# US 11,896,329 B1 · Robotic arthroscopic surgery with machine learning

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

- Number: US 11,896,329 B1
- Document ID: US11896329B1
- Status: Granted
- Kind: B1
- Application number: 18/157,903
- Filed: 2023-01-23
- Granted: 2024-02-13
- Earliest claimed: 2023-01-23
- Assignees: IX Innovation LLC
- Applicants: IX Innovation LLC
- Inventors: Jeffrey Roh; Mihir Patel; Justin Esterberg; John Cronin; Seth Cronin; Michael John Baker
- Web page: https://patentgrep.com/patent/US11896329B1

## CPC

- A61B 34/30: Surgical robots
- A61B 1/000096: {using artificial intelligence}
- A61B 1/00194: {adapted for three-dimensional imaging}
- A61B 1/317: for bones or joints, e.g. osteoscopes or arthroscopes
- A61B 34/10: Computer-aided planning, simulation or modelling of surgical operations
- A61B 34/25: {User interfaces for surgical systems}
- G16H 20/40: relating to mechanical, radiation or invasive therapies, e.g. surgery, laser therapy, dialysis or acupuncture
- G16H 50/20: for computer-aided diagnosis, e.g. based on medical expert systems
- G16H 50/70: for mining of medical data, e.g. analysing previous cases of other patients
- A61B 2034/101: {Computer-aided simulation of surgical operations}
- A61B 2034/105: {Modelling of the patient, e.g. for ligaments or bones}
- A61B 2034/107: {Visualisation of planned trajectories or target regions}
- A61B 2034/252: {indicating steps of a surgical procedure}
- A61B 2034/256: {having a database of accessory information, e.g. including context sensitive help or scientific articles}
- A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D)
- G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA

## Abstract

Methods, apparatuses, and systems for performing machine learning (ML) for robotic arthroscopic surgery are disclosed. The disclosed systems use ML to provide recommendations and methods for automated robotic ankle arthroscopic surgery. Historical patient data is filtered to match particular parameters of a patient. The parameters are correlated to the patient. A robotic surgical system or a surgeon reviews the historical patient data to select or adjust the historical patient data to generate a surgical workflow for a surgical robot to perform the robotic arthroscopic surgery.

## Claims

### Claim 1

A computer-implemented method for performing an arthroscopic surgical procedure on a joint using a robotic surgical system, the method comprising:
acquiring patient data of a patient;
analyzing, using at least one machine learning algorithm, the patient data to determine one or more ligament-attachment joint stabilization steps for the joint, wherein the at least one machine learning algorithm is trained using prior patient data with favorable outcomes;
receiving user input regarding the one or more ligament-attachment joint stabilization steps;
generating a pre-operative score for the one or more ligament-attachment joint stabilization steps based on one or more correlations of the patient data to prior patient data;
in response to the pre-operative score being above a threshold, generating a robotic-enabled surgical plan for the joint based on user input and the one or more ligament-attachment joint stabilization steps,
wherein the robotic-enabled surgical plan includes a sequence of surgical steps with corresponding surgical tools for arthroscopically attaching one or more connectors to at least one ligament of the joint and another structure of the patient to promote stabilization of the joint; and
controlling one or more robotic arms of surgical robot according to the robotic-enabled surgical plan to robotically perform the sequence of surgical steps on the patient.

### Claim 2 (depends on 1)

The computer-implemented method of claim 1, further comprising:
training the at least one machine learning algorithm to generate the robotic-enabled surgical plan based on historical patient data retrieved from at least one database;
selecting a set of the historical patient data from arthroscopic procedures based on one or more patient parameters for an arthroscopic surgical procedure;
determining data in the set of historical patient data correlated to the one or more patient parameters; and
displaying, via an electronic display, the data in the set of historical patient data.

### Claim 3 (depends on 1)

The computer-implemented method of claim 1, including:
receiving user input to adjust historical patient data that is used to train the at least one machine learning algorithm.

### Claim 4 (depends on 3)

The computer-implemented method of claim 3, wherein the user input includes selection of data of the historical patient data and at least one of changing, removing, or modifying the selected data.

### Claim 5 (depends on 1)

The computer-implemented method of claim 1, including:
correlating the patient data with historical patient data to identify one or more matching correlations;
selecting a set of the historical patient data with the identified one or more matching correlations; and
displaying at least a portion of the set of historical patient data for user review.

### Claim 6 (depends on 1)

The computer-implemented method of claim 1, further comprising:
determining the at least one machine learning algorithm trained with historical patient data meets a confidence score;
in response to the confidence score being less than a threshold confidence score,
obtaining additional historical patient data;
retraining the at least one machine learning algorithm using the additional historical patient data until the at least one machine learning algorithm meets the confidence score, wherein the one or more ligament-attachment joint stabilization steps are determined by the at least one machine learning algorithm that has been retrained.

### Claim 7 (depends on 1)

The computer-implemented method of claim 1, further comprising:
determining one or more arthroscopic access locations along the patient enabling the arthroscopic surgical procedure to be performed on the joint using arthroscopic imaging equipment; and
verifying a set arthroscopic surgical instruments is capable of being used by the surgical robot to the sequence of surgical steps on the patient prior to beginning the surgical steps on the patient.

