Bioptimus Releases Open Source AI for Medical Research

Paris-based startup Bioptimus has released a new artificial intelligence model named H-optimus-0, aimed at enhancing disease diagnosis and aiding medical research. With 1.1 billion parameters, this model has been trained using hundreds of millions of images, collected from over 500,000 histopathology slides across 4,000 clinical practices. Accessible and Open Source H-optimus-0 is distinguished as the largest open-source AI model developed specifically for use in pathology. By making this technology open source, Bioptimus makes cutting-edge diagnostic tools available to a broader audience, facilitating worldwide improvements and adaptations from researchers and healthcare practitioners. Using a unique dataset, the model underwent training to perform effectively in diverse diagnostic contexts. Its capabilities have been tested on various tasks, such as five tile-level tasks to classify tissue types, and six slide-level tasks for identifying biomarkers or metastasis in different cancers. Advancements in Patient Care Rodolphe Jenatton, Bioptimus’s Chief Technology Officer, emphasized that the introduction of H-optimus-0 marks an important step forward in the field of pathology. This model is expected to bolster medical research and enhance patient care by offering more precise and rapid disease detection. Jean-Philippe Vert, the company’s co-founder and CEO, stated that the launch of H-optimus-0 is just the beginning. Bioptimus intends to develop further models that will integrate genomics and proteomics data to establish a comprehensive foundational model of biological processes. AI in the Medical Industry AI companies are increasingly focusing on the healthcare industry. Just this week, OpenAI and Thrive launched an AI Health Coach. Named Thrive AI Health, this collaboration – which was announced in a joint press release – aims to improve daily health habits through highly personalized recommendations. This AI coach utilizes scientific research and Thrive’s established behavior change methodologies to provide users with data-driven health recommendations. By incorporating biometric, lab results, and other medical data, Thrive AI Health aims to deliver a holistic approach to personalized health management.