WO2016067276A1

WO2016067276A1 - Continuous non-invasive fetal/mother monitoring using fixed location sensors

Info

Publication number: WO2016067276A1
Application number: PCT/IL2015/050407
Authority: WO
Inventors: Oren Oz; Nathan Intrator
Original assignee: Nuvo Group Ltd
Current assignee: Nuvo Group Ltd
Priority date: 2014-10-28
Filing date: 2015-04-15
Publication date: 2016-05-06
Anticipated expiration: 2017-04-28

Abstract

A system, for continuous non-invasive: monitoring of one or more bloparameters of a fetus using multiple fixed location sensors, comprises a holder unit configured to hold the sensors adjacent a mother's body, one or more sensors of the first sensor modality fixedly placed at one or more: separate locations so as to: continually sense, at repeating time intervals, and produce at least one first signal, one or more sensors of the second sensor modality fixedly placed at one. or more separate locations to produce at least one second signal partially independent of the: at' least one first signal. One' or more processors configured to receive the first and second signals, generate a combined signal, from a combination of the at least one first signal and the at least one second signal and. process the combined signal to- produce an output indicating the one or more bioparameters.

Description

APPLICATION FOR PATENT lnvetttor(s): Oren Oz; Nathan Intrator

Title: Continuous Non-invasive Fetal/Mother Monitoring Using Fixed Location Sensors Field and Background of the invention

The present invention relates to apparatuses, systems and methods for continuous non-invasive monitoring: of a fetes and or mother, and more particularly, to such continuous or continual monitoring from multiple sensors held preferably adjacent the mother, to passively collect data continuously, in one preferred embodiment in a manner mat may be used by a mother at home. In some preferred embodiments, monitoring may include the utilization of sensors of multiple modalities,: such as both acoustic and electric signal sensors.

Fetal monitoring, besides needing to be sale, also has to contend with the following facts that make its effectiveness challenging.- First, the fetus moves freely in the womb. Accordingly; the fetal heart is very small, is located In fluid and is constantly moving. If the sensors are fixed in place, the fetal heart is moving with respect to the .sensors, therefore, no fixed transfer function can be used as a model for the fetal acoustic activity or the electrical activity. Second, the placenta has a -scattering effect on a fetal electrocardiogram (ECG). Third, the gastric activity sounds are loud during digestion and can mask the fetal cardiac sound completely and possibly saturate the acoustic sensor of these sounds. Fourth, the fetal electrocardiogram activity may be masked by the EMG signal emanating front the mother's abdominal skeletal muscles. The muscle activity is very strong as it has to support the extra weight of the extended womb. There is a -compelling need for a system ..and method (hat. can overcome the above problems and can safely monitor the fetus and/or mother accurately and such that the signals obtained more robustly correlate with the actual bioparamelers.

BRIEF DESCRIPTION OF THE DRAWINGS

Various embodiments are herein described, by way of example only, with reference to the accompanying drawings, wherein:

FIG . 1 is a schematic illustration of a medical grade muKi-rnodal remote continuous prenatal health & wellness monitoring, in accordance with one embodiment of the present invention;

FIG. 2 is a representati ve list of highlights, in accordance with one embodiment of the present i vention;

FIG. 3 is a schematic representation, of a system for fetal monitoring and processing, in accordance with one embodiment of the present invention;

^' FIG..4 is a schematic representation, of a wearable holder unit for sensors,, in- accordance with one embodiment of the present invention;

FIG. 5 is a schematic representation of a consumer continuous passive monitoring product, in accordance with one preferred embodiment of the present invention; FIG. 6 is a schematic representation of the system of FIG. 5 with certain highlights;

FIG. 7 is a schematic illustration of a fetal heart in different positions relative to sensors of different modalities, in accordance with one embodiment of the .present invention;

FIG. 8 is a schematic, depiction of elements of the system or device of the present invention, in accordance with one embodiment;

Ί FIG. 9 is a low chart showing a method of the present invention;

FIG. 10 depicts a multiple sensor analysis, in accordance with one embodiment of the present invention; and

FIG. ί .1 is a -flow chart showing a further method of the present invention.

DETAILED DESCRIPTION OF THE INVENTION

The following detailed description is of the best currently contemplated modes of ^•carrying out. the invention. The description is not to be taken in a limiting sense, but is made merely for the purpose of illustrating the general principles of the invention.

