Alexandria Balde
Research Mentor(s): Dawen Cai
Department or Program: Cell and Developmental Biology
Authors: Alexandria Balde, Logan Walker, Dawen Cai
Session: Session 1: 12:00pm-12:50pm
Poster: 7
Abstract
Spinning disk confocal microscopy (SDCM) is a powerful tool for biological applications, producing high-contrast images with spatial precision and resolution. This effect is achieved by separating and filtering light through a moving array of pinholes placed at the image’s focal point. However, it presents a great challenge in acquisition speed, imaging larger samples and hazy background signals due to fluorophore cross-talk compared to widefield microscopy leading to the need of additional technology to synthesize the smaller images from SDCM into larger mosaics. This paper will present a modular Python standardization interface for large scale images from the Zeiss Imager-M microscope mounted with CrestOptics CICERO spinning disk confocal solution. First we acquired a multi-channel fluorescent (blue, green, red, infrared), multi-axial large sample through the open-source plugin ImageJ MicroMagellan. In order to conscientiously remove the exposure radial falloff, we applied the flatfield correction method, BaSiC, across the randomly ordered individual frames. Then we rigid-stitched the image frames, dynamically removing the overlap factor which contends with current static/non-rigid stitching Python packages, to create a composite, large-sample image and finally wrote the dataset and preserved metadata in a HyperStack tagged image file format, with the goal that to be uploaded by widely-used, open-source scientific image analysis software, ImageJ. We anticipate this technology can then be applied for whole-slide multi-fluorescent images, and specifically in our lab, an efficient way to analyze and store whole slide images of the mouse brain with multiple fluorescent indicators. Keywords: confocal microscopy, Python software, image correction, whole slide imaging, neuroscience
