Purdue Research Foundation

Yale University

Fang Huang

Sheng Liu

Joerg Bewersdorf

Fan Xu

Peiyi Zhang

Donghan Ma

A fluorescent single molecule emitter simultaneously transmits its identity, location, and cellular context through its emission patterns. A deep neural network (DNN) performs multiplexed single-molecule analysis to enable retrieving such information with high accuracy. The DNN can extract three-dimensional molecule location, orientation, and wavefront distortion with precision approaching the theoretical limit of information content of the image which will allow multiplexed measurements through the emission patterns of a single molecule.

Analyzing complex single molecule emission patterns with deep learning

A method of in situ point spread function (PSF) retrieval is disclosed which includes encoding 3D location of molecules into PSFs, receiving molecule-generated images containing PSFs, segmenting the images into sub-PSF, initializing template PSFs from a pupil function, determining a statistical measure of a predetermined function between the sub- and template PSFs, associating each of the sub-PSFs with a template PSF, aligning and averaging the sub-PSF, applying a phase retrieval algorithm to the averaged sub-PSFs to update the pupil function, regenerating the template PSFs, repeating until a difference between a new and a prior generation pupil function is below a predetermined threshold, generating in situ PSFs from the last pupil function, and applying a maximum likelihood estimation algorithm based on the in situ PSFs and the sub-PSF to thereby generate lateral and axial locations of molecules.

Pinpointing single molecule positions with in situ point spread function retrieval for three dimensional nanoscopy of whole cells and tissues

A method of in situ point spread function (PSF) retrieval is disclosed which includes encoding 3D location of molecules into PSFs, receiving molecule-generated images containing PSFs, segmenting the images into sub-PSFs, initializing template PSFs from a pupil function, determining a maximum normalized cross correlation (NCC) coefficient (NCCmax) between the sub- and template PSFs, associating each of the sub-PSFs with a template PSF based on the NCCmax and storing the sub-PSFs in associated bins, aligning and averaging the binned sub-PSFs, applying a phase retrieval algorithm to the averaged sub-PSFs to update the pupil function, regenerating the template PSFs, repeating until a difference between a new and a prior generation pupil function is below a predetermined threshold, generating in situ PSFs from the last pupil function, and applying a maximum likelihood estimation algorithm based on the in situ PSFs and the sub-PSFs to thereby generate lateral and axial locations of molecules.

A system for reducing noise in a camera image is disclosed. The system includes one or more processors, and a camera operatively coupled to the processor, the processors are configured to reduce noise of camera images, the processors are configured to receive input image data from the camera representing pixel data from a plurality of pixels, segment the input image data to a plurality of segments and for each segment establish an initial segment image, pre-correct pixel data by modifying the pixel data to account for voltage offset and gain of each pixel based on a predetermined map of gain and offset, and obtain an estimate of an output image by minimizing a cost function and output and stitch the estimated image to other estimated and outputted image segments, and output a noise reduced image including the stitched estimated images.

Imaging noise reduction system and method

One aspect of the invention provides a method for drift correction to correct a 3D point collection dataset to compensate for drift over time. The method includes: (a) separating the 3D dataset into n segments, wherein n&#x3e;1; (b) for each of the n segments, reconstructing a volume image as a 3D histogram in which a count for each voxel in the histogram equals a number of localization estimates falling within the voxel; (c) performing 3D cross-correlation between pairs of the n segments; (d) identifying a correlation peak in a result of the 3D cross-correlation to determine a shift distance between pairs of the n segments; (e) solving an overdetermined system of shift distances to determine independent shifts; and (f) offsetting positions from a plurality of segments in the 3D point collection dataset with the independent shifts calculated in step (e) to correct for drift.

Systems, methods, and computer-readable media for ultra-high resolution 3D imaging of whole cells

Techniques for processing imaging data contaminated by sensor-dependent noise. An imaging method is described. In the imaging method, imaging data corresponding to an imaged region and acquired by at least first and second sensor elements is obtained. A parameterized model is fitted to the imaging data. The parameterized model includes a first sensor-dependent model of noise generated by the first sensor element in a first portion of the imaging data acquired by the first sensor element, and a second sensor-dependent model of noise generated by a second sensor element in a second portion of the imaging data acquired by the second sensor element. The first sensor-dependent noise model differs, at least in part, from the second sensor-dependent noise model.

Growing community of inventors

West Lafayette, IN, United States of America

Fang Huang