Aesclea: Our Oncology Foundation Model
Long-range temporal predictive modelling for oncology: building a unified transformer over 11M patients to forecast outcomes across a patient's lifespan.
Radical Labs
Working at the intersection of AI and oncology.
What we are working on
Cancer causes nearly one in six deaths worldwide. Medicine has seen every one of them and learned from almost none. Our mission is to build intelligence that learns from every cancer, and to put it within reach of every patient.
Our work so far
Radical Health's AI Achieves 100% on the US Medical Licensing Exam
Radical Health's AI scored a perfect 100% on the USMLE, validation of the knowledge engine powering personalized cancer reports for patients and doctors.
A medical embedding model that beats OpenAI 7x on note recall
How we fine-tuned EmbeddingGemma-300m on MIMIC-III clinical notes to predict disease at AUROC 0.934, outperforming OpenAI embeddings 7x on next-note recall.