Back to results
US 11,452,457 B2Granted
Method and device for improving prediction and detection of change in a physiological condition
AssigneeMedicus Engineering Aps
InventorsJesper Fleischer, Simon Lebech Cichosz
Life of this patent
Estimating the term…
Abstract
Devices, methods and systems for predicting and detecting a change in a physiological condition of a subject are disclosed. A device for detecting and predicting a change in physiological condition of a subject may include one or more processors configured to: (i) analyze changes in a heart rhythm of a subject; (ii) analyze a signal from one or more biomarkers that are associated with a physiological condition of the subject; and (iii) determine, based on the changes in the heart rhythm and the signal from the one or more biomarkers, whether there is a change in the physiological condition of the subject during a time period.
Claims28
A device comprising one or more processors configured to:
- obtain continuously over time, a heart rhythm signal related to a heart rhythm of a subject and a related blood glucose signal of the subject;
- analyze continuously over time changes in a pattern of a heart rate variability of the subject based on the heart rhythm signal;
- analyze continuously over the time changes in a pattern of the related blood glucose signal;
- extract features from the pattern of the heart rate variability and the pattern of the related blood glucose signal;
- reduce the features extracted based on a ranking algorithm to eliminate certain features that do not signify a physiological event;
- using pattern recognition of the heart rate variability and the pattern recognition of the related blood glucose signal to identify changes in a pattern of heart rate variability and the pattern of the related blood glucose signal correlated to a physiological condition;
- predict a future physiological condition, using a subset of the features extracted from a combined pattern recognition of the heart rate variability and the pattern recognition of the related blood glucose signal, via a classification model configured to predict whether there would be a change in a physiological condition of the subject during a future time period of the subject, a prediction being based on the changes in the pattern of heart rate variability and the pattern of the related blood glucose signal;
- wherein the related blood glucose signal is a surrogate measurement of blood glucose;
- and
- wherein the related blood glucose signal is measured by a continuous glucose monitor.
7 more elements
A method comprising:
- obtaining continuously over time, both a heart rhythm signal related to a heart rhythm of a subject and a related blood glucose signal;
- analyzing continuously over time, via one or more processors, changes in a dynamic pattern of a heart rate variability of the subject based on the heart rhythm signal;
- analyzing continuously over time, via the one or more processors, changes in a pattern of the related blood glucose signal;
- extracting features from a dynamic pattern of the heart rate variability and the pattern of the related blood glucose signal;
- combining the dynamic pattern of the heart rate variability and the pattern of the related blood glucose signal to identify changes;
- extracting features based on a ranking algorithm to eliminate a lag-time between a continuous glucose monitor measurement and an actual blood glucose level;
- utilizing a subset of the features extracted via a classification model for optimal classification;
- and
- predicting, via the one or more processors, whether there would be a change in a physiological condition of the subject during a future time period of the subject, a prediction being based on the changes;
- wherein the related blood glucose signal is a surrogate measurement of blood glucose of the subject;
- and
- wherein the related blood glucose signal is measured by an additional continuous glucose monitor.
9 more elements
A system comprising:
- one or more sensors;
- a device comprising one or more processors configured to:
- obtain continuously over time, via the one or more sensors, a pattern of a heart rhythm signal and a pattern of a related blood glucose signal of a subject;
- analyze continuously over time, changes in a pattern of a heart rate variability of the subject based on the heart rhythm signal;
- analyze continuously over time, changes in a pattern of the related blood glucose signal;
- combine the pattern of the heart rate variability and the pattern of the related blood glucose signal to identify changes;
- extract features from the pattern of the heart rate variability and the pattern of the related blood glucose signal;
- utilize a subset of the features extracted via a classification model for optimal classification;
- and
- predict whether there would be a change in a physiological condition of the subject during a future time period of the subject, a prediction being based on the changes;
- wherein the related blood glucose signal is a surrogate measurement of blood glucose of the subject;
- and wherein the related blood glucose signal is measured by an additional continuous glucose monitor.
9 more elements
Dependencies are inferred from the claim text.