Methods and systems for tracking physiological parameters of mother and fetus during pregnancy

Abstract

The invention provides systems and methods for monitoring the wellbeing of a fetus by the non-invasive detection and analysis of fetal cardiac electrical activity data.

What is claimed is: 
     
         1 . A computer-implemented method, comprising:
 receiving, by at least one computer processor, electrocardiogram (ECG) signals data from at least one pair of ECG sensors positioned on an abdomen of a pregnant human subject;   filtering, by the at least one computer processor, the ECG signals data to produce filtered ECG signals data;   detecting, by the at least one computer processor, maternal heart peaks in the filtered ECG signals data, the maternal heart peaks forming a maternal ECG signal;   subtracting, from the filtered ECG signals data, by the at least one computer processor, the maternal ECG signal to produce corrected ECG signals data;   extracting, by the at least one computer processor, raw fetal ECG signals data from the filtered N-ECG signals data based at least on the corrected ECG signals data;   improving, by the at least one computer processor, a signal-to-noise ratio of the raw fetal ECG signals data to form filtered fetal ECG signals data;   detecting, by the at least one computer processor, fetal heart peaks in the filtered fetal ECG signals data;   calculating, by the at least one computer processor, based on the detected fetal heart peaks, at least one of:   i) a fetal heart rate,   ii) a fetal heart curve,   iii) a beat-to-beat fetal heart rate, or,   iv) fetal heart rate variability; and   outputting, by the at least one computer processor, a result of the calculating step.   
     
     
         2 . The computer-implemented method of  claim 1 , wherein the subtracting is performed by using a non-linear subtraction procedure. 
     
     
         3 . The computer-implemented method of  claim 2 , wherein the non-linear subtraction procedure comprises iteratively performing:
 i) automatically dividing each ECG signal of ECG signals data into a second plurality of ECG signal segments,
 1) wherein each ECG signal segment of the second plurality of ECG signal segments corresponds to a beat interval of a full heartbeat; 
 2) wherein each beat interval is automatically determined based, at least in part, on automatically detecting an onset value and an offset value of the respective beat interval, and 
 3) wherein the automatically detecting of the onset value and the offset value of each beat interval is based, at least in part, on:
 a) a number of maternal peaks detected in each ECG signal, 
 b) changing a sample rate of each ECG signal by an integer number, 
 c) an onset of each P-wave, 
 d) an offset of each T-wave, and e) in-beat segmentation; 
 
   ii) automatically modifying each of the second plurality of ECG signal segments to form a plurality of modified ECG signal segments, wherein the modifying is performed using at least one inverse optimization scheme based on a set of parameters, wherein values of the set of parameters are determined based on iteratively performing:
 1) defining a global template based on a standard heartbeat profile of an adult human being; 
 2) setting a set of tentative values for a local template for each ECG signal segment; and 
 3) utilizing at least one optimization scheme to determine an adaptive template for each ECG signal segment based on the local template being matched to the global template within a pre-determined similarity value; and 
   iii) automatically eliminating the modified segments from each the ECG signals, by subtracting the adaptive template from each ECG signal thereby generating each corrected ECG signal, thereby generating corrected ECG signals data.   
     
     
         4 . The computer-implemented method of  claim 3 , wherein the pre-determined similarity value is based on an Euclidian distance, wherein the set of parameters is a local minima solution to a non-linear least squares problem solved by at least one of:
 1) minimizing a cost function taken as the Euclidian distance;   2) utilizing a Gauss-Newton algorithm;   3) utilizing a Steepest-Descent (Gradient-Descent) algorithm; or   4) utilizing a Levenberg-Marquardt algorithm.   
     
     
         5 . The computer-implemented method of  claim 1 , wherein the step of improving the signal-to-noise ratio is performed by
 i) applying a band-pass filter within a range of 15-65 Hz to break the plurality of fetal ECG signals in to a plurality of frequency filtered channels,   ii) scoring a fetal ECG signal per channel based on a peak-2-mean analysis to identify a plurality of fetal heartbeat channels, wherein each fetal heartbeat channel corresponds to a particular fetal ECG signal; and   iii) selecting the identified plurality of fetal heartbeat channels into filtered fetal ECG signals data, comprising a plurality of filtered fetal ECG signals, each of the filtered fetal ECG signals corresponding to one of the fetal ECG signals.   
     
     
         6 . The computer-implemented method of  claim 1 , wherein the step of filtering the ECG signals data comprises digital signal filtering. 
     
     
         7 . The computer-implemented method of 6, wherein the digital signal filtering utilizes at least one of: i) a baseline wander filtering, ii) a power line frequency filtering, iii) a high frequency filtering, and iv) a digital adaptive inverse-median filtering. 
     
     
         8 . The computer-implemented method of  claim 7 , wherein the processing of the raw fetal ECG signals data to improve the signal-to-noise ratio further comprises at least one of: 1) utilizing a Singular-Value-Decomposition (SVD) technique; and 2) utilizing a Wavelet-Denoising (WD) technique. 
     
     
         9 . The computer-implemented method of  claim 7 , wherein the digital adaptive inverse-median filtering comprises utilizing an adaptive median filter, having a length to be either: 1) a constant value, or 2) adapted depending on at least one local characteristic of the ECG signals data. 
     
     
         10 . The computer-implemented method of  claim 9 , wherein the at least one local characteristic is a duration of at least one of:
 a. a QRS complex;   b. a ST segment; or   c. a PR interval.   
     
     
         11 . The computer-implemented method of  claim 1 , wherein the result of the calculating step is utilized to determine the health of a fetus. 
     
     
         12 . A system, comprising:
 a sensor device comprising a plurality of ECG sensors,
 wherein the sensor device is configured to be worn by a pregnant human subject so as to position the plurality of ECG sensors on the abdomen of the pregnant human subject; and 
   a specifically programmed computing system, comprising:
 a non-transient memory storing particular computer-executable program code; and 
 at least one computer processor that, when executing the particular program code, becomes a specifically programmed computer processor that is configured to perform the computer-implemented method of  claim 1 .