Fluid Imaging Technologies has released a white paper that documents how image quality affects the ability of imaging particle analysers to produce accurate measurements.
It also considers the affect of image quality on how imaging particle analysers automatically recognise and classify different particles in a heterogeneous mixture.
Reporting the results of two experiments conducted on the FlowCAM particle imaging and analysis system, the timely white paper compares the measurements derived from sharp images to those derived from blurry images and concludes that deriving measurements from blurry or 'fuzzy' images reduces the accuracy of the resulting measurement and classification data.
The FlowCAM automatically takes high-resolution, digital images of microscopic particles and organisms, measures each one based on dozens of parameters in real time and automatically differentiates and classifies them for analysis.
In the first experiment, NIST-traceable calibrated size beads (10um +/- .08um) were imaged using two different flow cells to yield a group of sharp images and a group of blurry images.
The sharp images averaged 9.81um by mean equivalent spherical diameter (ESD) compared with 15.64um for the blurry images, which also yielded a higher standard deviation and co-efficient of variability.
In the second experiment, samples of Cosmarium algae were imaged to develop libraries of sharp and blurry images.
In the second experiment, samples of Cosmarium algae were imaged to develop libraries of sharp and blurry images.
Filtering with the FlowCAM's automated pattern recognition feature documented a concentration of 1,029 particles/ml for the sharp images compared with 197 particles/ml for the blurry images.
The white paper concludes that relying on blurry images results in undercounting of particles while also leading to false positives and negatives.
The paper is available at the Fluid Imaging Technologies website.
