Key Highlights:
- Bioptimus announced that its H-optimus-0 foundation model is now integrated into Proscia’s Concentriq Embeddings, enabling data scientists and researchers in pathology and life sciences to achieve breakthroughs in AI development.
- H-optimus-0 is an AI foundation model tailored specifically for pathology, delivering high performance in critical tasks advancing model development for AI-driven research, drug development, and diagnostics.
- Trained on a vast dataset of over 500,000 pathology slides, H-optimus-0 has been exposed to a diverse array of cases, enabling it to generalise effectively across different scenarios.
Bioptimus, a French startup building the reference AI foundation model in biology, announced that its H-optimus-0 foundation model is now integrated into Proscia’s Concentriq Embeddings, enabling data scientists and researchers in pathology and life sciences to achieve breakthroughs in AI development at rapid speed and scale. This integration adds H-optimus-0 to the collection of foundation models that Concentriq Embeddings makes available on the Concentriq software platform where an enterprise’s data is stored, enriched, and analysed from discovery to clinical trial.
David Cahané, co-founder and general manager at Bioptimus, said: “H-optimus-0 has set new benchmarks in AI performance, delivering best-in-class results. Our mission is to empower the scientific community and we are excited to discover what will be built on top of our cutting-edge histology foundation model. By integrating H-optimus-0 into Concentriq Embeddings, Proscia’s users now have access to a powerful tool that accelerates AI model development and drives breakthroughs not only in precision medicine, but also for therapeutic research and development.”
With its unprecedented scale and depth of training within the open-source community, H-optimus-0 leverages AI technology to drive the development of models that can enhance the precision and efficiency of pathology diagnostics, consistently meeting the performance or outperforming existing models and setting new standards in the field. Trained on a vast dataset of over 500,000 pathology slides, H-optimus-0 has been exposed to a diverse array of cases, enabling it to generalise effectively across different scenarios.
Real-world data-enhancing foundation models
Proscia’s real-world data (RWD) has been a key contributor to the success of foundation models like H-optimus-0. With high-quality, diverse, and clinically relevant datasets, Proscia’s RWD has enabled the creation of scalable models capable of powering large-scale AI development across multiple therapeutic areas. These models are transforming pathology AI development and playing a pivotal role in enabling innovations in pathology and beyond that can ultimately bring therapies to patients faster.
David West, CEO at Proscia, said: “From Concentriq Embeddings to our real-world data offering, we are committed to giving life sciences and pathology teams the tools they need to advance the next precision therapies and diagnostics. Adding Bioptimus’ H-optimus-0 to Concentriq Embeddings will help our users rapidly build high-performing algorithms at scale and unlock the promise of AI-driven therapeutic research and development.”