Fluorescence microscopy is a wonderful technique that allows us to differentiate between different structures by exploiting the nature of special molecules called fluorophores. Due to a phenomenon called fluorescence, these molecules can be driven to emit light which is then captured by the microscope. Therefore, the light comes only from those structures in cells or tissues which are chemically bonded to these molecules. This enables high contrast visualization of specific types of tissues and cells’ organelles, and with suitable modifications at a nanometer scales [1]. However, factors several practical experimental factors can drive the contrast of image to be low and thereby hinder interpretability.
Scientists have been using a simple and indirect way of enhancing the contrast through decreasing the noise. They take several fluorescence images of the same sample and average all the images so that the randomness contributed by noise gets averaged as well. Recently, in a paper published in Biomedical Optics Express [2], researchers at UiT have designed a new algorithm that uses the same data, but processes it enhance the signal strength directly and in a non-parametric manner. This simplifies the workflow for the user and produces high quality high contrast fluorescence images. In essence, it provides an alternative to simple averaging and comes in handy as a tool for microscopist and life scientists to improve the contrast of their images and therefore, their interpretability.
References:
- Bonnie O. Leung and Keng C. Chou, "Review of Super-Resolution Fluorescence Microscopy for Biology," Appl. Spectrosc. 65, 967-980 (2011)
- Sebastian Acuña, Mayank Roy, Luis E. Villegas-Hernández, Vishesh K. Dubey, Balpreet Singh Ahluwalia, and Krishna Agarwal, "Deriving high contrast fluorescence microscopy images through low contrast noisy image stacks," Biomed. Opt. Express 12, 5529-5543 (2021)