# US 2024/0102986 A1 · Systems and methods for particle analysis

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2024/0102986 A1
- Document ID: US20240102986A1
- Status: Published application
- Kind: A1
- Application number: 18/477,689
- Filed: 2023-09-29
- Published: 2024-03-28
- Earliest claimed: 2018-08-15
- Assignees: Deepcell, Inc.
- Applicants: Deepcell, Inc.
- Inventors: Mahdokht MASAELI; Mahyar SALEK; Hou-Pu CHOU; Soroush KAHKESHANI; Prashast KHANDELWAL; Syed Tariq SHAFAAT
- Web page: https://patentgrep.com/patent/US20240102986A1

## CPC

- G01N 33/491: {by separating the blood components (G01N15/05 takes precedence)}
- G01N 1/30: Staining; Impregnating {; Fixation; Dehydration; Multistep processes for preparing samples of tissue, cell or nucleic acid material and the like for analysis}
- G01N 15/0227: using imaging; using holography
- G01N 15/1429: Signal processing
- G01N 15/1434: Optical arrangements
- G01N 15/147: {the analysis being performed on a sample stream}
- G01N 15/1433: using image recognition
- G01N 15/149: specially adapted for sorting particles, e.g. by their size or optical properties
- G01N 2015/012: {Red blood cells}
- G01N 2015/016: {White blood cells}
- G01N 2015/018: {Platelets}
- G01N 2015/0288: {Sorting the particles}
- G01N 2015/0294: {Particle shape}
- G01N 2015/1006: {for cytology}
- G01N 2015/1445: {Three-dimensional imaging, imaging in different image planes, e.g. under different angles or at different depths, e.g. by a relative motion of sample and detector, for instance by tomography}
- G01N 2015/1493: {Particle size}
- G01N 2015/1497: {Particle shape}
- G01N: INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES (measuring or testing apparatus or processes other than immunoassay, involving enzymes or microorganisms C12M, C12Q)

## Abstract

The present disclosure provides systems and methods for sorting a cell. The system may comprise a flow channel configured to transport a cell through the channel. The system may comprise an imaging device configured to capture an image of the cell from a plurality of different angles as the cell is transported through the flow channel. The system may comprise a processor configured to analyze the image using a deep learning algorithm to enable sorting of the cell.

## Claims

### Claim at position 1

1-119. (canceled)

### Claim 120

A method, comprising:
using at least one deep learning algorithm to extract features from images of respective unstained cells and to generate classifications of the respective unstained cells using the extracted features; and
using the classification of each respective unstained cell to select a sub-channel, from a plurality of sub-channels, into which that respective unstained cell is to be flowed.

### Claim 121 (depends on 120)

The method of claim 120, further comprising actuating an actuator to cause that respective unstained cell to be flowed into the selected sub-channel.

### Claim 122 (depends on 121)

The method of claim 121, wherein the actuator is selected from the group consisting of a valve, a piezoelectric element, a magnetic element, and a pneumatic element.

### Claim 123 (depends on 120)

The method of claim 120, wherein the at least one deep learning algorithm comprises a neural network.

### Claim 124 (depends on 120)

The method of claim 120, wherein the at least one deep learning algorithm comprises a plurality of deep learning algorithms.

### Claim 125 (depends on 120)

The method of claim 120, wherein the extracted features are selected from the group consisting of cell shape, cell diameter, nuclear shape, nuclear diameter, nuclear texture, nuclear edges, nuclear area, nuclear average intensity, nucleus to cytoplasm ratio, cell texture, cell edges, cell area, cell average intensity, and DNA content.

### Claim 126 (depends on 120)

The method of claim 120, wherein at least some of the respective unstained cells are alive.

### Claim 127 (depends on 120)

The method of claim 120, wherein the at least one deep learning algorithm extracts the features from multiple images of each of the respective unstained cells.

### Claim 128 (depends on 120)

The method of claim 120, further comprising selecting the sub-channel using a response of the respective unstained cell to a biologically active molecule.

### Claim 129 (depends on 120)

The method of claim 120, wherein the sub-channels of the plurality of sub-channels meet at a junction, and wherein the sub-channel for the respective unstained cell is selected before that cell arrives at the junction.

### Claim 130

A system, comprising:
a processor to execute instructions to perform operations comprising:
using at least one deep learning algorithm to extract features from images of respective unstained cells and to generate classifications of the respective unstained cells using the extracted features; and
using the classification of each respective unstained cell to select a sub-channel, from a plurality of sub-channels, into which that respective unstained cell is to be flowed.

### Claim 131 (depends on 130)

The system of claim 130, further comprising an actuator to cause that respective unstained cell to be flowed into the selected sub-channel.

### Claim 132 (depends on 131)

The system of claim 131, wherein the actuator is selected from the group consisting of a valve, a piezoelectric element, a magnetic element, and a pneumatic element.

### Claim 133 (depends on 130)

The system of claim 130, wherein the at least one deep learning algorithm comprises a neural network.

### Claim 134 (depends on 130)

The system of claim 130, wherein the at least one deep learning algorithm comprises a plurality of deep learning algorithms.

### Claim 135 (depends on 130)

The system of claim 130, wherein the extracted features are selected from the group consisting of cell shape, cell diameter, nuclear shape, nuclear diameter, nuclear texture, nuclear edges, nuclear area, nuclear average intensity, nucleus to cytoplasm ratio, cell texture, cell edges, cell area, cell average intensity, and DNA content.

### Claim 136 (depends on 130)

The system of claim 130, wherein the at least one deep learning algorithm extracts the features from multiple images of each of the respective unstained cells.

### Claim 137 (depends on 130)

The system of claim 130, wherein the operations further comprise selecting the sub-channel using a response of the respective unstained cell to a biologically active molecule.

### Claim 138 (depends on 130)

The system of claim 130, wherein the sub-channels of the plurality of sub-channels meet at a junction, and wherein the sub-channel for the respective unstained cell is selected before that cell arrives at the junction.