Machine learning fast aberration measuring system and method based on wave-front sensor

Abstract

The invention discloses a machine learning fast aberration measuring system and method based on a wave-front sensor. The method comprises the steps of measuring a distortion phase distribution of an experimental sample using a Mach-Zehnder interferometer by phase shifting interferometry, processing the distortion phase distribution to obtain aberration prior information, and generating a phase distribution set using the aberration prior information; sequentially loading the phase distribution set to a spatial light modulator, then scanning undistorted fluorescence samples to obtain a distortion light intensity distribution pattern, and establishing a light intensity distribution set; inputting the light intensity distribution set and the phase distribution set into a machine learning modelfor training; scanning an experimental sample to be tested to obtain a new light intensity distribution pattern, and inputting the new light intensity distribution pattern into the trained machine learning model to obtain predicted aberration information, thus obtaining a corrected phase; and loading the corrected phase onto the spatial light modulator to form an aberration measurement. By combining a machine learning algorithm with a wave-front sensing technology, the system and the method improve the optical aberration measurement speed, achieve fast aberration measurement and correction inthe wide field fluorescence microscopic imaging, and have good application prospect.

Description

technical field [0001] The invention belongs to an aberration measurement system and method in the field of fast measurement of optical wavefront distortion, in particular to a machine learning fast aberration measurement system and method based on wavefront sensors, which can be applied to high-resolution optical microscopic imaging and optical Wavefront distortion measurement provides a new technical means for high-speed and high-resolution imaging in the field of biomedicine. Background technique [0002] In biomedical research, optical microscopic imaging technology is often used to obtain deep information of biological tissues. However, due to the inhomogeneity of the refractive index of biological tissues, the production accuracy error of optical components, and the mismatch of the refractive index between the media, optical aberrations often occur during the imaging process, which causes wavefront distortion and seriously affects the focus and alignment of the spot. ...

AI Technical Summary

Problems solved by technology

[0004] However, direct measurement methods are usually only able to correct low-order aberrations, and ideal correction effects cannot be obtained in deeper biological tissues; in addition, the wavefront measurement accuracy of direct measurement methods is limited, taking the Shack-Hartmann wavefront sensor as an example , its detection accuracy depends on the positioning accuracy of the centroid, and additional algorithms are often required to improve the positioning accuracy of the centroid; while the indirect measurement method can obtain ideal results, but it consumes a lot of time, and cannot take into account time cost and imaging quality, which is not conducive to Real-time imaging detection in living organisms restricts its practical application in the field of biomedicine

[0005] The patent with application number 201811314921.X in the prior art involves a high-speed and high-resolution scanning microscopy imaging system and method based on machine learning, but its ability to detect aberrations is limited

Images