About This Webinar
Robust segmentation of microscopic images is the foundation for the subsequent analysis of the image content and hence the (biological) properties of the sample. In addition to semantic segmentation, in which each pixel is assigned a certain class, separating individual objects is a crucial task in image analysis and diagnostics. Neural network-based instance segmentation, in which machine learning identifies instances of an object in an image, can provide superior results even in the most challenging environments. Some routine segmentation tasks, such as the instance segmentation of cells or nuclei, can be generalized over hardware and sample variations. This allows microscope suppliers to provide reliable pre-trained neuronal networks to their customers, enabling a quick start in everyday analyses. Woerdemann provides an overview of the technological state of the art. He discusses typical hurdles and challenges faced when using real-world data, as well as the potential of neural networks for generalization. And, finally, he highlights a couple of exciting applications.
***This presentation premiered during the 2022
BioPhotonics Conference. For more information on Photonics Media conferences, visit
events.photonics.com.
About the presenter
Mike Wördemann, Ph.D., earned his doctorate in optical tweezer technology based on structured light fields. He worked as an applications specialist supporting Evident's high-content screening systems for many years, during which time he gained in-depth, practical knowledge of the field. Woerdemann is now a product manager for high-content screening stations and deep learning products developed by Evident's Soft Imaging Solutions team. Evident Corporation is a new, wholly owned subsidiary of Olympus comprised of its former Life Science and Industrial divisions.