# US 2026/0229327 A1 · Method for health data and consent management

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

- Number: US 2026/0229327 A1
- Document ID: US20260229327A1
- Status: Published application
- Kind: A1
- Application number: 19/552,086
- Filed: 2026-02-27
- Published: 2026-08-06
- Earliest claimed: 2022-09-30
- Applicants: Cilag GmbH International
- Inventors: Frederick E. Shelton, IV; Matthew David Cowperthwait; Christopher Waid; Taylor W. Aronhalt; Jason L. Harris; Kevin M. Fiebig
- Web page: https://patentgrep.com/patent/US20260229327A1

## CPC

- G16H 10/60: for patient-specific data, e.g. for electronic patient records
- G16H 40/40: for the management of medical equipment or devices, e.g. scheduling maintenance or upgrades
- G16H 40/67: for remote operation
- G16H 50/20: for computer-aided diagnosis, e.g. based on medical expert systems
- G16H 80/00: ICT specially adapted for facilitating communication between medical practitioners or patients, e.g. for collaborative diagnosis, therapy or health monitoring
- G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA

## Abstract

Patient health data and consent data may be managed. Conditional consent by a patient may be provided for utilizing their private health data in one or more of the following: the adaptation of procedures, instrument operation, system configuration, display of results and/or plans for device usage. The patient granted consent may be associated with one or more condition(s), stipulation(s), and/or limitations of the scope, duration, and/or usage of their health data. A health data management system may check that patient data is only being used in accordance with the consent data (e.g., for the agreed upon purpose(s) and for the agreed upon time).

## Claims

### Claim at position 1

1-18. (canceled)

### Claim 19

A method comprising:
detecting a change in consent associated with a patient;
identifying private data associated with the change in consent;
identifying input data used to train a machine learning model, wherein the input data comprises the private data;
determining an impact score associated with removing the private data from the input data used to train the machine learning model;
generating modified input data based on comparing the impact score to a threshold, wherein:
in response to the impact score satisfying the threshold, replacing a portion of the input data that comprises the private data with replacement data; and
in response to the impact score not satisfying the threshold, removing the portion of the private data from the input data; and
adjusting the machine learning model based on the modified input data.

### Claim 20 (depends on 19)

The method of claim 19, further comprising:
identifying a quantity of model parameters impacted by removing the private data; and
for the impacted model parameters, determining a magnitude of change in at least one model weight associated with each impacted model parameter, wherein the impact score is determined based at least on the magnitude of change in the at least one model weight associated with each impacted model parameter.

### Claim 21 (depends on 19)

The method of claim 19, further comprising predicting a change in an output of the machine learning model based on removing the private data from the input data, wherein determining the impact score is based at least on the predicted change in the output.

### Claim 22 (depends on 19)

The method of claim 19, further comprising:
identifying model parameters impacted by removing the private data from the input data; and
determining an extent to which the model parameters are impacted by the removal of the private data, wherein the impact score is determined based at least on the extent.

### Claim 23 (depends on 19)

The method of claim 19, further comprising:
identifying, based on removing the private data from the input data used to train the machine learning model, at least one of: a number of affected nodes, a number of layers in which each affected node is located, a number of connections related to each affected node, or a number of weights associated with each connection; and
determining the impact score further based on the identifying.

### Claim 24 (depends on 19)

The method of claim 19, wherein the replacement data comprises synthetic data generated based on the private data.

### Claim 25 (depends on 19)

The method of claim 19, further comprising:
identifying an impacted model parameter associated with the impact score; and
applying a weight adjustment to a weight associated with the impacted model parameter, wherein the weight adjustment reduces the impact score.

### Claim 26 (depends on 19)

The method of claim 19, wherein the threshold is determined based on at least one of:
a number of affected nodes, a number of layers in which each affected node is located, a number of connections related to each affected node, or a number of weights associated with each connection.

