Granted Patent B2

US 12,646,610 · App. 18/351,759 · Granted Jun 2, 2026

Inventors: Jeffrey Roh (Seattle, WA); Justin Esterberg (Mesa, AZ); John Cronin (Jericho, VT); Seth Cronin (Essex Junction, VT); Michael D'Andrea (Burlington, VT)

Assignee: IX Innovation LLC

G16H40/40G06N20/00G16H10/60G16H20/40

Abstract

Claims (84)

1 . A computer-implemented method for surgical alert monitoring, the method comprising:

predicting, using a machine learning module, one or more adverse surgical events prior to a surgery,

wherein the machine learning module is trained based on stored surgical data and stored electronic health records;

receiving sensor readings during the surgery;

extracting, from the sensor readings, a feature vector indicative of one or more vital signs of a patient undergoing the surgery;

identifying, using the machine learning module, one or more patterns in the sensor readings based on the feature vector;

determining at least one alert threshold for a physiological condition of the patient based on the identified one or more patterns and the predicted one or more adverse surgical events; and

intra-operatively avoiding the predicted one or more adverse surgical events by:

transmitting a first alert to one or more robotic surgical controls to adjust at least one surgical step of the surgery to avoid the predicted one or more adverse surgical events; and

responsive to determining that the physiological condition of the patient has met the at least one alert threshold, transmitting a second alert to the one or more robotic surgical controls to terminate the surgery.

2 . The computer-implemented method of claim 1 , wherein the received sensor readings include one or more physiological parameters of the patient, and

wherein determining the at least one alert threshold comprises:

correlating the one or more physiological parameters to the predicted one or more adverse surgical events.

3 . The computer-implemented method of claim 1 , comprising:

determining a predicted time at which the physiological condition of the patient will meet the at least one alert threshold based on a rate of change of the received sensor readings,

wherein transmitting the first alert is performed prior to the predicted time.

4 . The computer-implemented method of claim 1 , wherein the received sensor readings include one or more physiological parameters of the patient, and

wherein determining the at least one alert threshold comprises:

comparing the one or more physiological parameters to the predicted one or more adverse surgical events using at least one of convolution, auto-correlation, or cross-correlation.

5 . The computer-implemented method of claim 1 , wherein the received sensor readings include one or more physiological parameters of the patient, and

wherein determining the at least one alert threshold comprises:

comparing the one or more physiological parameters to the stored electronic health records using regression analysis.

6 . The computer-implemented method of claim 1 , comprising:

generating a patient-specific monitoring alert plan based on patient data collected prior to the surgery; and

in response to receiving the sensor readings, adjusting the patient-specific monitoring alert plan.

7 . The computer-implemented method of claim 1 , wherein the received sensor readings include one or more physiological parameters of the patient, and

wherein the method comprises:

correlating the one or more physiological parameters to adverse event data in the stored electronic health records,

wherein the adverse event data is associated with prior surgeries matching the surgery.

8 . A computer system for surgical alert monitoring, the computer system comprising:

one or more computer processors; and

a non-transitory computer-readable storage medium storing computer instructions, which when executed by the one or more computer processors cause the computer system to:

predict, using a machine learning module, one or more adverse surgical events prior to a surgery,

wherein the machine learning module is trained based on stored surgical data and stored electronic health records;

receive sensor readings during the surgery;

extract, from the sensor readings, a feature vector indicative of one or more vital signs of a patient undergoing the surgery;

identify, using the machine learning module, one or more patterns in the sensor readings based on the feature vector;

determine at least one alert threshold for a physiological condition of the patient based on the identified one or more patterns and the predicted one or more adverse surgical events; and

intra-operatively avoid the predicted one or more adverse surgical events by:

transmitting a first alert to one or more robotic surgical controls to adjust at least one surgical step of the surgery to avoid the predicted one or more adverse surgical events; and

responsive to determining that the physiological condition of the patient has met the at least one alert threshold, transmitting a second alert to the one or more robotic surgical controls to terminate the surgery.

