Google MedPaLM Mayo

Mayo Clinic Begins Testing Google Med-PaLM 2 Generative AI Model

The Mayo Clinic has started integrating Google’s Med-PaLM 2 generative AI model into its research hospital, as first reported by The Wall Street Journal. Med-PaLM 2 is a large language model (LLM ) built for diagnosing medical conditions, a variant of the PaLM 2 model fueling Google Bard’s AI.

Generative Medical

Google unveiled Med-PaLM 2 at the I/O conference in May, boasting that the model hit had scored 85% on the US Medical Licensing Examination. Now, internal emails seen by WSJ indicate that the AI is helping fill in gaps where there are too few doctors, thanks to its specialized training and ability to converse about healthcare and medical issues. And though Med-PaLM 2 has some accuracy issues common to all LLM-derived chatbots, it performs just as well as real doctors when it comes to supporting its suggestions with research and explaining the logic behind its diagnosis.

The Mayo Clinic had hinted at broader plans in June when it announced a partnership with Google Cloud. The initial project tested the Enterprise Search in Generative AI App Builder feature, which has become available to healthcare providers now that the Gen App Builder can create HIPAA-compliant apps. That’s pointed out in the report, which mentions customers controlling any data processed by Med-PaLM 2, which won’t be available to Google and will be encrypted.

“Our prioritization of patient safety, privacy, and ethical considerations, means that generative AI can have a significant and positive impact on how we work and deliver healthcare,” Mayo Clinic chief information officer Cris Ross said when the Google Cloud Deal was announced.. “Google Cloud’s tools have the potential to unlock sources of information that typically aren’t searchable in a conventional manner, or are difficult to access or interpret, from a patient’s complex medical history to their imaging, genomics, and labs. Accessing insights more quickly and easily could drive more cures, create more connections with patients, and transform healthcare.”

Interest in generative AI for healthcare has been on the rise in many forms this year. Microsoft subsidiary Nuance started incorporating GPT-4 into the Dragon Ambient Intelligence platform medical professionals use to transcribe patient interactions back in March. And the British National Health Service is pushing hospitals to adopt generative AI tools with $27 million in grants. That’s before even considering pharmaceutical research, with human trials of a drug conceived and developed using generative AI underway.

UK Promises $27M to National Health Service to Deploy Generative AI for Healthcare

Microsoft Adds GPT-4 Generative AI to Nuance’s Dragon Medical Voice Assistant

Mayo Clinic Partners With Google Cloud to Bring Generative AI to Healthcare