### Claim 27

A surgical device comprising:
a processor configured to:
detect a change in consent associated with a patient;
identify private data associated with the change in consent;
identify input data used to train a machine learning model, wherein the input data comprises the private data;
determine an impact score associated with removing the private data from the input data used to train the machine learning model;
generate modified input data based on a comparison of the impact score to a threshold, wherein:
in response to the impact score satisfying the threshold, replace a portion of the input data that comprises the private data with replacement data;
in response to the impact score not satisfying the threshold, remove the portion of the private data from the input data; and
adjust the machine learning model based on the modified input data.

### Claim 28 (depends on 27)

The surgical device of claim 27, wherein the processor is further configured to:
identify a quantity of model parameters impacted by removing the private data; and
for the impacted model parameters, determine a magnitude of change in at least one model weight associated with each impacted model parameter, wherein the impact score is determined based at least on the magnitude of change in the at least one model weight associated with each impacted model parameter.

### Claim 29 (depends on 27)

The surgical device of claim 27, wherein the processor is further configured to predict a change in an output of the machine learning model based on removing the private data from the input data, wherein the impact score is determined based at least on the predicted change in the output.

### Claim 30 (depends on 27)

The surgical device of claim 27, wherein the processor is further configured to:
identify model parameters impacted by removing the private data from the input data; and
determine an extent to which the model parameters are impacted by the removal of the private data, wherein the impact score is determined based at least on the extent.

### Claim 31 (depends on 27)

The surgical device of claim 27, wherein the processor is further configured to:
identify, based on removing the private data from the input data used to train the machine learning model, at least one of: a number of affected nodes, a number of layers in which each affected node is located, a number of connections related to each affected node, or a number of weights associated with each connection; and
determine the impact score further based on the identification.

### Claim 32 (depends on 27)

The surgical device of claim 27, wherein the replacement data comprises synthetic data generated based on the private data.

### Claim 33 (depends on 27)

The surgical device of claim 27, wherein the processor is further configured to:
identify an impacted model parameter associated with the impact score; and
apply a weight adjustment to a weight associated with the impacted model parameter, wherein the weight adjustment reduces the impact score.

### Claim 34 (depends on 27)

The surgical device of claim 27, wherein the threshold is determined based on at least one of: a number of affected nodes, a number of layers in which each affected node is located, a number of connections related to each affected node, or a number of weights associated with each connection.

### Claim 35

A non-transitory, computer-readable medium comprising computer-executable instructions for adjusting a machine learning model, wherein the computer-executable instructions, when executed by a computer system, cause the computer system to:
detect a change in consent associated with a patient;
identify private data associated with the change in consent;
identify input data used to train a machine learning model, wherein the input data comprises the private data;
determine an impact score associated with removing the private data from the input data used to train the machine learning model;
generate modified input data based on a comparison of the impact score to a threshold, wherein:
in response to the impact score satisfying the threshold, replace a portion of the input data that comprises the private data with replacement data;
in response to the impact score not satisfying the threshold, remove the portion of the private data from the input data; and
adjust the machine learning model based on the modified input data.

### Claim 36 (depends on 35)

The non-transitory, computer-readable medium of claim 35, wherein the computer-executable instructions, when executed, further cause the computer system to:
identify a quantity of model parameters impacted by removing the private data; and
for the impacted model parameters, determine a magnitude of change in at least one model weight associated with each impacted model parameter, wherein the impact score is determined based at least on the magnitude of change in the at least one model weight associated with each impacted model parameter.

### Claim 37 (depends on 35)

The non-transitory, computer-readable medium of claim 35, wherein the computer-executable instructions, when executed, further cause the computer system to predict a change in an output of the machine learning model based on removing the private data from the input data, wherein the impact score is determined further based at least on the predicted change in the output.

### Claim 38 (depends on 35)

The non-transitory, computer-readable medium of claim 35, wherein the computer-executable instructions, when executed, further cause the computer system to:
identify model parameters impacted by removing the private data from the input data; and
determine an extent to which the model parameters are impacted by the removal of the private data, wherein the impact score is determined based at least on the extent.
