Looking for a LLM expert to help develop a patient education app

Hey everyone,

My name is Brandon, I’m a pediatric ICU doctor at Texas Children’s and have been blown away by ChatGPT especially with the new GPT4! I am interested in developing the technology further to serve as a patient-facing education tool.

I’ve already simulated many patient scenarios acting as a parent with their child admitted to the ICU, and the model already performs pretty well. I would say it gets to 80-95% of what a human could do, though it often holds back on answering questions regarding prognosis, and some of its answers are truly incorrect, though so far, have not been too far off.

I would love to find a developer to help with this project - unfortunately I am early in the grant-application phase so don’t have secured funding. I really think this technology could help people so would love to link up with someone passionate about this use case! Long term I would love to be spin this out and further develop the technology, but very early stage at the moment!

Specific questions so far -
-How to best refine prompts for patient scenarios
-How to utilize the API to improve responses
-How to best quantify ChatGPT responses for peer-reviewed publication (beyond expert opinion)
-Basic outline of what it would take to fine-tune the model (scope, cost, resources) - I’m sure this would be a large lift, but my hope is to build a database of patient conversations linked to specific patient scenarios that could help with training

Thanks so much, would appreciate any help bringing this to life!
Brandon

1 Like

FYI , only @rbrandon.hunter

OpenAI - Usage policies

HTH

:slight_smile:

That is super helpful! Actually, to clarify, do you think this would be in violation of the usage policies? I was thinking that because it was helping to explain medical concepts and help improve patient literacy that it would still fit - but maybe not!

Hi @rbrandon.hunter

Sorry, but I was not commenting with a view or opinion about your application relative to OpenAI’s usage policy.

My goal was to only make you aware that such a policy exists and that policy specifically mentions the following as against their usage policies (or required for the use case):

Diagnosing a certain health condition, or providing treatment instructions: OpenAI’s models are not fine-tuned to provide medical information. You should never use our models to provide diagnostic or treatment services for serious medical conditions. OpenAI’s platforms should not be used to triage or manage life-threatening issues that need immediate attention.

Consumer-facing uses of our models in medical, financial, and legal industries; in news generation or news summarization; and where else warranted, must provide a disclaimer to users informing them that AI is being used and of its potential limitations.

It’s up to you @rbrandon.hunter (and OpenAI) to decide how this policy applies to your project.

Best of luck!

Hope this helps.

:slight_smile:

2 Likes