Automated detection and repositioning of micro-objects in microfluidic devices
Inventors
Du, Fenglei • Lundquist, Paul M. • Tenney, John A. • Lionberger, Troy A.
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12272048-B2
Publication Date
2025-04-08
Expiration Date
Abstract
Methods are provided for the automated detection of micro-objects in a microfluidic device. In addition, methods are provided for repositioning micro-objects in a microfluidic device. In addition, methods are provided for separating micro-objects in a spatial region of the microfluidic device.
Core Innovation
The invention provides a method for determining a number density of micro-objects disposed within a predetermined area of a microfluidic device. The method captures a digital image of a region that contains a micro-object of interest, determines periodic structures in the digital image using a Fourier transform, and generates a filtered image or a differential image by removing the periodic structures from the digital image.
After forming the filtered or differential image, the method identifies a micro-object of interest based on the processed image and calculates the number density of micro-objects in the predetermined area of that region of the microfluidic device. Related refinements describe generating the differential image from a first image and a second image after fluid motion or device shifting, with computational alignment before subtraction so the micro-object current location is reflected in positive-value pixels.
Further refinements include using a discrete Fourier transform to determine periodic structures and filtering pixels in the frequency-domain representation of the filtered or differential image. The processed images are further described as being used for pixel clustering, feature extraction, and micro-object density determination in device regions.
Claims Coverage
The independent claim provides four main inventive features: imaging a microfluidic region, using a Fourier transform to determine periodic structures, generating a filtered image or a differential image by removing periodic structures, and identifying micro-objects and calculating number density from the processed image. Dependent claims further refine differential-image formation, computational alignment, discrete Fourier transform filtering, pixel classification, and the predetermined area.
Fourier-transform periodic-structure filtering for density determination
Determining periodic structures in the digital image using a Fourier transform; generating a filtered image or a differential image by removing the periodic structures from the digital image; identifying a micro-object of interest based on the filtered or differential image; and calculating the number density of micro-objects in the predetermined area of that region of the microfluidic device using the filtered or differential image.
Differential-image formation by subtracting first and second images
Generating the filtered image or a differential image by removing the periodic structures from the digital image using a differential image that is created by subtracting the first image from the second image.
Computational alignment before differential processing
Computationally aligning the first and second images before subtracting them, where positive-value pixels indicate the current location of a micro-object.
Discrete Fourier-transform frequency-domain pixel filtering
Determining periodic structures using a discrete Fourier transform and filtering one or more pixels in the frequency-domain representation of the filtered or differential image.
Pixel classification using light-intensity thresholding
Comparing each pixel’s light intensity value (Li) to a predetermined threshold (Lo) and classifying pixels with Li greater than Lo as positive and pixels with Li less than −1*Lo as negative.
Predetermined area limited to channel, sequestration pen, or trap
Specifying that the pre-determined area of the microfluidic circuit includes a channel, a sequestration pen, or a trap of the microfluidic device.
Overall, the claim coverage centers on using a Fourier transform to determine and remove periodic structures from a microfluidic image, then identifying micro-objects and calculating micro-object number density from the resulting filtered or differential image. Dependent refinements specify differential-image subtraction, computational alignment and interpretation via positive-value pixels, discrete Fourier transform and frequency-domain pixel filtering, intensity-based pixel classification rules, and constraining the counted area to device features including channels, sequestration pens, and traps.
Stated Advantages
Removes periodic structures from a captured digital image to enable identification of micro-objects for number density calculation.
Enables micro-object number density determination in a predetermined area of a microfluidic device using the filtered or differential image.
Supports differential imaging based on images after fluid motion or device shifting, including computational alignment for micro-object current location representation via positive-value pixels.
Supports micro-object identification using pixel clustering, feature extraction, and threshold-based pixel classification using light intensity values.
Documented Applications
Automated micro-object detection and number density determination in microfluidic devices for micro-objects disposed within predetermined areas, including regions such as channels, sequestration pens, and traps.
Micro-object repositioning and separation within microfluidic devices using computed trajectories toward sequestration pens with assignment optimized by minimized travel distance and separation using modified light cages derived from Delaunay triangulation and Voronoi partitioning so cages do not overlap.
Contextual use for identifying cells including mammalian cells, blood cells, hybridoma, cancer cells, and transformed cells as micro-objects.
Interested in licensing this patent?