

Tever
58 posts

@devTever
Ornitogames 🦦 | Educación digital con innovación y ética. Profesor de FP Hago proyectos chulis 😎












After watching Sid’s presentation on his own case don’t think this is a story about how AI transforms oncology, its more a story about what you can do in healthcare, if you have unlimited financial resources. He had advanced sarcoma, which is a cancer from the connective tissue like bones, cartilage or muscles. It’s a group of very different types of diseases with very distinct characteristics, very rare and hard to treat at the same time. As there tends to be not even an established first line therapy at times for some forms of sarcoma, it’s essential to perform a broad genetic testing of the cancer cells in order to find a personalized treatment with already existing drugs, which might help your case. And that’s exactly what in his case was done and most probably (we don’t know every detail) saved his life. Many sarcoma patients unfortunately are not that lucky with their cancer and its genetic fingerprint. This is nothing what can’t be done without AI and this would most likely be done in every cancer center around the world. He then did a lot of stuff on top like PRRT (which is a special form of radiation via the blood) of some bone metastases in Germany and a special form of a PET scan in China, which again is only possible with the financial background. The PRRT alone sets you back 100k USD. Was all of that necessary to beat the cancer? I’m not so sure, but we will never know. AI will definitely be a game changer for oncology and medicine in general going forward, but it can’t solve everything. The essential parts of his therapy path can already be done today and for many patients. I’ll link his full presentation on his case below.





Today we're introducing TRIBE v2 (Trimodal Brain Encoder), a foundation model trained to predict how the human brain responds to almost any sight or sound. Building on our Algonauts 2025 award-winning architecture, TRIBE v2 draws on 500+ hours of fMRI recordings from 700+ people to create a digital twin of neural activity and enable zero-shot predictions for new subjects, languages, and tasks. Try the demo and learn more here: go.meta.me/tribe2




