Powering worldwide collaborative research
Owkin’s federated research ecosystem enables academic medical centers and pharmaceutical partners to collaborate easily via specific private and public consortia. Our goal is to fuel collaborative research, break down data silos and accelerate medical research.
Machine Learning Ledger Orchestration for Drug Discovery
MELLODDY is a first of its kind initiative enabling 10 leading pharmaceutical companies to collaborate on machine learning for drug discovery without sharing sensitive assay data. This three year Innovative Medicines Initiative (IMI) project gathers 17 partners led by Owkin and Janssen. Owkin Connect is used to remove the bottleneck of data sharing within pharmaceutical research and to allow the 10 competitors to collaborate without exposing proprietary data. Our aim: enhancing predictive Machine Learning models on decentralized proprietary pharmaceutical companies’ data creating the world’s largest collection of small molecules.
AI on Clinical & Histology data
This EU funded consortium connects 4 top tier clinical centers in a federated network to build collaborative AI models on siloed clinical and histology data. Started in 2018, HealthChain gathers 9 partners, led by Owkin. The aim of the project is to predict treatment responses in breast cancer and melanoma to empower oncologists to make the best therapeutic decision for each patient. Owkin Connect is used to operate three federated learning channels between partners who are developing collaborative AI models to gain insights in dermato-oncology, anatomo-pathology and fertility.
More at substra.ai
Central Repository for Digital Pathology
BIGPICTURE’s vision is to become the catalyst in digital transformation in pathology. The consortium will build a data repository of around 3 million slides covering a range of diseases to develop digital pathology tools based on artificial intelligence. This GDPR compliant platform will be the first of its kind to host both quality-controlled whole slide imaging data and advanced artificial intelligence algorithms. Owkin will collaborate with a number of consortium members to enable the training of histology AI models on sensitive data while ensuring full GDPR compliance and protection of privacy and confidentiality.
IMI project sheet: IMI2 – 18 – 945358
COVID-19 Open AI Consortium
Owkin leads the Covid-19 Open AI Consortium (COAI) to bring breakthrough medical discoveries to the Covid-19 pandemic fight. In May 2020, Owkin developed a machine learning model in partnership with Gustave Roussy, Hôpital Kremlin-Bicêtre & INRIA to predict the severity of SARS-CoV-2 infection from initial CT-scans and clinical variables, published in Nature Communications.
More at Covid-19 Open AI Consortium
We are always looking for new challenges to use federated learning to improve patient outcomes and propel medical research.
Have an idea or a project to share? We would love to hear from you!