US 11,445,975 B2 · Methods and systems for improved prediction of fluid responsiveness

Abstract

The present disclosure provides systems and methods for predicting fluid responsiveness. Embodiments include sensors configured to obtain a high-resolution electrocardiogram signal and a computer system connected to the sensors that detects and processes changes in at least one of length, amplitude, slope, area, depth, and height of at least one of P, Q, R, S, T, and U complex of the electrocardiogram signal caused by the influence of physiological variables on each other to create a prognostic index. The prognostic index of the changes in the electrocardiogram signal may then be used to generate a fluid responsiveness prediction. The fluid responsiveness prediction may then be used to control fluid administration for patients that are receiving anesthesia during surgery or that are critically ill or unresponsive. Disclosed embodiments may also be used as a diagnostic tool for athletes during training, who are undergoing endurance tests, for bike riders, etc.

Claims18

Dependent claims

A processor-implemented method of determining fluid responsiveness and of providing treatment to a human subject based on the determined fluid responsiveness, the method comprising:

continuously measuring, by a sensor, an electrocardiogram signal of a human subject, the electrocardiogram signal having a frequency greater than 250 Hz to about 1000 Hz;
performing, by a processor, operations of a first machine learning algorithm using the electrocardiogram signal as input to the first machine learning algorithm to continuously determine features of the electrocardiogram signal;
performing operations of a second machine learning model, using the determined features as input to the second machine learning model, to generate a fluid responsiveness index as output of the second machine learning model, wherein one of the determined features used as input by the second machine learning algorithm to generate a fluid responsiveness index is a rise angle of the QRS complex of the electrocardiogram signal;
and
administering a course of treatment to the human subject based on the fluid responsiveness index.

2 more elements

A system that determines fluid responsiveness and determines a course of treatment for a human subject based on the determined fluid responsiveness, the system comprising:

a sensor that is configured to continuously measure an electrocardiogram signal of a human subject, the electrocardiogram signal having a frequency greater than 250 Hz to about 1000 Hz;
a processor that is configured to continuously perform operations comprising:
receiving the measured electrocardiogram signal of a human subject;
performing operations of a first machine learning algorithm using the electrocardiogram signal as input to determine features of the electrocardiogram signal;
performing operations of a second machine learning model, using the determined features as input to the second machine learning model, to generate a fluid responsiveness index as output of the second machine learning model, wherein one of the determined features used as input by the second machine learning algorithm to generate a fluid responsiveness index is a rise angle of the QRS complex of the electrocardiogram signal;
and
determining a course of treatment to be administered to the human subject based on the fluid responsiveness index.

4 more elements

Dependencies are inferred from the claim text.

Methods and systems for improved prediction of fluid responsiveness

Life of this patent

Methods and systems for improved prediction of fluid responsiveness

Life of this patent

Abstract

Claims18