Electrocardiogram device and methods

• US 10,441,184 B2
• Filed: 03/15/2017
• Issued: 10/15/2019
• Est. Priority Date: 03/16/2016
• Status: Active Grant

• Current Assignee: Ventrilink Corporation
• Original Assignee: Ventrilink Corporation
• Inventors: Baumann, Eric, Korzinov, Lev, Churchville, David
• Primary Examiner(s): Levicky, William J
• Application Number:
  US15/460,036
• Publication Number:
  US 20170265765A1
• Time in Patent Office: 944 Days
• CPC Class Codes: A61B 5/002 Monitoring the patient using a local or closed circuit, e.g. in a room or building A61B5/0017 takes precedence
• CPC Class Codes: A61B 5/30 Input circuits therefor
• CPC Class Codes: A61B 5/316 Modalities, i.e. specific diagnostic methods
• CPC Class Codes: A61B 5/339 Displays specially adapted therefor
• CPC Class Codes: A61B 5/361 Detecting fibrillation
• CPC Class Codes: A61B 5/7257 using Fourier transforms
• CPC Class Codes: A61B 5/726 using Wavelet transforms
• CPC Class Codes: A61B 5/7264 Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems neural networks per se G06N3/00; expert systems per se G06N5/00
• CPC Class Codes: A61B 5/7465 Arrangements for interactive communication between patient and care services, e.g. by using a telephone network telemetry of measured physiological signal A61B5/0002
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

Devices and methods are described that provide improved diagnosis from the processing of physiological data. The methods include use of transforms prior to submitting the data to a multiple level neural network. In one embodiment for ECG analysis, a template is used to subtract data that is not pertinent to the diagnosis and then a Fourier transform is applied to the time series data. Examples are shown with applications to electrocardiogram data, but the methods taught are applicable to many types of physiological data.

Claims

1. A system for providing a health diagnosis, to a medical practitioner, from electrocardiogram data acquired from a patient over a time period, said electrocardiogram data including a plurality of heartbeats, and, each heartbeat including PQRST wave data, and, said system comprising:

a) a sensor for measuring the electrocardiogram data over the time period, and, storing the electrocardiogram data, and, transferring the electrocardiogram data to a computing device,

b) the computing device further programmed to;

i) perform a mathematical transform on the electrocardiogram data, where the mathematical transform is at least one selected, by the medical practitioner, from;

(1) a wavelet transform,

(2) creating a template of QRS data from the electrocardiogram data and subtracting the template of QRS data from each of the plurality of heartbeats,

(3) a Fourier transform, and

where the mathematical transform is selected on the basis of a health diagnosis to be tested, thereby producing transformed physiological data, and

ii) encode the transformed physiological data using a multiple layer neural network, thereby producing an encoded output,

iii) classify the encoded output using a classifier that has been previously trained using a library of time varying physiological data with confirmed diagnoses, thereby producing a classified output,

c) wherein the classified output is the health diagnosis for the patient.

2. The system of claim 1 wherein the library of time varying physiological data with confirmed diagnoses is produced by collecting and interpreting time varying physiological data from the patient in a first time period, said first time period occurring prior to the time period of claim 1, and wherein the confirmed diagnoses are produced by a human technician.

3. The system of claim 1 wherein the mathematical transform includes:

creating a template of QRS data from the electrocardiogram data and subtracting the template of QRS data from each of the plurality of heartbeats, and, a Fourier transform, and, the health diagnosis to be tested is P wave blockage.

4. The system of claim 3 wherein the library of time varying physiological data with confirmed diagnoses is produced by collecting and interpreting time varying physiological data from the patient in a first time period, said first time period occurring prior to the time period of claim 1, and wherein the confirmed diagnoses are produced by a human technician.

5. The system of claim 1 wherein the mathematical transform includes:

creating a template of QRS data from the electrocardiogram data and subtracting the template of QRS data from each of the plurality of heartbeats, and, a Wavelet transform, and, the health diagnosis to be tested is P wave flutter.

6. The system of claim 5 wherein the library of time varying physiological data with confirmed diagnoses is produced by collecting and interpreting time varying physiological data from the patient in a first time period, said first time period occurring prior to the time period of claim 1, and wherein the confirmed diagnoses are produced by a human technician.