US-12616387-B2 - Derivation of physiological parameters from a radar signal

US12616387B2US 12616387 B2US12616387 B2US 12616387B2US-12616387-B2

Abstract

Apparatus and methods are described including deriving a subject's heart rate from in-phase and the quadrature signals received by a radar. The in-phase and quadrature signals are processed to generate two or more outputs using two or more respective methods. For each of the two or more outputs, for each of a plurality of time segments, the subject's heart rate is derived from the filtered signal, and quality scores are assigned to the subject's heart rate as derived for each of the plurality of time segments from each of the two or more outputs. At least partially based upon the quality scores, the subject's heart rate is derived using the output of one of the methods. Other applications are also described.

Inventors

Reinier DOELMAN
Ehud Fishler

Assignees

NETEERA TECHNOLOGIES LTD.

Dates

Publication Date: 20260505
Application Date: 20220626

Claims (20)

1 . A method for deriving a heart rate of a subject comprising the steps of: receiving in-phase and quadrature signals using a radar; and deriving a subject's heart rate from the in-phase and the quadrature signals by: processing the in-phase and quadrature signals to generate two or more outputs using two or more respective methods, the two or more methods comprising at least two out of: (a) calculating a phase of each of the in-phase and the quadrature signals with respect to a center for each point in time, (b) calculating a complex signal based upon the in-phase and the quadrature signals, and (c) calculating a linear combination of the in-phase and quadrature signals; for each of the two or more outputs: filtering the outputsignal to at least partially dampen frequencies that are below a minimum frequency to generate a filtered signal; dividing the filtered signal into a plurality of time segments; and for each of the plurality of time segments, deriving the subject's heart rate from the filtered signal; assigning quality scores to the subject's heart rate as derived for each of the plurality of time segments from each of the two or more outputs; and at least partially based upon the quality scores, selecting to derive the subject's heart rate using the output of either method (a), (b), or (c); and outputting the subject's heart rate, as derived using the selected output.
2 . The method according to claim 1 , wherein filtering the output to at least partially dampen frequencies that are below a minimum frequency comprises at least one filtering selected from the group consisting of: filtering the output to at least partially dampen frequencies that are above a maximum frequency and below a minimum frequency; filtering the output to remove frequencies that are below the minimum frequency; filtering the signaloutput to remove frequencies that are above a maximum frequency and below a minimum frequency; filtering the output to at least partially dampen frequencies that are below 5 Hz; filtering the output to at least partially dampen frequencies that are below 8 Hz; filtering the output to at least partially dampen frequencies that are below 5 Hz and above 50 Hz; and filtering the output to at least partially dampen frequencies that are below 8 Hz and above 40 Hz.
3 . The method according to claim 1 , further comprising, based upon the outputted subject's heart rate, deriving one or additional physiological parameters of the subject, the one or additional physiological parameters of the subject being selected from the group consisting of: heart rate variability, heart rate interval, pulse wave velocity, blood pressure, mean arterial pressure, systolic pressure, diastolic pressure, vascular resistance, pulse pressure variability, stroke volume, and stroke volume variability.
4 . The method according to claim 1 , wherein for each of a plurality of time segments deriving the subject's heart rate from the filtered signal comprises, at least one operation selected from the group consisting of: for each of a plurality of time segments, calculating a power spectrum using the filtered signal; for each of a plurality of overlapping time segments, deriving the subject's heart rate from the filtered signal; for each of a plurality of non-overlapping time segments, deriving the subject's heart rate from the filtered signal; for each of a plurality of time segments having a duration of at least 2 seconds, deriving the subject's heart rate from the filtered signal; and for each of a plurality of time segments having a duration of between 5 seconds and 15 seconds, deriving the subject's heart rate from the filtered sign.
5 . The method according to claims 1 , wherein receiving in-phase and quadrature signals using a radar comprises receiving in-phase and quadrature signals using at least one of: a continuous-wave (CW) radar system; and a frequency-modulated continuous-wave (FMCW) radar system.
6 . The method according to claims 1 , wherein outputting the subject's heart rate as derived using the selected output comprises, using the selected output, deriving the subject's heart rate from the filtered signal for each of the time segments, and calculating a weighted average of the subject's heart rate as derived for each of the time segments.
7 . The method according to claim 6 , wherein calculating the weighted average of the power spectra of the time segments comprises assigning weights to respective time segments based upon the quality scores of the time segments.
8 . The method according to claim 1 , further comprising at least partially based upon the quality scores as derived for each of the plurality of time segments selecting a frequency range to utilize for filtering the output to at least partially dampen frequencies that are below the minimum frequency.
9 . The method according to claim 8 , wherein at least partially based upon the quality scores, selecting to derive the subject's heart rate using the output of either method (a), (b), or (c) comprises performing a grid search over the two or more outputs and over a plurality of frequency ranges to utilize for filtering the output to at least partially dampen frequencies that are below the minimum frequency, and selecting to derive the subject's heart rate using a given methodology based upon the grid search.
10 . The method according to claims 1 , wherein assigning quality scores to the subject's heart rate as derived for each of the plurality of time segments from each of the two or more outputs comprises at least one selected from the group consisting of: assigning quality scores based upon the presence of a prominent peak within a power spectrum corresponding to that time segment from each of the two or more outputs; assigning high quality scores to time segments having a prominent peak within the power spectrum and low quality scores to time segments having a smeared signal; and assigning high quality scores to time segments having a prominent peak within the power spectrum and low quality scores to time segments in which the signal is indicative of large body motion of the subject.
