Heart beat signal recognition
First Claim
1. A computer implemented method to determine a heart rate for a subject, comprising:
- detecting a heartbeat signal from the subject;
extracting a set of two or more types of heartbeat features from the heartbeat signal;
accessing, from stored data, one or more heartbeat patterns and one or more noise patterns, the one or more heartbeat patterns comprising at least one of;
(a) one or more heartbeat models and (b) one or more heartbeat templates, and the one or more noise patterns comprising at least one of;
(c) one or more noise models and (d) one or more noise templates;
determining whether the set of two or more types of heartbeat features more closely matches the one or more heartbeat patterns or the one or more noise patterns;
if the set of two or more types of heartbeat features more closely matches the one or more noise patterns, determining that the set of two or more types of heartbeat features is not to be used to update the heart rate for the subject;
if the set of two or more types of heartbeat features more closely matches the one or more heartbeat patterns, determining that the set of two or more types of heartbeat features represents a heartbeat candidate;
performing post-processing to determine that the heartbeat candidate represents a heartbeat; and
updating the heart rate for the subject based on timing of the heartbeat.
1 Assignment
0 Petitions
Accused Products
Abstract
A subject'"'"'s heart rate is determined. A heart rate monitor receives a Doppler signal reflected from an artery of a target, performs demodulation and heart beat recognition techniques to determine a set of features in each frame of the signal. Pattern classification is performed to determine if the extracted feature sequence is associated with heart beats. The pattern classification may include finding the optimal state sequence by calculating the probability of each allowable state sequence based on the extracted feature sequence and heart beat models or additional noise models. Or, a heart beat candidate is determined using frame energy and dynamic thresholding followed by computing the probabilities between the feature sequence and each stored heart beat model or additional noise models. Or, heart beat candidates are determined using frame energy and dynamic thresholding which compute the similarity between the feature sequences and each of the stored heart beat templates.
-
Citations
41 Claims
-
1. A computer implemented method to determine a heart rate for a subject, comprising:
-
detecting a heartbeat signal from the subject; extracting a set of two or more types of heartbeat features from the heartbeat signal; accessing, from stored data, one or more heartbeat patterns and one or more noise patterns, the one or more heartbeat patterns comprising at least one of; (a) one or more heartbeat models and (b) one or more heartbeat templates, and the one or more noise patterns comprising at least one of;
(c) one or more noise models and (d) one or more noise templates;determining whether the set of two or more types of heartbeat features more closely matches the one or more heartbeat patterns or the one or more noise patterns; if the set of two or more types of heartbeat features more closely matches the one or more noise patterns, determining that the set of two or more types of heartbeat features is not to be used to update the heart rate for the subject;
if the set of two or more types of heartbeat features more closely matches the one or more heartbeat patterns, determining that the set of two or more types of heartbeat features represents a heartbeat candidate;performing post-processing to determine that the heartbeat candidate represents a heartbeat; and updating the heart rate for the subject based on timing of the heartbeat. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26)
-
-
27. One or more processor readable storage devices having processor readable code embodied on said processor readable storage devices, said processor readable code for programming one or more processors to perform a method to determine a heart rate for a subject, the method performed comprising:
-
extracting a set of features from each frame of a plurality of frames of a heartbeat signal of the subject; performing pattern classification of the set of features using one or more heartbeat patterns and one or more noise patterns, the one or more heartbeat patterns comprising at least one of;
(a) one or more heartbeat models and (b) one or more heartbeat templates, and the one or more noise patterns comprising at least one of;
(c) one or more noise models and (d) one or more noise templates;determining whether the set of features more closely matches the one or more heartbeat patterns or the one or more noise patterns, based on the performing pattern classification; if the set of features more closely matches the one or more noise patterns, determining that the set of features is not to be used to update the heart rate for the subject; if the set of features more closely matches the one or more heartbeat patterns, determining that the set of features represents a heartbeat candidate; performing post-processing to determine whether the heartbeat candidate represents a heartbeat; and updating the heart rate for the subject based on timing of the heartbeat, if the post-processing determines that the heartbeat candidate represents the heartbeat. - View Dependent Claims (28, 29, 30, 31, 32, 33, 34, 35)
-
-
36. A heart rate monitor, comprising:
-
at least one transducer which detects a heartbeat signal from a subject; at least one microcontroller associated with the at least one transducer; at least one data store associated with the microcontroller, the data store comprising one or more heartbeat patterns and one or more noise patterns, the one or more heartbeat patterns comprising at least one of;
(a) one or more heartbeat models and (b) one or more heartbeat templates, and the one or more noise patterns comprising at least one of;
(c) one or more noise models and (d) one or more noise templates; andan amplifier and a demodulator arranged between the at least one transducer and the at least one microcontroller; the at least one microcontroller;
(i) extracts a set of two or more types of heartbeat features from the heartbeat signal, (ii) accesses, from the at least one data store, the one or more heartbeat patterns and one or more noise patterns, (iii) determines whether the set of two or more types of heartbeat features more closely matches the one or more heartbeat patterns or the one or more noise patterns, (iv) if the set of two or more types of heartbeat features more closely matches the one or more noise patterns, determines that the set of two or more types of heartbeat features is not to be used to update a heart rate for a subject, (v) if the set of two or more types of heartbeat features more closely matches the one or more heartbeat patterns, determines that the set of two or more types of heartbeat features represents a heartbeat candidate, and (iv) updates the heart rate for the subject based on timing of the heartbeat candidate. - View Dependent Claims (37, 38, 39)
-
-
40. A heart rate monitor, comprising:
-
at least one transducer which detects a heartbeat signal from a subject; at least one microcontroller associated with the at least one transducer; and at least one data store associated with the microcontroller, the data store comprising one or more heartbeat patterns and one or more noise patterns, the one or more heartbeat patterns comprising at least one of;
(a) one or more heartbeat models and (b) one or more heartbeat templates, and the one or more noise patterns comprising at least one of;
(c) one or more noise models and (d) one or more noise templates;the at least one microcontroller;
(i) extracts a set of heartbeat features from the heartbeat signal, (ii) accesses, from the at least one data store, the one or more heartbeat patterns and one or more noise patterns, (iii) determines whether the set of features more closely matches the one or more heartbeat patterns or the one or more noise patterns, (iv) if the set of features more closely matches the one or more noise patterns, determines that the set of features is not to be used to update a heart rate for a subject, (v) if the set of features more closely matches the one or more heartbeat patterns, determines that the set of features represents a heartbeat candidate, and (vi) updates the heart rate for the subject based on timing of the heartbeat candidate. - View Dependent Claims (41)
-
Specification