Owkin
French medical company From Wikipedia, the free encyclopedia
Owkin is a French artificial intelligence and biotech company that aims to identify new treatments, optimize clinical trials and develop AI diagnostics.[1][2] The company uses federated learning, a type of privacy preserving technology, to access multimodal patient data from academic institutions and hospitals to train its AI models for drug discovery, development, and diagnostics. Owkin has collaborated with pharmaceutical companies around the world to improve their therapeutic programs.[3]
 | |
| Industry | Biotechnology |
|---|---|
| Founded | August 3, 2016 |
| Founder | Thomas Clozel, Gilles Wainrib |
| Headquarters | , |
Area served | US, France, UK, Switzerland, Germany, Spain |
| Products | MSIntuit CRC, RlapsRisk BC |
| Services | AI Drug Discovery, AI Drug Development, AI Diagnostics |
Number of employees | 350 (2023) |
| Website | owkin |
History
Owkin was founded in Paris in 2016, by Thomas Clozel, a clinical research doctor and son of Jean-Paul and Martine Clozel founders of Swiss biotech Actelion, and Gilles Wainrib, a professor of Artificial Intelligence.[4]
Owkin has raised over $255 million and became a ‘unicorn’ – a startup valued at more than $1 billion – in November 2021 through a $180 million investment from French biopharma company Sanofi.[5]
Technologies
Federated learning
Owkin uses federated learning, a decentralized machine learning technique, to train machine learning models with multiple data providers.[6][7][8] Federated learning allows data providers to collaborate without moving or sharing their data.[8][7]
The MELLODDY project, an initiative that included Owkin, 10 pharmaceutical companies, and six other partners, applied federated learning to train AI on datasets without having to share proprietary data.[9][8][10] The aim was to improve drug discovery and they built a shared platform called MELLODDY (Machine Learning Ledger Orchestration for Drug Discovery).[9][10][8] The first results of the project were published in July 2022.[8]
Transfer learning
Transfer learning is a machine learning technique that allows a model pre-trained on one task to be used on another related task.[11] Owkin uses transfer learning to work on very small datasets.[11] Owkin's model (CHOWDER) is able to understand high-level graphic patterns, such as tumors, that are themselves relying on very low-level visual patterns, in order to fully learn the tumor's visual pattern.[12]
Products and Services
MSIntuit CRC
MSIntuit CRC is an AI-powered digital pre-screening diagnostic tool to improve colorectal cancer diagnosis and treatment.[13] It screens patients for microsatellite instability (MSI), which is a key genomic biomarker in colorectal cancer.[13] MSIntuit CRC is approved for use across the European Union.[14] It underwent a blind validation in 2023, made possibly partly by its availability within Medipath, the largest pathology lab network in France.[1]
Dx RlapsRisk BC
Dx RlapsRisk BC uses AI to predict if breast cancer patients will relapse within a few years of initial treatment.[14] It is used by pathologists and oncologists to help determine the right treatment pathway for breast cancer patients.[14]
Partnerships
Summarize
Perspective
Amgen
Owkin collaborated with Amgen to test the ability of AI to improve cardiovascular prediction.[3]
Sanofi
In November 2021 Owkin entered a strategic alliance with Sanofi.[15] The alliance included a $180 million equity investment, and a $90 million discovery and development partnership focused on Sanofi's oncology efforts in four different cancers.[16] Sanofi used Owkin's technology to find new biomarkers and therapeutic targets, build prognostic models, and predict response to treatment.[17]
Bristol-Myers Squibb
In June 2022, Owkin entered a strategic alliance with Bristol-Myers Squibb to help them design potentially more precise and efficient clinical trials.[17] The collaboration initially focused on cardiovascular disease, and has the potential to expand into projects in other therapeutic areas.[18]
MSD
In December 2023, Owkin entered a strategic alliance with MSD to develop and commercialize AI-powered digital pathology diagnostics for the EU market that could be used to identify patients suitable for immunotherapies.[19]
Servier
In October 2023, Owkin and Servier started a multi-year partnership focused on developing “better-targeted therapies” in oncology and other disease areas.[20] The partnership's first two projects were in translational medicine and digital pathology.[20]
MOSAIC
MOSAIC (Multi Omic Spatial Atlas in Cancer) was formed by Owkin, Nanostring Technologies, the University of Pittsburgh, Gustave Roussy, Lausanne University Hospital, Uniklinikum Erlangen/Friedrich-Alexander-Universität Erlangen-Nürnberg, and Charité-Universitätsmedizin Berlin.[21][22] It uses spatial omics, multimodal patient data, and artificial intelligence, and aims to “offer unprecedented information on the structure of tumors” and guide new treatments.[22][21]
Publications
Summarize
Perspective
Owkin's research on AI/ML has led to a number of publications that focus on machine learning methodologies and the development of predictive models for different disease areas, mainly oncology.
- Courtiol, Pierre et al. “Deep learning-based classification of mesothelioma improves prediction of patient outcome”, Nat Med 25, 1519–1525 (2019)[23]
- Schmauch, Benoît et al. “A deep learning model to predict RNA-Seq expression of tumours from whole slide images”, Nature Communications volume 11, Article number: 3877 (2020)[24]
- Jean Ogier du Terrail et al. “Federated learning for predicting histological response to neoadjuvant chemotherapy in triple-negative breast cancer" Nat Med (2023). 10.1038/s41591-022-02155[25]
- Saiilard et al., “Pacpaint: a histology-based deep learning model uncovers the extensive intratumor molecular heterogeneity of pancreatic adenocarcinoma” Nat Commun 14, 3459 (2023)[26]
- Saillard et al., “Validation of MSIntuit as an AI-based pre-screening tool for MSI detection from colorectal cancer histology slides” Nature Communications 14, 6695 (2023)[12]
- Saillard et al., "Predicting Survival After Hepatocellular Carcinoma Resection Using Deep Learning on Histological Slides" Hepatology 72 (2020)[27]
Awards
References
Wikiwand - on
Seamless Wikipedia browsing. On steroids.