9 . The computer system of claim 8 , wherein the received sensor readings include one or more physiological parameters of the patient, and

wherein the computer instructions to determine the at least one alert threshold cause the computer system to:

correlate the one or more physiological parameters to the predicted one or more adverse surgical events.

10 . The computer system of claim 8 , wherein the computer instructions cause the computer system to:

determine a predicted time at which the physiological condition of the patient will meet the at least one alert threshold based on a rate of change of the received sensor readings,

wherein transmitting the first alert is performed prior to the predicted time.

11 . The computer system of claim 8 , wherein the received sensor readings include one or more physiological parameters of the patient, and

wherein the computer instructions to determine the at least one alert threshold cause the computer system to:

compare the one or more physiological parameters to the predicted one or more adverse surgical events using at least one of convolution, auto-correlation, or cross-correlation.

12 . The computer system of claim 8 , wherein the received sensor readings include one or more physiological parameters of the patient, and

wherein the computer instructions to determine the at least one alert threshold cause the computer system to:

compare the one or more physiological parameters to the stored electronic health records using regression analysis.

13 . The computer system of claim 8 , wherein the computer instructions cause the computer system to:

generate a patient-specific monitoring alert plan based on patient data collected prior to the surgery; and

in response to receiving the sensor readings, adjust the patient-specific monitoring alert plan.

14 . The computer system of claim 8 , wherein the received sensor readings include one or more physiological parameters of the patient, and

wherein the computer instructions cause the computer system to:

correlate the one or more physiological parameters to adverse event data in the stored electronic health records,

wherein the adverse event data is associated with prior surgeries matching the surgery.

15 . A non-transitory computer-readable storage medium storing computer instructions, which when executed by one or more computer processors cause the one or more computer processors to:

predict, using a machine learning module, one or more adverse surgical events prior to a surgery, wherein the machine learning module is trained based on stored surgical data and stored electronic health records;

receive sensor readings during the surgery;

extract, from the sensor readings, a feature vector indicative of one or more vital signs of a patient undergoing the surgery;

identify, using the machine learning module, one or more patterns in the sensor readings based on the feature vector;

determine at least one alert threshold for a physiological condition of the patient based on the identified one or more patterns and the predicted one or more adverse surgical events; and

intra-operatively avoid the predicted one or more adverse surgical events by:

transmitting a first alert to one or more robotic surgical controls to adjust at least one surgical step of the surgery to avoid the predicted one or more adverse surgical events; and

responsive to determining that the physiological condition of the patient has met the at least one alert threshold, transmitting a second alert to the one or more robotic surgical controls to terminate the surgery.

16 . The non-transitory computer-readable storage medium of claim 15 , wherein the received sensor readings include one or more physiological parameters of the patient, and

wherein the computer instructions to determine the at least one alert threshold cause the one or more computer processors to:

correlate the one or more physiological parameters to the predicted one or more adverse surgical events.

17 . The non-transitory computer-readable storage medium of claim 15 , wherein the computer instructions cause the one or more computer processors to:

determine a predicted time at which the physiological condition of the patient will meet the at least one alert threshold based on a rate of change of the received sensor readings,

wherein transmitting the first alert is performed prior to the predicted time.

18 . The non-transitory computer-readable storage medium of claim 15 , wherein the received sensor readings include one or more physiological parameters of the patient, and

wherein the computer instructions to determine the at least one alert threshold cause the one or more computer processors to:

compare the one or more physiological parameters to the predicted one or more adverse surgical events using at least one of convolution, auto-correlation, or cross-correlation.

19 . The non-transitory computer-readable storage medium of claim 15 , wherein the received sensor readings include one or more physiological parameters of the patient, and

wherein the computer instructions to determine the at least one alert threshold cause the one or more computer processors to:

compare the one or more physiological parameters to the stored electronic health records using regression analysis.

20 . The non-transitory computer-readable storage medium of claim 15 , wherein the computer instructions cause the one or more computer processors to:

generate a patient-specific monitoring alert plan based on patient data collected prior to the surgery; and

in response to receiving the sensor readings, adjust the patient-specific monitoring alert plan.

Continuity (2)

References Cited (16)