Non-invasive method and system for characterizing cardiovascular systems

• US 9,655,536 B2
• Filed: 03/04/2016
• Issued: 05/23/2017
• Est. Priority Date: 08/17/2012
• Status: Active Grant

• Current Assignee: Analytics For Life
• Original Assignee: Analytics For Life
• Inventors: Gupta, Sunny, Ramchandani, Shyamlal, Burton, Timothy William Fawcett, Exner, Derek Vincent, Yazdi, Mohsen Najafi
• Primary Examiner(s): MAHMOOD, NADIA AHMAD
• Application Number:
  US15/061,090
• Publication Number:
  US 20160183822A1
• Time in Patent Office: 445 Days
• Field of Search: 600523, 600508, 600509
• CPC Class Codes: A61B 2576/023 for the heart
• CPC Class Codes: A61B 5/0205 Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
• CPC Class Codes: A61B 5/021 Measuring pressure in heart or blood vessels A61B5/0205 takes precedence
• CPC Class Codes: A61B 5/14551 for measuring blood gases
• CPC Class Codes: A61B 5/316 Modalities, i.e. specific diagnostic methods
• CPC Class Codes: A61B 5/318 Heart-related electrical modalities, e.g. electrocardiography [ECG]
• CPC Class Codes: A61B 5/339 Displays specially adapted therefor
• CPC Class Codes: A61B 5/341 Vectorcardiography [VCG]
• CPC Class Codes: A61B 5/7235 Details of waveform analysis A61B5/349 takes precedence
• CPC Class Codes: G16H 30/40 for processing medical images, e.g. editing

Abstract

The present disclosure uses physiological data, ECG signals as an example, to evaluate cardiac structure and function in mammals. Two approaches are presented, e.g., a model-based analysis and a space-time analysis. The first method uses a modified Matching Pursuit (MMP) algorithm to find a noiseless model of the ECG data that is sparse and does not assume periodicity of the signal. After the model is derived, various metrics and subspaces are extracted to image and characterize cardiovascular tissues using complex-sub-harmonic-frequencies (CSF) quasi-periodic and other mathematical methods. In the second method, space-time domain is divided into a number of regions, the density of the ECG signal is computed in each region and inputted into a learning algorithm to image and characterize the tissues.

Claims

1. A method for localizing and characterizing both the architectural features and function of cardiovascular tissues, comprising the steps of:

obtaining data associated with measured cardiovascular physiological signals;

processing the data to display an abnormality associated with the tissues in a 3-D image without use of invasive procedures, wherein the processing comprises;

creating a phase space diagram based on the data;

dividing the phase space diagram into a plurality of regions; and

computing one or more space-time density values of each divided regions, wherein the one or more space-time density values contain specific information about non-linear variability of the physiological signal; and

linking, via learning algorithms, one or more nonlinear nested sinusoidal Gaussian equations to a plurality of locations of the heart based on the one or more space-time density values, each location being associated with an architectural feature or function of the tissues to display the abnormality associated with the tissues in the 3-D image.

2. The method of claim 1, wherein the plurality of regions comprise 12 regions for ventricular tissues.

3. The method of claim 2, wherein the 12 regions for the ventricular tissues comprise an associated 12 quantities of space-time density values.

4. The method of claim 3, wherein the step of processing the data further comprises:

computing spatial changes in a phase space matrix comprising the 12 quantities of space-time density values based on non-Fourier multi-dimensional fractional integral summation across leads that acquired the cardiovascular physiological signals.

5. The method of claim 3, wherein the step of processing the data further comprises:

computing spatial changes in a phase space matrix comprising the 12 quantities of space-time density values based on Fourier multi-dimensional fractional integral summation across leads that acquired the cardiovascular physiological signals.

6. The method of claim 1, wherein the step of processing the data further comprises:

computing dynamical signal density using non-Fourier or Fourier n dimensional fractional integral summation across leads that acquired the cardiovascular physiological signals, wherein the order of the fractional integral is in the range of −

1.5 to −

2.5.

7. The method of claim 1, wherein the step of processing the data further comprises:

computing dynamical signal density using non-Fourier or Fourier n dimensional fractional integral summation across leads that acquired the cardiovascular physiological signals, wherein the order of the fractional integral is an irrational number.

8. The method of claim 1, wherein the step of processing the data further comprises:

computing dynamical signal density using non-Fourier or Fourier n dimensional fractional integral summation across leads that acquired the cardiovascular physiological signals, wherein the order of the fractional integral is a complex number.

9. The method of claim 1, wherein the step of processing the data further comprises:

computing dynamical signal density using non-Fourier or Fourier n dimensional fractional integral summation across leads that acquired the cardiovascular physiological signals, wherein the order of the fractional integral is a real number.

10. The method of claim 1, wherein the plurality of regions comprise 6 regions for atrial tissues.

11. The method of claim 10, wherein the 6 regions for the atrial tissues comprise an associated 6 quantities of space-time density values.

12. The method of claim 1, wherein the plurality of locations correspond to a 17-segment model of the heart.

13. The method of claim 1, wherein the one or more nonlinear nested sinusoidal Gaussian equations are used to produce a probability value of the abnormality associated with the tissues being present.

14. The method of claim 1, wherein the abnormality associated with the tissues is associated with a pathophysiological abnormality selected from the group consisting of hypertrophy, atrophy, scar, ischemia, edema, and fibrosis.

15. The method of claim 1, wherein the data associated with cardiovascular physiological signals comprise high resolution ECG data.

16. The method of claim 1, wherein the learning algorithms comprise a genetic algorithm.

17. The method of claim 1, wherein the data associated with the cardiovascular physiological signals are acquired via a lead selected from the group consisting of a single lead ECG, a 3 lead ECG, and a 12 lead ECG.

18. The method of claim 1, wherein the step of processing the data further comprises:

removing a baseline wander from the data.

19. The method of claim 1, wherein the step of processing the data further comprises:

processing the data with a Wavelet Packets and using the resulting data as a basis for the created phase space diagram.

20. The method of claim 1, wherein the step of processing the data further comprises:

processing the data with a Cosine Packets and using the resulting data as a basis for the created phase space diagram.

21. The method of claim 1, wherein the step of processing the data further comprises:

processing the data with one or more Chirplets and using the resulting data as a basis for the created phase space diagram.