Device for image-based cell classification, method therefor and use thereof

Patent metadata

Field	Value
Publication number	US-11610416-B2
Application number	US-202017018033-A
Country	US
Kind code	B2
Filing date	Sep 11, 2020
Priority date	Sep 23, 2019
Publication date	Mar 21, 2023
Grant date	Mar 21, 2023

Abstract

A classifying device for classifying cells in real-time, comprising: as alignment unit configured to align a cell to be classified along the cell's major axis; and a classifying unit configured to classify the aligned cell using a multilayer perceptron, MLP; wherein the MLP classifies the aligned cell based on one or more images of the aligned cell. By executing the classifying device, an improved and efficient cell classification in real-time based on cell images can be provided, while labelling of the cells to be classified can be avoided.

First claim

The invention claimed is:

1. A classifying device for classifying cells in real-time, the classifying device comprising:
an alignment unit configured to align a cell to be classified, along the cell's major axis, prior to being imaged by an imaging unit, wherein the alignment unit is further configured to deform and extend the cell along the cell's major axis during alignment; and
a classifying unit configured to classify the so-aligned deformed and extended cell using a multilayer perceptron, MLP, wherein the MLP has been trained based on a plurality of images of- aligned deformed and extended cells;
wherein the MLP classifies the so-aligned deformed and extended cell based on one or more images of the- aligned deformed and extended cell.

2. The classifying device according to claim 1 , wherein a cross-sectional diameter of the alignment unit is less than 100 μm.

3. The classifying device according to claim 2 , a cross-sectional diameter of the alignment unit is less than 50 μm.

4. The classifying device according to claim 1 , further comprising the imaging unit configured to obtain the one or more images of the- aligned deformed and extended cell to be classified.

5. The classifying device according to claim 1 , wherein the alignment unit comprises a microfluidic channel through which the cell to be classified flows.

6. The classifying device according to claim 1 , wherein the MLP has been trained on an input-output pair, the input being the plurality of images of the- aligned deformed and extended cells and the output being corresponding classification results, wherein the one or more images of the- aligned deformed and extended cell used for classification are images of a marker-free cell.

7. The classifying device according to claim 1 , wherein the one or more images of the-aligned deformed and extended cell used for classification are bright-field images.

8. A sorting device comprising the classifying device of claim 1 , further comprising a sorting unit configured to sort the classified cell to a target outlet or to a default outlet based on a classification result output by the classifying device.

9. The classifying device according to claim 5 , wherein the alignment unit corresponds to a narrowing portion of the microfluidic channel.

10. The classifying device according to claim 1 , wherein the alignment unit is configured to cause the cells flowing therein to align, deform and extend along their major axis.

11. A classifying method for classifying cells in real-time, comprising the steps of:
aligning a cell to be classified, along the cell's major axis, prior to being imaged by an imaging unit;
deforming and extending the cell along the cell's major axis during aligning, prior to being imaged by an imaging unit and classifying the so- aligned deformed and extended cell using a multilayer perceptron, MLP;
wherein the MLP classifies the so- aligned deformed and extended cell based on one or more images of the so- aligned deformed and extended cell, wherein the MLP has been trained based on a plurality of images of- aligned deformed and extended cells.

12. The classifying method according to claim 11 , wherein the cell to be classified flows through a microfluidic channel for alignment.

13. The classifying method according to claim 12 , wherein the diameter of the cell to be classified is 30% to 90% of the minimum width of the microfluidic channel.

14. The classifying method according to claim 11 , further comprising the step of obtaining, by the imaging unit, the one or more images of the so- aligned deformed and extended cell to be classified.

15. The classifying method according to claim 11 , wherein the MLP has been trained on an input-output pair, the input being the plurality of images of- aligned deformed and extended cells and the output being corresponding classification results.

16. The classifying method according to claim 11 , wherein the cell to be classified is a marker-free cell and/or wherein the one or more images of the cell used for classification are bright-field images.

17. A sorting method comprising the steps of the classifying method of claim 11 , further comprising the step of sorting the classified cell to a target outlet or to a default outlet based on a classification result of the classifying method.

Assignees

• Max Planck Gesellschaft
• Univ Dresden Tech
• Max Planck Ges Zur Foerderung Der Wissensc

Inventors

• Herbig Maik
• Nawaz Ahmad Ahsan
• Nötzel Martin
• Guck Jochen

Classifications

• G06V2201/03

  Recognition of patterns in medical or anatomical images · CPC title

• G06F18/24 Primary

  Classification techniques · CPC title

• G06V10/764

  using classification, e.g. of video objects · CPC title

• G06V20/698 Primary

  Matching; Classification · CPC title

• B01L2200/0652

  Sorting or classification of particles or molecules · CPC title