# US 2025/0148367 A1 · Systems and methods for training and application of machine learning algorithms for microscope images

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

- Number: US 2025/0148367 A1
- Document ID: US20250148367A1
- Status: Published application
- Kind: A1
- Application number: 18/838,654
- Filed: 2023-02-14
- Published: 2025-05-08
- Earliest claimed: 2022-02-16
- Applicants: Leica Instruments (Singapore) Pte. Ltd.
- Inventors: Milos SORMAZ; Jochen SCHWEIZER; George THEMELIS
- Web page: https://patentgrep.com/patent/US20250148367A1

## CPC

- G06N 20/00: Machine learning
- G16H 30/20: for handling medical images, e.g. DICOM, HL7 or PACS
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA

## Abstract

A system for training of a machine-learning algorithm includes one or more processors and one or more storage devices. The system is configured to receive training data. The training data includes images showing a tissue. The system is further configured to adjust the machine-learning algorithm to obtain a trained machine-learning algorithm based on the training data, such that the trained machine-learning algorithm generates instruction data for at least a part of the tissue shown in the images. The instruction data is indicative of an action to be performed on the tissue. The system is further configured to provide the trained machine-learning algorithm.

## Claims

### Claim 1

A system for training of a machine-learning algorithm, the system comprising:
one or more processors and one or more storage devices, wherein the system is configured to:
receive training data, the training data comprising images showing a tissue;
adjust the machine-learning algorithm to obtain a trained machine-learning algorithm based on the training data, such that the trained machine-learning algorithm generates instruction data for at least a part of the tissue shown in the images, wherein the instruction data is indicative of an action to be performed on the tissue during a surgery, and wherein the instruction data comprises a resection boundary indicative of an area of the tissue in which the action is to be performed on the tissue; and
provide the trained machine-learning algorithm.

### Claim 2 (depends on 1)

The system of claim 1, wherein the training data further comprises patient data, the patient data including information about patients correlated to the tissue shown in the images.

### Claim 3 (depends on 2)

The system of claim 2, wherein the patient data comprise at least one of following parameters for the patients: age, gender, type and dosage of oncological treatment, remaining life time, blood pressure, cholesterol, tumour recurrence, or loss of at least part of one or more cognitive functions.

### Claim 4 (depends on 2)

The system of claim 2, wherein the machine-learning algorithm is adjusted such that the trained machine-learning algorithm generates the instruction data based at least in part on at least parts of the patient data, wherein the patient data is to be provided to the trained machine-learning algorithm as input data for application of the trained machine-learning algorithm.

### Claim 5 (depends on 1)

The system of claim 1, wherein the training data further comprise annotations on the images,
wherein the annotations are indicative of at least one of: classes or types of the tissue shown in the images, and the action to be performed or actions having been performed on the tissue.

### Claim 6 (depends on 1)

The system of claim 1, wherein the trained machine-learning algorithm is obtained based on supervised learning.

### Claim 7 (depends on 5)

The system of claim 5, wherein the machine-learning algorithm is based on classification, wherein the classification is based on at least a part of the annotations on the images.

### Claim 8 (depends on 6)

The system of claim 6, wherein the machine learning algorithm is based on regression.

### Claim 9 (depends on 1)

The system of claim 1, wherein the trained machine-learning algorithm is obtained based on unsupervised learning.

### Claim 10 (depends on 1)

The system of claim 1, wherein the images showing the tissue comprise microscope images from a surgical microscope obtained during a surgery.

### Claim 11 (depends on 1)

The system of claim 1, wherein the training data further comprise at least one of: radiology images or scans, ultrasound images, endoscope images, or neuromonitoring information.

### Claim 12 (depends on 10)

The system of claim 10, wherein the images showing the tissue further comprise at least one of: radiology images, ultrasound images, or endoscope images, each showing the tissue shown in the microscope images, and having a same field of view as the corresponding microscope images.

### Claim 13

(canceled)

### Claim 14 (depends on 1)

The system of claim 1, further configured to determine the boundary out of multiple boundaries with different values of a parameter relating to the tissue or a patient from whom the tissue is from, such that a value of the parameter for the boundary is maximized.

### Claim 15 (depends on 1)

The system of claim 1, wherein the instruction data comprise different indicators for different actions to be performed on the tissue.

### Claim 16 (depends on 1)

The system of claim 1, wherein the machine-learning algorithm is adjusted such that the trained machine-learning algorithm generates the instruction data being overlaid to the microscope images.

### Claim 17

A computer-implemented method for training of a machine-learning algorithm, the method comprising:
receiving training data, the training data comprising images showing a tissue;
adjusting the machine-learning algorithm to obtain a trained machine-learning algorithm based on the training data, such that the trained machine-learning algorithm generates instruction data for at least a part of the tissue shown in the images, wherein the instruction data is indicative of an action to be performed on the tissue during a surgery, and wherein the instruction data comprises a resection boundary indicative of an area of the tissue in which the action is to be performed on the tissue; and
providing the trained machine-learning algorithm.

### Claim 18

A trained machine-learning algorithm, trained by:
receiving training data, the training data comprising: images showing tissue; and
adjusting the machine learning algorithm based on the training data, such that the machine-learning algorithm generates instruction data for at least part of the tissue shown in the images, wherein the instruction data is indicative of an action to be performed on the tissue during a surgery, and wherein the instruction data comprises a resection boundary indicative of an area of the tissue in which the action is to be performed on the tissue.

### Claim 19

A system for providing instruction data for a tissue shown in a microscope image, the system comprising:
one or more processors, and one or more storage devices, wherein the system is configured to:
receive input data, the input data comprising: a microscope image from a surgical microscope obtained during a surgery, the microscope image showing a tissue of a patient,
generate instruction data for at least a part of the tissue shown in the microscope image, by applying a machine-learning algorithm, wherein the instruction data is indicative of an action to be performed on the tissue during a surgery, and wherein the instruction data comprises a resection boundary indicative of an area of the tissue in which the action is to be performed on the tissue; and
provide output data, the output data comprising the instruction data for the microscope image.

### Claim 20 (depends on 19)

The system of claim 19, wherein the input data further comprise patient data, wherein the instruction data is determined based at least in part on at least parts of the patient data.

### Claim 21 (depends on 19)

The system of claim 19, wherein the output data comprise the microscope image with the instruction data being overlaid.

### Claim 22

(canceled)

### Claim 23 (depends on 19)

A surgical microscopy system, comprising a surgical microscope, an image sensor, and the system of claim 19.

### Claim 24

A computer-implemented method comprising:
receiving input data, the input data comprising a microscope image from a surgical microscope obtained during a surgery, the microscope image showing a tissue of a patient,
generating instruction data for at least part of the tissue shown in the microscope image, by applying a machine-learning algorithm, wherein the instruction data is indicative of an action to be performed on the tissue during a surgery, and wherein the instruction data comprises a resection boundary indicative of an area of the tissue in which the action is to be performed on the tissue; and
providing output data, the output data comprising: the instruction data for the microscope image.

### Claim 25

A method for providing an image and instruction data to a user using a surgical microscope, the method comprising:
illuminating a tissue of a patient,
capturing a microscope image of the tissue, the microscope image showing the tissue,
