Computer Vision-Based Image Analysis of Bacteria

Research output: Chapter in Book/Report/Conference proceedingBook chapter

Abstract

Microscopy is an essential tool for studying bacteria, but is today mostly used in a qualitative or possibly semi-quantitative manner often involving time-consuming manual analysis. It also makes it difficult to assess the importance of individual bacterial phenotypes, especially when there are only subtle differences in features such as shape, size, or signal intensity, which is typically very difficult for the human eye to discern. With computer vision-based image analysis - where computer algorithms interpret image data - it is possible to achieve an objective and reproducible quantification of images in an automated fashion. Besides being a much more efficient and consistent way to analyze images, this can also reveal important information that was previously hard to extract with traditional methods. Here, we present basic concepts of automated image processing, segmentation and analysis that can be relatively easy implemented for use with bacterial research.

Details

Authors
Organisations
External organisations
  • Lund University
Research areas and keywords

Subject classification (UKÄ) – MANDATORY

  • Medical Image Processing
  • Radiology, Nuclear Medicine and Medical Imaging
Original languageEnglish
Title of host publicationBacterial Pathogenesis
Subtitle of host publicationMethods and Protocols
EditorsPontus Nordenfelt, Matthias Collin
PublisherSpringer New York
Pages161-172
Number of pages12
Volume1535
ISBN (Electronic)978-1-4939-6673-8
ISBN (Print)978-1-4939-6671-4
StatePublished - 2017
Publication categoryResearch
Peer-reviewedYes

Publication series

NameMethods in Molecular Biology
Volume1535
ISSN (Print)1064-3745