Cusp-artifacts in high order Superresolution Optical Fluctuation Imaging

Superresolution Optical Fluctuation Imaging (SOFI) offers a simple and affordable alternative to the more sophisticated (and expensive) super-resolution imaging techniques such as STED, PALM, STORM, structured illumination, and other derivative methods. In SOFI, the calculation of high order cumulants provides higher resolution but drastically expands the dynamic range of the resulting image. In this study, we have identified another type of artifact for high order SOFI cumulants, dubbed as ‘cusp artifacts.’ A series of realistic simulations are performed to study the cusp artifacts under the influences of various factors, including the blinking statistics, the spatial distribution of photophysical properties of the sample, the total number of frames processed per dataset, photobleaching, and noise. Experiments, simulations, and theory all show that high order cumulants and odd-order moments could suffer from cusp artifacts. These cusp artifacts also degrade the fidelity of bSOFI that has been proposed to solve the dynamic range expansion of image pixel intensities. Alternatively, cusp-artifacts could be altogether eliminated by utilizing even-order moments constructed directly or from cumulants for image reconstruction. Together with dynamic range compression, these approaches yield improved SOFI images. Our study provides new insight into the nature of high order SOFI images, outlines guidelines for developing and screening SOFI-optimized fluorescence probes, and suggests improved strategies for SOFI data acquisition.

ANALYSIS OF THE INFLUENCE OF FILTRATION ON THE SENSITIVITY OF THE CUMULANT METHOD OF DETECTING THE ACOUSTIC SIGNALS OF LEAKAGE

An experimental study of the ratio of cumulants of acoustic signals of fluid leakage in a pipeline and interference is carried out. It is shown that the use of bandpass filtering increases the ratio of cumulants by orders of magnitude. The purpose of the article is to analyze the sensitivity of the cumulant method of detecting fluid leaks in a pipeline based on the statistical processing of the acoustic leakage signal and interference generated by the moving fluid flow. To analyze the possibilities of a cumulant leak detection method, it is necessary to have specific values of cumulants of real leakage signals and interference, which can be obtained as a result of a full-scale experiment. To generate a leakage signal, an installation is used that allows physical modeling of the leakage of liquid in the pipe in standing water. To simulate a leak in a pipe, a nozzle with copper walls and a through hole with a diameter of 1 mm was used, and a pressure of 6 atmospheres was maintained to produce leakages in the pipe. As a interference, noise generated by a moving fluid flow in the pipeline is used. The interference was recorded on the pipeline of the distribution network at a pressure of 7 atmospheres, the volume flow of the coolant averaged more than 1000 liters per hour, which ensured the flow rate of the fluid at which its motion can be considered turbulent. The results of experimental studies of acoustic signals of fluid leakage in the pipeline and interference showed that the main interference power is concentrated in the 0...5 kHz band, and the leakage signal is in the band 5...10 kHz. With the use of bandpass filtering, the ratio of the mean square deviation of the leakage signals and interference increased by more than an order of magnitude for the frequency bands 5...10 kHz and 10...15 kHz, as compared with the absence of filtering. In this case, the ratio of the absolute values of the cumulants of the leakage signal and the interference is much greater than the ratio of the variances, and the ratio of cumulants increases substantially with increasing order. This confirms the sensitivity of high-order cumulants to the presence of a leak.