Algae through the looking glass

Abstract

Microalgae are one of the most suitable subjects for testing the potentiality of light microscopy and image analysis, because of the size of single cells, their endogenous chromaticity, and their metabolic and physiological characteristics. Microscope observations and image analysis can use microalgal cells from lab cultures or collected from water bodies as model to investigate metabolic processes, behavior/reaction of cells under chemical or photic stimuli, and dynamics of population in the natural environment in response to changing conditions. In this paper we will describe the original microscope we set up together with the image processing techniques we improved to deal with these topics. Our system detects and recognizes in-focus cells, extracts their features, measures cell concentration in multi-algal samples, reconstructs swimming cell tracks, monitors metabolic processes, and measure absorption and fluorescent spectra of subcellular compartments. It can be used as digital microscopy station for algal cell biology and behavioral studies, and field analysis applications.

Applications of artificial neural networks in three agro-environmental systems: microalgae production, nutritional characterization of soils and meteorological variables management

“…Microalgae identification using ANN has been previously reported with taxonomic accuracy of up to 99%, using micrographic image analysis (Coltelli et al, 2017), similar accuracy to that achieved in the present study for monoalgal cultures. The method reported here demonstrated its efficiency in discriminating mixed algal cultures, whereas it was less efficient when there were smaller percentages of another species in the samples.…”

“…Moreover, the absorption spectrum in the 400 to 700 nm range is used to determine the extinction coefficient of the biomass, a specific microalgae strain and culture conditions indicator (Rubio Camacho et al, 2003). Based on this, absorption properties could be a possible approach to distinguish between microalgae species (Coltelli et al, 2017).…”

“…In microalgae identification using ANN, extracted features such as the perimeter, shape, area and Fourier Transform of microalgae micrographics were used to train a model capable of identifying the genera Navicula, Scenedesmus, Microcystis, Oscillatoria and Chroococcus (Mosleh et al, 2012). Microalgae micrographic image processing and color analysis were used with ANN to achieve taxonomic accuracy of up to 99% by first detecting a cell in the image and subsequently extracting the detected cell color (Coltelli et al, 2017).…”

Monoalgal and mixed algal cultures discrimination by using an artificial neural network

“…Moreover, the absorption spectrum in the 400 to 700 nm range is used to determine the extinction coefficient of the biomass, such as a specific microalgae strain characteristic or the culture conditions [16]. Based on this, absorption properties could be a possible approach to distinguish between microalgae species [17]. Artificial neural networks (ANNs) are a powerful tool for finding relationships between experimental data and the phenomena behind these data.…”

“…In microalgae identification using ANN, extracted features such as the perimeter, shape, area and Fourier transformations of microalgae micrographics were used to train a model capable of identifying the genera Navicula, Scenedesmus, Microcystis, Oscillatoria and Chroococcus [22]. Microalgae micrographic image processing and colour analysis were used with ANN to achieve taxonomic accuracy of up to 99% by first detecting a cell in the image and subsequently extracting the detected cell colour [17].…”