IP Library › Granted Patent US 12,496,394
Granted Patent B2
US 12,496,394 · App. 17/565,812 · Granted Dec 16, 2025

Systems and methods for occlusion detection in infusion pumps

Inventors: Steven Ward Fischer (Gurnee, IL); Ye Chen (Lake Forest, IL)
Assignees: Baxter International Inc.; Baxter Healthcare SA
A61M5/16831A61M5/142A61M2205/18A61M2205/332A61M2205/502A61M2205/52
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,496,394
App. No.
17/565,812
Granted
Dec 16, 2025
Kind
B2
Abstract

Systems and methods for detecting an occlusion in an infusion device are disclosed. One method includes generating, during an infusion session at predetermined intervals, a feature vector including an initial measurement of a physical force caused by fluid motion through an infusion device at the start of an infusion session, a current measurement of the physical force, a long term change in the measurement of the physical force, and a short term change in the measurement of the physical force. The feature vector is inputted, during the predetermined intervals, into a support vector machine (“SVM”) to output an indication of a presence or an absence of an occlusion in the infusion device. The SVM may be trained using reference data from reference infusion sessions having known consequences regarding the presence or the absence of an occlusion at various times during the reference infusion sessions.

Claims (83)

1 . An infusion pump apparatus for detecting an occlusion, the apparatus comprising:

intravenous (“IV”) tubing having a first end that is fluidly coupled to a patient and a second end that is fluidly coupled to a container;

a pumping mechanism operable with a portion of the IV tubing for providing controlled delivery of a fluid from the container to the patient;

a force sensor coupled to the IV tubing to measure a physical force caused by the fluid;

one or more processors; and

a memory storing instructions that, when executed by the one or more processors, cause the one or more processors to:

receive, via the force sensor and during a start of an infusion session, an initial measurement of a physical force caused by fluid motion through the IV tubing,

identify, based on one or more parameters associated with the infusion session, a trained support vector machine for occlusion detection;

generate, during the infusion session at predetermined intervals, a feature vector comprising:

the initial measurement of the physical force measured by the force sensor,

a current measurement of the physical force measured by the force sensor,

a long term change in the measurement of the physical force, and

a short term change in the measurement of the physical force;

input, during the infusion session at the predetermined intervals, the feature vector into the trained support vector machine to output a detection of a presence or an absence of an occlusion in the IV tubing; and

when the presence of the occlusion is detected, pause or terminate the infusion session.

2 . The apparatus of claim 1 , wherein generating the feature vector further comprises:

receiving, via the force sensor, a current measurement of the physical force;

determining, based on a first prior measurement of the physical force, a long term change in the measurement of the physical force; and

determining, based on a second prior measurement of the physical force, a short term change in the measurement of the physical force.

3 . The apparatus of claim 1 , wherein the instructions, when executed, further cause the one or more processors to train the support vector machine using a training dataset from a plurality of reference infusion sessions sharing the one or more parameters of the infusion session.

4 . The apparatus of claim 3 , wherein training the support vector machine further comprises:

receiving, for each of the plurality of reference infusion sessions:

one or more reference feature vectors corresponding to one or more times during the respective infusion session, wherein a given reference feature vector comprises:

a reference initial measurement of the physical force at the start of the respective reference infusion session,

a reference current measurement of the physical force at a given time of the one or more times during the reference infusion session,

a reference long term change in the measurement of the physical force, and

a reference short term change in the measurement of the physical force, and

an indication of a presence or an absence of an occlusion at the one or more times during the respective infusion session;

associating, for each of the plurality of reference infusion sessions, the one or more feature vectors with either the indication of the presence of the occlusion or the indication of the absence of the occlusion;

determining, based on an optimized margin, a decision boundary for the training dataset, wherein the decision boundary divides the reference feature vectors associated with the indication of the presence of the occlusion from the reference feature vectors associated with the indication of the absence of the occlusion; and

generating, based on the decision boundary, the trained support vector machine.

5 . The apparatus of claim 1 , wherein the long term change in the measurement of the physical force includes a slope value represented by the current measurement of the physical force over a measurement of the physical force at a distant past.

6 . The apparatus of claim 5 , wherein the distant past includes one or more measurements of the physical force recorded at one to two minutes previously.

7 . The apparatus of claim 1 , wherein the long term change in the measurement of the physical force includes a slope value that is equal to the current measurement of the physical force minus a measurement of the physical force at a distant past, divided by a number of measurements recorded between the current measurement of the physical force and the measurement of the physical force at the distant past.

8 . The apparatus of claim 7 , wherein the distant past includes one or more measurements of the physical force recorded at one to two minutes previously.

9 . The apparatus of claim 1 , wherein the short term change in the measurement of the physical force includes a slope value represented by the current measurement of the physical force over a measurement of the physical force at a near past.

10 . The apparatus of claim 9 , wherein the near past includes one or more measurements of the physical force recorded at 50 milliseconds to 100 milliseconds previously.

