Patent US8755877B2 successfully refreshed from Google Patents.
← Back to Patent Explorer
US8755877B2 Active / In Force US ⏱️ 5.4 years remaining

Real time QRS detection using adaptive threshold

Assignee: Texas Instruments Inc
•
Filing: 2012-03-29
•
Granted: 2014-06-17
•
Expires: 2032-03-29
📄 Original PDF 🌐 Google Patents 🔄 Sync Now

Abstract

Abstract A mobile system for analyzing ECG data includes an analog front end module coupled to a mobile consumer device. The analog front end module is configured to collect ECG data from one or more leads and is operable to convert the analog ECG data to digital ECG data. The mobile consumer device is coupled to receive the digital ECG data, and is configured to perform QRS detection using a filter whose cutoff frequency is adapted to noise level in real time. The ECG signal is amplified non-linearly and three windowed threshold signals (D, E, J) are derived. The cutoff frequency for the QRS detection is dynamically selected as a function of the threshold signals. A sample in the amplified signal is identified to be a heart beat point only when the sample value is equal to the first threshold signal and greater than the filtered threshold signal.

Inventors

  • 👤 Vasile Zoica

Bibliographic Details

Application Number —
Jurisdiction US
Data Source google_patents
Last Synchronized 2026-11-01 20:38
CPC Classifications
A; A61; A61B; A61B5/00; A61B5/68; A61B5/6887; A61B5/6898; A61B5/02; A61B5/024; A61B5/02405; A61B5/24; A61B5/316; A61B5/318; A61B5/346; A61B5/349; A61B5/352; A61B5/72; A61B5/7203; A61B5/7235; A61B5/725; A61B5/0002; A61B5/0004; A61B5/0006; A61B5/0015; A61B5/0022; A61B5/25; A61B5/279; A61B5/28; A61B5/282; A61B5/6801; A61B5/6802; A61B5/6804; A61B5/6805; A61B5/74; A61B5/742; A61B5/743; A61B5/746