11 . Apparatus for deriving a heart rate of a subject, for use with a radar, the apparatus comprising: at least one computer processor configured to: receive in-phase and quadrature signals from the radar; derive a subject's heart rate from the in-phase and the quadrature signals by: processing the in-phase and quadrature signals to generate two or more outputs using two or more respective methods, the two or more methods comprising at least two out of: (a) calculating a phase of each of the in-phase and the quadrature signals with respect to a center for each point in time, (b) calculating a complex signal based upon the in-phase and the quadrature signals, and (c) calculating a linear combination of the in-phase and quadrature signals; for each of the two or more outputs: filtering the output to at least partially dampen frequencies that are below a minimum frequency to generate a filtered signal; dividing the filtered signal into a plurality of time segments; and for each of the plurality of time segments, deriving the subject's heart rate from the filtered signal; assigning quality scores to the subject's heart rate as derived for each of the plurality of time segments from each of the two or more outputs; and at least partially based upon the quality scores, selecting to derive the subject's heart rate using the output of either method (a), (b), or (c); and output the subject's heart rate, as derived using the selected output.
12 . The apparatus according to claim 11 , wherein the at least one computer processoris configured to filter the output to at least partially dampen frequencies by at least one filtering selected from the group consisting of: filtering the output to at least partially dampen frequencies that are above a maximum frequency and below a minimum frequency; filtering the output to remove frequencies that are below the minimum frequency; filtering the output to remove frequencies that are above a maximum frequency and below a minimum frequency; filtering the output to at least partially dampen frequencies that are below 5 Hz; filtering the output to at least partially dampen frequencies that are below 8 Hz; filtering the output to at least partially dampen frequencies that are below 5 Hz and above 50 Hz; and filtering the output to at least partially dampen frequencies that are below 8 Hz and above 40 Hz.
13 . The apparatus according to claim 11 , wherein, based upon the outputted subject's heart rate, the at least one computer processor is configured to derive one or additional physiological parameters of the subject, the one or additional physiological parameters of the subject being selected from the group consisting of: heart rate variability, heart rate interval, pulse wave velocity, blood pressure, mean arterial pressure, systolic pressure, diastolic pressure, vascular resistance, pulse pressure variability, stroke volume, and stroke volume variability.
14 . The apparatus according to claim 11 , wherein the at least one computer processor is configured to derive the subject's heart rate from the filtered signal for each of the plurality of time segments by at least one operation selected from the group consisting of: calculating a power spectrum using the filtered signal, for each of the plurality of time segments; deriving the subject's heart rate from the filtered signal, for each of a plurality of overlapping time segments; deriving the subject's heart rate from the filtered signal, for each of a plurality of non-overlapping time segments; deriving the subject's heart rate from the filtered signal, for each of a plurality of time segments having a duration of at least 2 seconds; deriving the subject's heart rate from the filtered signal, for each of a plurality of time segments having a duration of between 5 seconds and 15 seconds.
15 . The apparatus according to claims 11 , wherein the radar includes at least one of: a continuous-wave (CW) radar system; and a frequency-modulated continuous-wave (FMCW) radar system, and wherein the at least one computer processor is configured to receive the in-phase and quadrature signals from at least one of: the continuous-wave (CW) radar system and the frequency- modulated continuous-wave (FMCW) radar system.
16 . The apparatus according to claims 11 , wherein the at least one computer processor is configured to output the subject's heart rate as derived using the selected output by, using the selected output, deriving the subject's heart rate from the filtered signal for each of the time segments, and calculating a weighted average of the subject's heart rate as derived for each of the time segments.
17 . The apparatus according to claim 16 , wherein the at least one computer processor is configured to calculate the weighted average of the power spectra of the time segments by assigning weights to respective time segments based upon the quality scores of the time segments.
18 . The apparatus according to claim 11 , wherein the at least one computer processoris configured, at least partially based upon the quality scores as derived for each of the plurality of time segments, to select a frequency range to utilize for filtering the output to at least partially dampen frequencies that are below the minimum frequency.
19 . The apparatus according to claim 18 , wherein the at least one computer processor is configured to perform a grid search over the two or more outputs and over a plurality of frequency ranges to utilize for filtering the output to at least partially dampen frequencies that are below the minimum frequency, and to select to derive the subject's heart rate using a given methodology based upon the grid search.
20 . The apparatus according to claims 11 , wherein the at least one computer processor is configured to assign quality scores to the subject's heart rate as derived for each of the plurality of time segments from each of the two or more outputs by at least one selected from the group consisting of: assigning quality scores based upon the presence of a prominent peak within a power spectrum corresponding to that time segment from each of the two or more outputs; assigning high quality scores to time segments having a prominent peak within the power spectrum and low quality scores to time segments having a smeared signal; and assigning high quality scores to time segments having a prominent peak within the power spectrum and low quality scores to time segments in which the signal is indicative of large body motion of the subject.