11 . The apparatus of claim 1 , wherein the short term change in the measurement of the physical force includes a slope value that is equal to the current measurement of the physical force minus a measurement of the physical force at a near past, divided by a number of measurements recorded between the current measurement of the physical force and the measurement of the physical force at the near past.

12 . The apparatus of claim 11 , wherein the near past includes one or more measurements of the physical force recorded at 50 milliseconds to 100 milliseconds previously.

13 . The apparatus of claim 1 , wherein the one or more processors are configured to cause an alert or an alarm to be displayed on a user interface when the presence of the occlusion is detected.

14 . An infusion system for detecting an occlusion, the system comprising:

a force sensor coupled to IV tubing to measure a physical force caused by a fluid within the IV tubing, the IV tubing having a first end that is fluidly coupled to a patient and a second end that is fluidly coupled to a container; and

an infusion device communicatively coupled to the force sensor, the infusion device including:

a pumping mechanism operable with a portion of the IV tubing for providing controlled delivery of a fluid from the container to the patient,

a user interface,

one or more processors, and

a memory storing a plurality of trained support vector machines and instructions that, when executed by the one or more processors, cause the one or more processors to:

receive, via the force sensor and during a start of an infusion session, an initial measurement of a physical force caused by fluid motion through the IV tubing,

identify, based on one or more parameters associated with the infusion session, a trained support vector machine for occlusion detection among the plurality of the trained support vector machines;

generate, during the infusion session at predetermined intervals, a feature vector comprising:

the initial measurement of the physical force measured by the force sensor,

a current measurement of the physical force measured by the force sensor,

a long term change in the measurement of the physical force, and

a short term change in the measurement of the physical force;

input, during the infusion session at the predetermined intervals, the feature vector into the trained support vector machine to output a detection of a presence or an absence of an occlusion in the IV tubing; and

when the presence of the occlusion is detected, cause an alert or an alarm to be displayed on the user interface that is indicative of the occlusion;

wherein the instructions, when executed, further cause the one or more processors to train the support vector machine using a training dataset from a plurality of reference infusion sessions sharing the one or more parameters of the infusion session, wherein the training the support vector machine further comprises:

receiving, for each of the plurality of reference infusion sessions:

one or more reference feature vectors corresponding to one or more times during the respective infusion session, wherein a given reference feature vector comprises:

a reference initial measurement of the physical force at the start of the respective reference infusion session,

a reference current measurement of the physical force at a given time of the one or more times during the reference infusion session,

a reference long term change in the measurement of the physical force, and

a reference short term change in the measurement of the physical force, and

an indication of a presence or an absence of an occlusion at the one or more times during the respective infusion session;

associating, for each of the plurality of reference infusion sessions, the one or more feature vectors with either the indication of the presence of the occlusion or the indication of the absence of the occlusion;

determining, based on an optimized margin, a decision boundary for the training dataset, wherein the decision boundary divides the reference feature vectors associated with the indication of the presence of the occlusion from the reference feature vectors associated with the indication of the absence of the occlusion; and

generating, based on the decision boundary, the trained support vector machine.

15 . The system of claim 14 , wherein the one or more processors are configured to pause or terminate the infusion session when the presence of the occlusion is detected.

16 . The system of claim 14 , wherein the force sensor is integrally formed or fluidly connected to the IV tubing.

17 . The system of claim 14 , wherein generating the feature vector further comprises:

receiving, via the force sensor, a current measurement of the physical force;

determining, based on a first prior measurement of the physical force, a long term change in the measurement of the physical force; and

determining, based on a second prior measurement of the physical force, a short term change in the measurement of the physical force.

18 . An infusion pump method for detecting an occlusion, the method comprising:

receiving, in a processor of an infusion device from a force sensor and during a start of an infusion session, an initial measurement of a physical force caused by fluid motion through IV tubing having a first end that is fluidly coupled to a patient and a second end that is fluidly coupled to a container;

identifying, via the processor based on one or more parameters associated with the infusion session, a trained support vector machine for occlusion detection;

generating, via the processor during the infusion session at predetermined intervals, a feature vector comprising:

the initial measurement of the physical force measured by the force sensor,

a current measurement of the physical force measured by the force sensor,

a long term change in the measurement of the physical force, and

a short term change in the measurement of the physical force;

inputting, via the processor during the infusion session at the predetermined intervals, the feature vector into the trained support vector machine to output a detection of a presence or an absence of an occlusion in the IV tubing; and

when the presence of the occlusion is detected, pausing or terminating the infusion session via the processor.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 7, 2022
From: FISCHER, STEVEN WARD; CHEN, YE
To: BAXTER INTERNATIONAL INC.; BAXTER HEALTHCARE SA
Reel/Frame 058583/0919 →
Continuity (2)
Provisional Application 63132183 · Dec 30, 2020
Related Publication 20220203026A1 · Jun 30, 2022
References Cited (1)
US 10792420B2 · Carothers · 2020 [cited by examiner]