Claims Analysis

Total Claims: 16
Claim 1
Independent Claim 226 words
A method for processing ECG data, the method comprising: receiving a stream of raw ECG data samples comprising PQRST patterns; filtering the raw ECG data to form filtered ECG data by using a non-linear filter to minimize baseline wander and to suppress a T wave portion of each PQRST pattern; performing a non-linear operation on the filtered ECG data to form an amplified signal that exaggerates an R peak of each PQRST pattern; deriving a first threshold signal and a third threshold signal using a moving max filter on the amplified signal; deriving a second threshold signal using a moving average filter on the amplified signal; filtering the third threshold signal to form a filtered threshold signal using a filter in which a cutoff frequency is dynamically selected as a function of the first threshold signal and the second threshold signal; and identifying a sample in the amplified signal to be a heart beat point only when the sample value is equal to the first threshold signal and greater than the filtered threshold signal, wherein filtering the raw ECG data comprises: filtering the raw ECG data using a cascade of median filter and low pass filter; delaying the raw ECG data and subtracting an output signal of the low pass filter from the delayed raw ECG data in order to deliver a high pass filtered data.
Claim 2
Independent Claim 226 words
A method for processing ECG data, the method comprising: receiving a stream of raw ECG data samples comprising PQRST patterns; filtering the raw ECG data to form filtered ECG data by using a non-linear filter to minimize baseline wander and to suppress a T wave portion of each PQRST pattern; performing a non-linear operation on the filtered ECG data to form an amplified signal that exaggerates an R peak of each PQRST pattern; deriving a first threshold signal and a third threshold signal using a moving max filter on the amplified signal; deriving a second threshold signal using a moving average filter on the amplified signal; filtering the third threshold signal to form a filtered threshold signal using a filter in which a cutoff frequency is dynamically selected as a function of the first threshold signal and the second threshold signal; and identifying a sample in the amplified signal to be a heart beat point only when the sample value is equal to the first threshold signal and greater than the filtered threshold signal, wherein filtering the raw ECG data comprises: filtering the raw ECG data using a cascade of median filter and low pass filter; delaying the raw ECG data and subtracting an output signal of the low pass filter from the delayed raw ECG data in order to deliver a high pass filtered data.
Claim 3
Dependent Claim 43 words
The method of claim 1 , further comprising: filtering the raw ECG data using a non-linear filter to minimize baseline wander of each PQRST pattern to form a viewable ECG signal; and annotating the viewable ECG signal to indicate each heart beat point.
Claim 4
Dependent Claim 58 words
The method of claim 2 , wherein filtering the raw ECG data comprises: filtering the raw ECG data using a cascade of median filter and low pass filter; delaying the raw ECG data; and subtracting an output signal of the low pass filter from the delayed raw ECG data in order to deliver a high pass filtered data.
Claim 5
Dependent Claim 22 words
The method of claim 1 , wherein the amplified signal is formed by squaring the filtered ECG data at least one time.
Claim 6
Dependent Claim 59 words
The method of claim 1 , wherein filtering the third threshold signal comprises: filtering the third threshold signal with a first order infinite impulse response (IIR) filter; and dynamically selecting a cutoff frequency for the IIR as a linear function of a difference between a logarithm of the first threshold signal and a logarithm of the second threshold signal.
Claim 7
Dependent Claim 53 words
The method of claim 1 , wherein filtering the third threshold signal comprises: filtering the third threshold signal with a first order infinite impulse response (IIR) filter; and dynamically selecting a cutoff frequency for the IIR as a linear function of a ratio of the first threshold signal and the second threshold signal.
Claim 8
Dependent Claim 46 words
The method of claim 1 , wherein deriving a first threshold signal and a third threshold signal using a moving max filter on the amplified signal uses a first window width for the first threshold signal and a different window width for the third threshold signal.
Claim 9
Dependent Claim 255 words
A computer-readable non-transitory storage media having computer-readable instructions stored therein that when executed by a computer performs a method of processing ECG data in real time, the method comprising: receiving a stream of raw ECG data samples comprising PQRST patterns from an analog front end coupled to the computer; filtering the raw ECG data to form filtered ECG data by using a non-linear filter to minimize baseline wander and to suppress a T wave portion of each PQRST pattern; performing a non-linear operation on the filtered ECG data to form an amplified signal that exaggerates an R peak of each PQRST pattern; deriving a first threshold signal and a third threshold signal using a moving max filter on the amplified signal; deriving a second threshold signal using a moving average filter on the amplified signal; filtering the third threshold signal to form a filtered threshold signal using a filter in which a cutoff frequency is dynamically selected as a function of the first threshold signal and the second threshold signal; and identifying a sample in the amplified signal to be a heart beat point only when the sample value is equal to the first threshold signal and greater than the filtered threshold signal, wherein filtering the raw ECG data comprises: filtering the raw ECG data using a cascade of median filter and low pass filter; delaying the raw ECG data; and subtracting an output signal of the low pass filter from the delayed raw ECG data in order to deliver a high pass filtered data.
Claim 10
Dependent Claim 255 words
A computer-readable non-transitory storage media having computer-readable instructions stored therein that when executed by a computer performs a method of processing ECG data in real time, the method comprising: receiving a stream of raw ECG data samples comprising PQRST patterns from an analog front end coupled to the computer; filtering the raw ECG data to form filtered ECG data by using a non-linear filter to minimize baseline wander and to suppress a T wave portion of each PQRST pattern; performing a non-linear operation on the filtered ECG data to form an amplified signal that exaggerates an R peak of each PQRST pattern; deriving a first threshold signal and a third threshold signal using a moving max filter on the amplified signal; deriving a second threshold signal using a moving average filter on the amplified signal; filtering the third threshold signal to form a filtered threshold signal using a filter in which a cutoff frequency is dynamically selected as a function of the first threshold signal and the second threshold signal; and identifying a sample in the amplified signal to be a heart beat point only when the sample value is equal to the first threshold signal and greater than the filtered threshold signal, wherein filtering the raw ECG data comprises: filtering the raw ECG data using a cascade of median filter and low pass filter; delaying the raw ECG data; and subtracting an output signal of the low pass filter from the delayed raw ECG data in order to deliver a high pass filtered data.
Claim 11
Dependent Claim 58 words
The computer-readable non-transitory storage media of claim 8 , further comprising: filtering the raw ECG data using a non-linear filter to minimize baseline wander of each PQRST pattern to form a viewable ECG signal; annotating the viewable ECG signal to indicate each heart beat point; and displaying the annotated viewable ECG signal on a display coupled to computer.
Claim 12
Dependent Claim 25 words
The computer-readable non-transitory storage media of claim 8 , wherein the amplified signal is formed by squaring the filtered ECG data at least one time.
Claim 13
Dependent Claim 62 words
The computer-readable non-transitory storage media of claim 8 , wherein filtering the third threshold signal comprises: filtering the third threshold signal with a first order infinite impulse response (IIR) filter; and dynamically selecting a cutoff frequency for the IIR as a linear function of a difference between a logarithm of the first threshold signal and a logarithm of the second threshold signal.
Claim 14
Dependent Claim 56 words
The computer-readable non-transitory storage media of claim 8 , wherein filtering the third threshold signal comprises: filtering the third threshold signal with a first order infinite impulse response (IIR) filter; and dynamically selecting a cutoff frequency for the IIR as a linear function of a ratio of the first threshold signal and the second threshold signal.
Claim 15
Dependent Claim 22 words
The computer readable non-transitory storage media of claim 8 , wherein the computer readable storage media is comprised within a smart phone.
Claim 16
Dependent Claim 22 words
The computer readable non-transitory storage media of claim 8 , wherein the computer readable storage media is comprised within a medical system.

USPTO Annuity & Maintenance Windows

Annuity Stage Window Opens Payment Deadline Surcharge Period Until Estimated Fee Status
3.5 Years (Year 4) 2017-06-16 2017-12-16 2018-06-17 $2000.00 Window Closed / Lapsed
7.5 Years (Year 8) 2021-06-16 2021-12-16 2022-06-17 $3760.00 Window Closed / Lapsed
11.5 Years (Year 12) 2025-06-16 2025-12-16 2026-06-17 $7700.00 Window Closed / Lapsed

Assigned Categories & Tags

No categories assigned to this patent yet.

Assign Category Tag