Go to symposium website → www.slas.org/HighContentAnalysis
Back To Schedule
Tuesday, October 22 • 11:10am - 11:30am
Advances in Imaging and Analysis- Extracting Meaning From Big Data in Volume Electron Microscopy

Sign up or log in to save this to your schedule, view media, leave feedback and see who's attending!

Many different imaging modalities now routinely produce huge amounts of data thanks to increased acquisition speeds and extensive automation. Volume electron microscopy techniques, such as serial block face SEM (SBF SEM), focused ion beam SEM (FIB SEM) and array tomography (AT), produce datasets in the terabyte regime. Multimodal imaging methods such as correlative light and electron microscopy (CLEM) can be used to navigate the sample more efficiently, reducing the amount of data that must be analysed. We have developed a set of tools that allow us to "find the needle in the haystack" using these CLEM techniques.

Electron microscopy image data has remained stubbornly resistant to automatic computational analysis, with painstaking manual segmentation (finding and delineating a structure or object of interest) still being the gold standard. However, recent generations of microscope produce data far more quickly than a small team of experts can thoroughly analyse. Whilst new deep learning techniques offer significant promise in automating this analysis, the collection of sufficient annotations to provide training data is a major bottleneck. In collaboration with the Zooniverse (zooniverse.org) we have developed a citizen science project called “Etch a Cell” (etchacell.org) in which we ask volunteers to contribute segmentations. By collecting multiple segmentations per image, we are able to aggregate the volunteers’ annotations into accurate and robust data that can be used to train our machine learning system. Beyond the research applications, we have also found the project to be an effective outreach and education tool, letting students and non-experts gain an insight into the scientific process at the raw data level.

avatar for Martin Jones, DPhil

Martin Jones, DPhil

Deputy Head of Microscopy Prototyping, The Francis Crick Institute
Martin Jones works in the Electron Microscopy Science Technology Platform at the Francis Crick Institute in London, developing new hardware and software for extracting meaning from ever more complex datasets.Martin's DPhil from Sussex University was in experimental atomic and quantum... Read More →


Tuesday October 22, 2019 11:10am - 11:30am BST
Wellcome Auditorium