#40 - Cracking biology through AI - Jean-Philippe Vert - Owkin

All experts agree: if there's one sector that AI will transform, it's healthcare. Whether it's understanding pathologies, developing personalized treatments, the doctor-patient relationship, diagnostic methods, or creating tools to facilitate the work of healthcare professionals, everything is affected by the ongoing transformation! This revolution raises many expectations and questions about the real impact these new technologies will have on tomorrow's medicine. One of the leading figures in this movement on the international scene is none other than the French-American company Owkin, which aims to be "the first end-to-end AI biotech company," whose R&D activities are overseen by Jean-Philippe Vert, himself a true pioneer in the field of machine learning and computational biology. The first “AI-first” biotech company of its kind, bringing a breath of fresh air to an industry known for being extremely costly and time-consuming. So how does AI accelerate the development of new therapies? What is its role in identifying new molecules and targets? How can it be used to create new diagnostic tools? Can we leverage patient data without compromising their safety and privacy? And what can we expect from future models and their impact on our understanding of biology? These are all fascinating questions that Jean-Philippe breaks down with pedagogy and realism, shedding light on one of the most exciting fields of medicine in this era! Timeline: 04:21 - Jean-Philippe’s background at the interface of mathematics, computer science, and life sciences 07:10 - Owkin’s mission to become the first end-to-end AI biotech firm 09:48 - Leveraging AI for drug discovery and new treatment modalities 14:35 - Selecting target indications and therapeutic areas 17:13 - Changing the way we run clinical trials through AI 22:03 - Unlocking patient data across hospitals while preserving safety and privacy 25:40 - Developing AI-based diagnostics and driving clinical adoption 34:50 - Towards the first AI foundation model to understand biology What we also talked about with Jean-Philippe Google DeepMind Large Language Models (LLMs) Digital twins External control arms Federated learning Digital pathology MSIntuit CRC Bioptimus Neural networks As mentioned by Jean-Philippe during the episode, we recommend taking a look at Owkin and Bioptimus websites as well as their respective Publications & News section to learn more about their technology and stay posted on their latest activities. If you are a data scientist or machine learning enthusiast, you can tackle one of the many biology-related challenges proposed by Kaggle on their platform. To learn more about Owkin’s portfolio and the diversity of biomarkers and health conditions they cover, take a look at their Pipeline section. You can follow Owkin’s activities on LinkedIn, X, and YouTube, and Bioptimus on LinkedIn and X. To get in touch with Jean-Philippe, contact him over LinkedIn. If you want to give me feedback on the episode or suggest potential guests, contact me over LinkedIn or via email! And if you liked the episode, please share it, subscribe to the podcast, and leave a 5-star review on streaming platforms! 🙏 There’s now a bi-monthly newsletter around the podcast where you will be informed of the latest episodes and kept posted on the latest medical tech progress, subscribe here! Lastly, don’t forget to follow our activities on LinkedIn and our website!