it questions usually that take months to answer:

• What are the different outcomes of breast cancer patients in my population who received breast surgery followed by radiation?
• What biomarkers are associated with advanced lung cancer patients treated with Tagrisso who received radiation or a surgery?
• Which side effects of my drug make physicians stop prescribing it?

How much easier would it be to design and enroll clinical studies, understand population health, and commercialize effectively?

Large language models (LLMs) like ChatGPT have arisen as a transformational means of processing massive amounts of data and putting information at users’ fingertips in seconds. If fed with the right data, these tools could displace traditional data analysis.

Health data, however, is often dense, complex, and contradictory, which means getting insights isn’t as easy as feeding information to a standard LLM. The machine learning methods that fuel standard LLMs are great at pattern recognition, but machines trained this way don’t have the “common sense” to reason and understand underlying concepts.

But what goes into creating a machine that can think like a physician? And how are such approaches transforming healthcare and life sciences today? In this session, you will learn about:

• Core components healthcare-specific AI requires to go beyond extracting data to actually understanding it
• Getting usable, accurate insight from health data instantaneously
• Case studies for powering use cases across the organization with clinically intelligent AI

You’ll also get a first look at Mendel.ai’s new solution that allows users to ask the AI questions and generate responses in English – just as one might engage with Google or ChatGPT.