There are few specializations in medicine that require more precision or data than radiation oncology. Built with Meta Llama 2, RadOnc-GPT is a fine-tuned LLM that has the potential to significantly improve radiation therapy decision-making. More details ➡️ https://go.fb.me/2fes0g Paper ➡️ https://go.fb.me/dny6by RadOnc-GPT improves specificity and clinical relevance compared to general LLMs. Building with an open model like Llama 2 enabled the researchers to fine-tune on their data in such a way that no patient data is ever shared outside of a secure network, as they can train the model using their own local GPU server.
Jigyasa C.
2w
One of the decisions that will have to be made looks like the following: do we optimize a sub-par healthcare modality/strategy w debilitating side effects or do we pour brute compute + human intelligence into finding alternatives. I guess the answer is the latter; in the long run. Over short term work to minimize harm.
Scarlett Lee
1mo
Would love to see your upcoming multimodal version of Llama 3 fine-tuned to the same use case with images from X-ray imaging integrated into the workflow! Imagine a generative AI-powered diagnostic and treatment development process that is triggered automatically by X-ray imaging events and not only gives doctors a great base of suggestions to start with but also discovers other health issues that the specialists are not specifically looking for.
Very promising. Is the chatbot for post-radiation questions mentioned in the blog post meant for use by patients or clinic personnel?
In healthcare applications, since ROUGE is not a relevant metric due to its lack of semantic depth, we need a new metric that can accurately capture clinical accuracy.
#AI is THE future 🚀
Giovanni Toschi
1mo
RadOnc-GPT sounds like a game-changer for radiation therapy! Can't wait to see it in action and hear more about the impact on treatment decisions.
abdul samad
1mo
wow
This is groundbreaking work, AI at Meta! Leveraging Meta Llama for the development of RadOnc-GPT is truly impressive. The potential to improve radiation therapy decision-making is invaluable. Great job!