Vice President, Life Sciences Solution Group Enthought Austin, Texas
Organoids are three-dimensional models of an organ or tissue that is grown in vitro from stem cells or tissue-specific progenitor cells. Given this structure, the generation and cultivation of these structures can be a complex and labor-intensive process. The analysis of organoid culture health often requires the expertise and intuition of a senior researcher, inherently making the process unscalable and variable between researchers. Rather than relying solely on researcher experience and intuition, we believe that it should be codified into an automated process that can enable companies to scale their R&D workforce. Here, using a combination of computer vision and machine learning, we developed a technique that automatically classifies the health of organoid cultures and recommends when the researcher should passage the organoids. Careful tracking and data recording throughout the culturing process will allow researchers to not only optimize the machine learning model and computer vision algorithms to the organoid model in question, but also enable that information to be integrated into the experiments performed with the organoid cultures. This will allow researchers to investigate any effect that the culture process may have on the experimental results.
Learning Objectives:
Upon completion, participants will be able to understand how machine learning and computer vision can be used to analyze organoid culture state.
Upon completion, participants will be able to better understand how they can use machine learning to codify their expertise to scale their research organizations.
Upon completion, participants will be able to envision how they can codify and scale their R&D workforce with digital tools and capabilities.