Fetal heart rate extraction within a processor constrained environment
AssigneeOwlet Baby Care, Inc.
InventorAjay Sundaram Iyer
Life of this patent
Estimating the term…
Abstract
An abdominal electrocardiogram processing system for determining a fetal heart rate of a fetus in a pregnant woman obtains a combined electrocardiogram measurement of a maternal heart rate and a fetal heart rate. The system then transforms the combined electrocardiogram measurement into a wavelet domain to create a combined electrocardiogram measurement wavelet transform. The system can then remove the maternal heart rate from the combined electrocardiogram measurement wavelet transform. The system then identifies a fetal heart rate from the compensated combined electrocardiogram measurement wavelet transform.
Claims20
An abdominal electrocardiogram processing computer system for determining a fetal heart rate of a fetus in a pregnant woman, the computer system comprising:
- one or more processors;
- and
- one or more computer-readable media having stored thereon executable instructions that when executed by the one or more processors configure the computer system to perform at least the following:
- obtain a combined electrocardiogram measurement of a maternal heart rate and a fetal heart rate, wherein the combined electrocardiogram measurement is sampled at a first sample rate;
- transform the combined electrocardiogram measurement into a wavelet domain to create a combined electrocardiogram measurement wavelet transform;
- determine an estimate of the maternal heart rate from the combined electrocardiogram measurement wavelet transform;
- remove the maternal heart rate from the combined electrocardiogram measurement wavelet transform using the estimate of the maternal heart rate to create a compensated combined electrocardiogram measurement wavelet transform;
- identify a fetal heart rate from the compensated combined electrocardiogram measurement wavelet transform, wherein identifying a fetal heart rate comprises:
- identifying at least one fetal wavelet transform modulus maxima in a particular detail level of a wavelet transform of the compensated combined electrocardiogram measurement wavelet transform,
- determining one or more fetal features of the at least one fetal wavelet transform modulus maxima,
- evaluating each fetal feature of the one or more fetal features against one or more predetermined criteria to identify a plurality of candidate fetal peaks,
- determining a set of preliminary fetal peaks using a probabilistic likelihood search of the plurality of candidate fetal peaks,
- determining a kurtosis score and a signal score of each preliminary fetal peak in the set of preliminary fetal peaks for a given time step, and
- selecting a set of final fetal peaks with kurtosis scores in a first predetermined range and signal scores in a second predetermined range;
- and
- display an indication of a final fetal heart rate based on the set of final fetal peaks.
A method for processing abdominal electrocardiogram signals comprising:
- obtaining a combined electrocardiogram measurement of a maternal heart rate and a fetal heart rate, wherein the combined electrocardiogram measurement is sampled at a first sample rate;
- transforming the combined electrocardiogram measurement into a wavelet domain to create a combined electrocardiogram measurement wavelet transform;
- determining an estimate of the maternal heart rate from the combined electrocardiogram measurement wavelet transform;
- removing the maternal heart rate from the combined electrocardiogram measurement wavelet transform using the estimate of the maternal heart rate to create a compensated combined electrocardiogram measurement wavelet transform;
- identifying a fetal heart rate from the compensated combined electrocardiogram measurement wavelet transform, wherein identifying a fetal heart rate comprises:
- identifying at least one fetal wavelet transform modulus maxima in a particular detail level of a wavelet transform of the compensated combined electrocardiogram measurement wavelet transform,
- determining one or more fetal features of the at least one fetal wavelet transform modulus maxima,
- evaluating each fetal feature of the one or more fetal features against one or more predetermined criteria to identify a plurality of candidate fetal peaks,
- determining a set of preliminary fetal peaks using a probabilistic likelihood search of the plurality of candidate fetal peaks,
- determining a kurtosis score and a signal score of each preliminary fetal peak in the set of preliminary fetal peaks for a given time step, and
- selecting a set of final fetal peaks with kurtosis scores in a first predetermined range and signal scores in a second predetermined range;
- and
- displaying an indication of a final fetal heart rate based on the set of final fetal peaks.
An abdominal electrocardiogram processing computer system for determining a fetal heart rate of a fetus in a pregnant woman, the computer system comprising:
- one or more processors;
- and
- one or more computer-readable media having stored thereon executable instructions that when executed by the one or more processors configure the computer system to perform at least the following:
- obtain a combined electrocardiogram measurement of a maternal heart rate and a fetal heart rate, wherein the combined electrocardiogram measurement is sampled at a first sample rate;
- transform the combined electrocardiogram measurement into a wavelet domain to create a combined electrocardiogram measurement wavelet transform;
- determine an estimate of the maternal heart rate from the combined electrocardiogram measurement wavelet transform, wherein determining the estimate of the maternal heart rate from the combined electrocardiogram measurement wavelet transform comprises:
- identifying at least one maternal wavelet transform modulus maxima in a first detail level of the combined electrocardiogram measurement wavelet transform,
- determining one or more maternal features of the at least one maternal wavelet transform modulus maxima,
- evaluating each maternal feature of the one or more maternal features against one or more predetermined criteria to identify a plurality of candidate maternal peaks,
- obtaining a set of windowed candidate maternal peaks, where each windowed candidate maternal peak in the set of windowed candidate maternal peaks is a window of a first number of samples around a candidate maternal peak of the plurality of candidate maternal peaks,
- determining a set of template matching scores for the set of windowed candidate maternal peaks, wherein each template matching score of the set of template matching scores is found using a distance metric between each windowed candidate maternal peak of the set of windowed candidate maternal peaks and a maternal heart rate template,
- determining a set of maternal peaks using a probabilistic likelihood search of the plurality of candidate maternal peaks that is based at least on the set of template matching scores,
- updating one or more parameters of the maternal heart rate template based at least on the set of maternal peaks to create an updated maternal heart rate template, and
- determining an estimate of the maternal heart rate based on the set of maternal peaks and the updated maternal heart rate template;
- remove the maternal heart rate from the combined electrocardiogram measurement wavelet transform using the estimate of the maternal heart rate to create a compensated combined electrocardiogram measurement wavelet transform;
- identify a fetal heart rate from the compensated combined electrocardiogram measurement wavelet transform, wherein identifying a fetal heart rate comprises:
- identifying at least one fetal wavelet transform modulus maxima in a particular detail level of a wavelet transform of the compensated combined electrocardiogram measurement wavelet transform,
- determining one or more fetal features of the at least one fetal wavelet transform modulus maxima,
- evaluating each fetal feature of the one or more fetal features against one or more predetermined criteria to identify a plurality of candidate fetal peaks,
- determining a set of preliminary fetal peaks using a probabilistic likelihood search of the plurality of candidate fetal peaks,
- determining a kurtosis score and a signal score of each preliminary fetal peak in the set of preliminary fetal peaks for a given time step, and
- selecting a set of final fetal peaks with kurtosis scores in a first predetermined range and signal scores in a second predetermined range;
- and
- display an indication of a final fetal heart rate based on the set of final fetal peaks.
Dependencies are inferred from the claim text.