### Claim 8

A system comprising:
one or more processors; and
one or more memories storing instructions that, when executed by the one or more processors, cause the system to perform a process for performing an arthroscopic surgical procedure on a joint using a robotic surgical system, the process comprising:
acquiring patient data of a patient;
analyzing, using at least one machine learning algorithm, the patient data to determine one or more ligament-attachment joint stabilization steps for the joint, wherein the at least one machine learning algorithm is trained using prior patient data with favorable outcomes;
receiving user input regarding the one or more ligament-attachment joint stabilization steps;
generating a pre-operative score for the one or more ligament-attachment joint stabilization steps based on one or more correlations of the patient data to prior patient data;
in response to the pre-operative score being above a threshold, generating a robotic-enabled surgical plan for the joint based on user input and the one or more ligament-attachment joint stabilization steps, wherein the robotic-enabled surgical plan includes a sequence of surgical steps with corresponding surgical tools for arthroscopically attaching one or more connectors to at least one ligament of the joint and another structure of the patient to promote stabilization of the joint; and
controlling one or more robotic arms of surgical robot according to the robotic-enabled surgical plan to robotically perform the sequence of surgical steps on the patient.

### Claim 9 (depends on 8)

The system according to claim 8, wherein the process further comprises:
training the at least one machine learning algorithm to generate the robotic-enabled surgical plan based on historical patient data retrieved from at least one database;
selecting a set of the historical patient data from arthroscopic procedures based on one or more patient parameters for an arthroscopic surgical procedure;
determining data in the set of historical patient data correlated to the one or more patient parameters; and
displaying, via an electronic display, the data in the set of historical patient data.

### Claim 10 (depends on 8)

The system according to claim 8, wherein the process further comprises:
receiving user input to adjust historical patient data that is used to train the at least one machine learning algorithm.

### Claim 11 (depends on 10)

The system according to claim 10, wherein the user input includes selection of data of the historical patient data and at least one of changing, removing, or modifying the selected data.

### Claim 12 (depends on 8)

The system according to claim 8, wherein the process further comprises:
correlating the patient data with historical patient data to identify one or more matching correlations;
selecting a set of the historical patient data with the identified one or more matching correlations; and
displaying at least a portion of the set of historical patient data for user review.

### Claim 13 (depends on 8)

The system according to claim 8, wherein the process further comprises:
determining the at least one machine learning algorithm trained with historical patient data meets a confidence score;
in response to the confidence score being less than a threshold confidence score,
obtaining additional historical patient data;
retraining the at least one machine learning algorithm using the additional historical patient data until the at least one machine learning algorithm meets the confidence score, wherein the one or more ligament-attachment joint stabilization steps are determined by the at least one machine learning algorithm that has been retrained.

### Claim 14 (depends on 8)

The system according to claim 8, wherein the process further comprises:
determining one or more arthroscopic access locations along the patient enabling the arthroscopic surgical procedure to be performed on the joint using arthroscopic imaging equipment; and
verifying a set arthroscopic surgical instruments is capable of being used by the surgical robot to the sequence of surgical steps on the patient prior to beginning the surgical steps on the patient.

### Claim 15

A non-transitory computer-readable medium storing instructions that, when executed by a computing system, cause the computing system to perform operations for performing an arthroscopic surgical procedure on a joint using a robotic surgical system, the operations comprising:
acquiring patient data of a patient;
analyzing, using at least one machine learning algorithm, the patient data to determine one or more ligament-attachment joint stabilization steps for the joint, wherein the at least one machine learning algorithm is trained using prior patient data with favorable outcomes;
receiving user input regarding the one or more ligament-attachment joint stabilization steps;
generating a pre-operative score for the one or more ligament-attachment joint stabilization steps based on one or more correlations of the patient data to prior patient data;
in response to the pre-operative score being above a threshold, generating a robotic-enabled surgical plan for the joint based on user input and the one or more ligament-attachment joint stabilization steps, wherein the robotic-enabled surgical plan includes a sequence of surgical steps with corresponding surgical tools for arthroscopically attaching one or more connectors to at least one ligament of the joint and another structure of the patient to promote stabilization of the joint; and
controlling one or more robotic arms of surgical robot according to the robotic-enabled surgical plan to robotically perform the sequence of surgical steps on the patient.

### Claim 16 (depends on 15)

The non-transitory computer-readable medium of claim 15, wherein the operations further comprise:
training the at least one machine learning algorithm to generate the robotic-enabled surgical plan based on historical patient data retrieved from at least one database;
selecting a set of the historical patient data from arthroscopic procedures based on one or more patient parameters for an arthroscopic surgical procedure;
determining data in the set of historical patient data correlated to the one or more patient parameters; and
displaying, via an electronic display, the data in the set of historical patient data.

### Claim 17 (depends on 15)

The non-transitory computer-readable medium of claim 15, wherein the operations further comprise:
receiving user input to adjust historical patient data that is used to train the at least one machine learning algorithm.

### Claim 18 (depends on 17)

The non-transitory computer-readable medium of claim 17, wherein the user input includes selection of data of the historical patient data and at least one of changing, removing, or modifying the selected data.

### Claim 19 (depends on 15)

The non-transitory computer-readable medium of claim 15, wherein the operations further comprise:
correlating the patient data with historical patient data to identify one or more matching correlations;
selecting a set of the historical patient data with the identified one or more matching correlations; and
displaying at least a portion of the set of historical patient data for user review.

### Claim 20 (depends on 15)

The non-transitory computer-readable medium of claim 15, wherein the operations further comprise:
determining the at least one machine learning algorithm trained with historical patient data meets a confidence score;
in response to the confidence score being less than a threshold confidence score,
obtaining additional historical patient data;
retraining the at least one machine learning algorithm using the additional historical patient data until the at least one machine learning algorithm meets the confidence score, wherein the one or more ligament-attachment joint stabilization steps are determined by the at least one machine learning algorithm that has been retrained.