# US 2025/0157010 A1 · Systems and methods for artifact detection and removal from image data

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

- Number: US 2025/0157010 A1
- Document ID: US20250157010A1
- Status: Published application
- Kind: A1
- Application number: 18/999,793
- Filed: 2024-12-23
- Published: 2025-05-15
- Earliest claimed: 2023-11-10
- Assignees: Insitro, Inc.
- Applicants: Insitro, Inc.
- Inventors: Varun KANWAR; Christopher PROBERT; Benjamin DULKEN; Adelaide WOICIK; Zachary R. MCCAW
- Web page: https://patentgrep.com/patent/US20250157010A1

## CPC

- G06T 5/20: using local operators
- G06T 5/70: Denoising; Smoothing
- G06T 5/77: Retouching; Inpainting; Scratch removal
- G06T 7/0012: {Biomedical image inspection}
- G06T 7/168: involving transform domain methods
- G06T 2207/10056: Microscopic image
- G06T 2207/20024: Filtering details
- G06T 2207/20072: Graph-based image processing
- G06T 2207/20081: Training; Learning
- G06T 2207/20084: Artificial neural networks [ANN]
- G06T 2207/30024: Cell structures in vitro; Tissue sections in vitro
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

A method for filtering out artifacts from a microscopic image of a tissue includes determining a plurality of frequency values corresponding to a plurality of pixels in the microscopic image of the tissue; grouping the plurality of pixels into a plurality of pixel clusters based on the plurality of frequency values corresponding to the plurality of pixels; identifying, from the plurality of pixel clusters, one or more pixel clusters corresponding to one or more artifacts in the microscopic image; and filtering the microscopic image by removing one or more regions in the microscopic image corresponding to the one or more pixel clusters corresponding to the one or more artifacts.

## Claims

### Claim 1

A system for filtering out artifacts from a microscopic image of a tissue, the system comprising one or more processors, memory, and one or more programs stored in the memory for execution by the one or more processors, the one or more programs including instructions that when executed by the one or more processors cause the system to:
determine a plurality of frequency values corresponding to a plurality of pixels in the microscopic image of the tissue;
group the plurality of pixels into a plurality of pixel clusters based on the plurality of frequency values corresponding to the plurality of pixels;
identify, from the plurality of pixel clusters, one or more pixel clusters corresponding to one or more artifacts in the microscopic image; and
filter the microscopic image by removing one or more regions in the microscopic image corresponding to the one or more pixel clusters corresponding to the one or more artifacts.

### Claim 2 (depends on 1)

The system of claim 1, wherein the microscopic image of the tissue comprises any of a whole slide image, a spatial omics image, an immunohistochemistry image, a trichrome image, and a brightfield image of a 2D cell culture.

### Claim 3 (depends on 1)

The system of claim 1, wherein the one or more artifacts comprise: a tissue fold, a pen marking, an air bubble, defocus, or any combination thereof.

### Claim 4 (depends on 1)

The system of claim 1, wherein the one or more programs include instructions that when executed by the one or more processors cause the system to obtain the microscopic image of the tissue by:
reducing a resolution of an original microscopic image of the tissue; and
converting the original microscopic image of the tissue with the reduced resolution into grayscale.

### Claim 5 (depends on 1)

The system of claim 1, wherein determining the plurality of frequency values corresponding to the plurality of pixels in the microscopic image of the tissue comprises computing the Laplacian of the microscopic image of the tissue.

### Claim 6 (depends on 5)

The system of claim 5, wherein the one or more programs include instructions that when executed by the one or more processors cause the system to reduce noise in the plurality of frequency values corresponding to the plurality of pixels in the microscopic image.

### Claim 7 (depends on 6)

The system of claim 6, wherein reducing noise in the plurality of frequency values corresponding to the plurality of pixels in the microscopic image comprises performing a Gaussian smoothing algorithm on the plurality of frequency values corresponding to the plurality of pixels in the microscopic image.

### Claim 8 (depends on 7)

The system of claim 7, wherein one or more parameters of the Gaussian smoothing algorithm are determined based on a resolution of the microscopic image of the tissue, one or more cellular structures in the microscopic image of the tissue, or any combination thereof.

### Claim 9 (depends on 1)

The system of claim 1, wherein grouping the plurality of pixels into the plurality of pixel clusters comprises identifying a plurality of initial pixel clusters via a K-means algorithm.

### Claim 10 (depends on 9)

The system of claim 9, wherein the one or more programs include instructions that when executed by the one or more processors cause the system to assign all pixels in each initial pixel cluster of the plurality of initial pixel clusters to a representative frequency value of the respective initial pixel cluster.

### Claim 11 (depends on 10)

The system of claim 10, wherein the representative frequency value of the respective initial pixel cluster comprises a centroid of the respective initial pixel cluster.

### Claim 12 (depends on 11)

The system of claim 11, wherein the one or more programs include instructions that when executed by the one or more processors cause the system to merge one or more initial pixel clusters of the plurality of initial pixel clusters via a Region Adjacency Graph (RAG).

### Claim 13 (depends on 1)

The system of claim 1, wherein identifying the one or more pixel clusters corresponding to the one or more artifacts in the microscopic image comprises: identifying a foreground portion and a background portion of the microscopic image, wherein the one or more artifacts are located in the background portion of the microscopic image.

### Claim 14 (depends on 13)

The system of claim 13, wherein the foreground portion and the background portion of the microscopic image are identified via a binary thresholding algorithm.

### Claim 15 (depends on 1)

The system of claim 1, wherein the one or more programs include instructions that when executed by the one or more processors cause the system to further filter the microscopic image by removing a region in the microscopic image corresponding to a pixel cluster below a predefined frequency threshold.

### Claim 16 (depends on 1)

The system of claim 1, wherein the one or more programs include instructions that when executed by the one or more processors cause the system to fill one or more holes in the filtered microscopic image.

### Claim 17 (depends on 1)

The system of claim 1, wherein the one or more programs include instructions that when executed by the one or more processors cause the system to input the filtered microscopic image or a representation of the filtered microscopic image into a trained machine learning model.

### Claim 18 (depends on 17)

The system of claim 17, wherein the trained machine learning model is configured to provide an output indicative of a diagnosis, a treatment, an association between phenotypes, an outcome prediction, a subtype classification, an imputed value, or any combination thereof.

### Claim 19

A method for filtering out artifacts from a microscopic image of a tissue, the method comprising:
determine a plurality of frequency values corresponding to a plurality of pixels in the microscopic image of the tissue;
group the plurality of pixels into a plurality of pixel clusters based on the plurality of frequency values corresponding to the plurality of pixels;
identify, from the plurality of pixel clusters, one or more pixel clusters corresponding to one or more artifacts in the microscopic image; and
filter the microscopic image by removing one or more regions in the microscopic image corresponding to the one or more pixel clusters corresponding to the one or more artifacts.

### Claim 20

A non-transitory computer-readable medium storing instructions for filtering out artifacts from a microscopic image of a tissue, wherein the instructions are executable by a system comprising one or more processors to cause the system to:
determine a plurality of frequency values corresponding to a plurality of pixels in the microscopic image of the tissue;
group the plurality of pixels into a plurality of pixel clusters based on the plurality of frequency values corresponding to the plurality of pixels;
identify, from the plurality of pixel clusters, one or more pixel clusters corresponding to one or more artifacts in the microscopic image; and
filter the microscopic image by removing one or more regions in the microscopic image corresponding to the one or more pixel clusters corresponding to the one or more artifacts.
