Genedata and AstraZeneca Receive 2018 Bio-IT World Best Practices Award for Innovative AI Project
Deep Learning for Phenotypic Image Analysis automates high content screening workflows
May 23, 2018
Genedata, a leading provider of advanced software solutions for biopharmaceutical R&D, today announced that its collaborative project with AstraZeneca – Deep Learning for Phenotypic Image Analysis – received the 2018 Bio-IT World Best Practices Award for Informatics & Knowledge Management. The project validated Genedata’s innovative approach to automating the analysis workflow for high content screens (HCS), which reduces project completion times by an order of magnitude while increasing data quality and reproducibility of analysis results. The award was presented last week at the 2018 Bio-IT World Conference and Expo in Boston, MA.
Deep Learning-based Analysis Outperforms Traditional Approaches
Biopharmaceutical R&D increasingly relies on complex phenotypic assays in HCS workflows. Traditional image analysis approaches, however, do not scale well with the analytical complexity and amount of data produced by these assays, which require complex analysis procedures and lengthy data analysis setups. Genedata and AstraZeneca worked together in addressing these challenges with a deep learning-based approach, resulting in software that successfully automates the analysis of phenotypic HCS. The software:
- automates time-consuming and repetitive tasks during the analysis of phenotypic HCS;
- improves data quality through reliable detection of complex phenotypes and by eliminating the biased selection of handcrafted features; and
- increases reproducibility by providing a standardized end-to-end process for image and data analysis for HCS.
Ultimately, the project enables a single biologist to run an analysis workflow from image production to results interpretation, bypassing complicated analysis set-ups and numerous manual steps. The solution effectively reduces the time and costs required for typical analysis of phenotypic HCS, producing quality results from a new experiment in just seconds vs. days of work in multi-disciplinary teams.
Genedata currently collaborates with several partners from the biopharmaceutical industry to develop a new software solution for deep-learning based HCS analysis. The solution, which will be applicable to a wide range of phenotypic assays and harmonize highly complex workflows, can be licensed by interested parties in the near future.
"Genedata is excited about receiving the Bio-IT World Best Practices Award with our long-time partner AstraZeneca," said Dr. Othmar Pfannes, CEO of Genedata. "The award validates Genedata’s vision and is another proof point of our success in helping our partners to become more efficient in their R&D operations through adopting our highly innovative software solutions. Genedata remains committed to automating complex data analysis workflows in close collaboration with biopharmaceutical R&D leaders."
Genedata transforms data into intelligence with innovative software solutions and domain-specific consulting services that automate complex, large-scale experimental processes and enable organizations to maximize the ROI in their R&D, spanning early discovery all the way to the clinic. Founded in 1997, Genedata is headquartered in Switzerland with additional offices in Germany, Japan, Singapore, the UK, and the US.
LinkedIn | Twitter | YouTube
The statements in this press release that relate to future plans, events or performance are forward-looking statements that involve risks and uncertainties, including risks associated with uncertainties related to contract cancellations, developing risks, competitive factors, uncertainties pertaining to customer orders, demand for products and services, development of markets for the Company's products and services. Readers are cautioned not to place undue reliance on these forward-looking statements, which speak only as of the date hereof. The Company undertakes no obligation to release publicly the result of any revisions to these forward-looking statements that may be made to reflect events or circumstances after the date hereof or to reflect the occurrence of unanticipated events.
All product and service names mentioned are the trademarks of their respective companies.