Presentation Information
[2ASPR-08]Deep Nanospectrometry: High-speed Label-free Single-Bacterial and Particle Phenotyping by Representation Learning Cytometry
○Sadao Ota1 (1. Research Center for Advanced Science and Technology, The University of Tokyo (Japan))
動画講演
Keywords:
Bacterial analysis,label-free optics,machine learning,representation learning
Phenotyping bacteria and other small particles typically relies on biomarkers, which limit what we are able to detect. Label-free optical methods provide an attractive alternative, but they are often slow for scalable, sensitive, and data-driven analysis.
In this talk, I will introduce Deep Nanospectrometry (DNS), an experimental and computational framework for label-free single-particle phenotyping. DNS combines high-throughput optical measurements with representation learning to capture informative features without the need for labels. Using this approach, we can distinguish bacterial species and strains, and detect stress-induced phenotypic bifurcation within just one hour of antibiotic exposure.
In this talk, I will introduce Deep Nanospectrometry (DNS), an experimental and computational framework for label-free single-particle phenotyping. DNS combines high-throughput optical measurements with representation learning to capture informative features without the need for labels. Using this approach, we can distinguish bacterial species and strains, and detect stress-induced phenotypic bifurcation within just one hour of antibiotic exposure.
Comment
To browse or post comments, you must log in.Log in