Description

This application is a national phase of International Application No. PCT/IL2022/050682 filed Jun. 26, 2022, which claims the benefit of United States of America Application No. 63/216,005 filed Jun. 29, 2021, each of which is hereby incorporated herein by reference in its entirety. FIELD OF THE INVENTION Some applications of the presently disclosed subject matter relate generally to the derivation of a subject's physiological parameters in a non-invasive manner, and, in particular, to using a non-contact radar to derive a subject's heart rate. BACKGROUND There are a variety of methods that can be used for measuring a subject's physiological parameters (such as heart rate) in a non-invasive and/or non-contact manner. For example, a radar system can be used to detect motion of the subject without contacting the subject. A challenge associated with the detection of a subject's physiological parameters (such as heart rate) when using such systems is the removal of unrelated artifacts that are present within the subject's motion signal. For example, artifacts may be present within the signal as a result of instrument noise, muscle spasms, external or subject motion artifacts, and/or signal bias drift. The presence of noise within the radar signal may result in a such a system presenting the wrong heart rate value or no heart rate value at all. SUMMARY OF EMBODIMENTS In accordance with some applications of the present invention, a radar is configured to detect physiological parameters of a subject in a non-invasive and non-contact manner. For some applications, the radar is a frequency-modulated radar, for example, a frequency-modulated continuous-wave radar. For some applications, the radar transmitter is configured to transmit a signal having a frequency of between 500 MHz and 15 GHz, e.g., between 500 MHz and 2 GHz, or between 2 GHz and 10 GHz. Typically, a receiver of the radar detects a reflection of the signal, a portion of which is reflected from the subject's body, and the computer processor analyzes the transmitted and the received signals to thereby detect motion of the subject's body and derive physiological parameters of the subject based upon the motion (micro-motion) of the subject's body. For some applications, the computer processor is configured to periodically sample the signals from the radar as measured over a given time period (or over an entire session) and to derive a representative value of the physiological parameter for the time period. For some applications, in order to derive a representative value of the physiological parameter for the time period, the computer processor analyses a radar signal using an algorithm as described hereinbelow. Typically, by applying the algorithm as described hereinbelow, the computer processor optimizes the accuracy of the derived heart rate by first determining a processing method that yields the highest quality measure of the heart rate and using that processing method for deriving the heart rate. As noted above, the algorithm described hereinbelow is typically applied in order to derive a representative value of a subject's heart rate over a given time period. Thus, the algorithm may be referred to as a periodic algorithm in that it is run periodically in order to derive a representative value of a subject's heart rate over a given time period. For some applications, in addition to deriving a representative value of a subject's heart rate over a given time period (e.g., using the periodic algorithm), the computer processor is configured to run a continuous algorithm that is configured to determine the subject's heart rate in real time, as data is received by the radar. For some applications, the periodic algorithm is used to validate and/or correct the heart rate as determined using the continuous algorithm, as described in further detail hereinbelow. For example, the continuous algorithm may be steered toward detecting a frequency range that corresponds to the representative heart rate as detected using continuous algorithm. There is therefore provided, in accordance with some applications of the present invention, apparatus for use with a radar, the apparatus including: at least one computer processor configured to:receive in-phase and quadrature signals from the radar;derive a subject's heart rate from the in-phase and the quadrature signals by: processing the in-phase and quadrature signals to generate two or more outputs using two or more respective methods, the two or more methods including at least two out of: (a) calculating a phase of each of the in-phase and the quadrature signals with respect to a center for each point in time, (b) calculating a complex signal based upon the in-phase and the quadrature signals, and (c) calculating a linear combination of the in-phase and quadrature signals;for each of the two or more outputs: filtering the signal to at least partially dampen frequencies that are below a minimum frequency