The present invention generally provides continuous passive monitoring of a fetus using multiple sensors enabling continuous monitoring of fetal and mother heart activity, including but not limited to heart rate and heart rate variability during pregnancy, in the case of home monitoring, processed, measurements from home monitoring can deli ver to doctors and patients more in-depth, real-time indications of the well-being of both the mother and fetus. From a doctor's perspective, the system can collect, monitor and analyze data from: a large cohort of pregnancies to obtain more detailed baseline pregnancy data representing normal conditions of the fetus and/or mother as well as correlation function between fetal and mother heart activity and well being. Such baseline data can enable early borne detection of potential danger or abnormal behavior during the pregnancy. Fro the pregnant mother's viewpoint, the information can hel in reducing pregnancy associated stress and risks related to nutrition, physical activity or lack thereof: sleep and other life-style factors that may affect the pregnancy. According to the present invention, all this can be achieved, with a mobile/remote continuous and passive monitoring, which does not pose discomfort or risk to the fetus and the mother. The parameters that may be monitored typically but not necessarily relate to vital signs and may include, but are not limited to, one or more of the following fetal bioparameters: cardiac activity, heart rate, cardiac vibro aeoustic sounds, pulse rate, heart rate variability,, bradycardia event, tachycardia event, desaturation, fetal movement, fetal position/orientation, fetal sounds, fetal kicks, fetal brain activity fetal temperature, glucose, sleeping state, blood flow and blood pressure, activity level and/or one or more of the following bioparameters of the mother: cardiac activity, ejection fraction, heart rate, heart sounds, breathing depth and duration, pulse rate, heart rate variability, bradycardia event, tachycardia event, desaturation, brain acti vity, temperature, glucose, inflammation level, uterus activity, sleeping stages, blood pressure, physical activity level, posture, body position.

The present, invention may utilize an array of fixed location sensors placed at various points on the mother's body, for example her abdomen, in one preferred embodiment, the present invention may utilize both acoustic and electric sensors for monitoring cardiac activity. In one preferred embodiment, the present in vention utilizes two or more, and preferably four, fixed location acoustic sensors and two or more, and preferably four, electric signal sensors such as ECG sensors, All the sensors may be. integrated into a holding unit that may be wont by the mother, The holder unit; may be a state of the art pregnancy belt that is wearable (i.e. by the mother) and washable. It may have a unique textile with embedded wiring. The belt may be irritation-free. These features are to enable practical continuous monitoring, thai is comprehensive and hassle free for the mother. In contrast to prior art monitoring systems, and in order to provide a reliable: readout of fetal cardiac activity with fixed location sensors, the present invention addresses the following facts;

A. The fetus moves freely in the womb. Accordingly, the fetal heart, which is very small in comparison with the mother heart, is located inside fluid and is constantly moving. Since the sensors are fixed i place for continuous monitoring, the fetal heart is mo ving with respect to the sensors.

B . The placenta- has a scattering effect on the fetal electrocardiogram (ECG).

C. The gastric act ity sounds are loud during digestion and can mask the fetal cardiac sound completely since an acoustic signal is very sensitive to gastric activity.

D, In addition to gastric internal sounds, acoustic monitoring i affected b external background sounds as well as the own c rdio respiratory arid gastric activity of the mother, as well as her vocal sounds.

n contrast to prior art monitoring systems, where application of Independent Components Analysis (ICA) to partiaily independent variables may involve variables whose independence derives from the: fact that they relate to different objects, for example a heart of a mother versus a heart of a fetus, or a heart of a mother versus a different organ of that mother, the system and method of the present invention fo monitoring a fetus may apply (ICA) to partially independent variables whose independence derives from the fact that they may relate to the same object but may be sensed by sensors having two or more different modalities. The two different modalities may be selected from among acoustic sensors, electric signal sensors (i.e. ECG), Dojppler, inffa red sensors or sensors of other modalities. For example, one or more acoustic sensors may sense a heart of fetus and one or more ECG sensors may sense: the same heart of the same fetus. ECG waveforms comprise electric cardiac signals. The acoustic sensors may be affected differently than the ECG sensors by the. envelope of fluid around the fetus that causes background noise, thereby creating partial independence between, a signal from the acoustic. sensors and^' a signal from the ECG sensors despite the fact that both types of signals emanate from the same fetal, heart. In furthe contrast to prior art monitoring systems that utilize 1CA, the present invention may input one or more processors with a combined signal made of adding together, or otherwise combining, a: first signal from one or more sensors of a first sensor modality type and a second signal from one o more sensors of a second sensor modality type. For example, the first and second signals may he combined by adding a magnitude of the first signal at a given point in time to the .magni tude of the second signal at the: same point in. time to provide a. combined signal. In contrast to prior art fetal monitoring systems and methods, the present: invention may fuse different cardiac (or other vital sign c«- bioparameter) activity sensing modalitie to provide die most robust correlation to actual cardiac, (or other vital sign or bioparameter) activity of the fetus and/or the .mother, i.e. more robust than "by sensing either acoustic signals alone or electric signals alone, in further contrast to prior art monitoring systems, the present invention is non-invasive and passive. In still further contrast to prior art fetal monitoring, the present nvention, may involve fixed location acoustic sensors and fixed location electric signal sensors, in further contrast to prior art fetal heart monitoring, the present invention enables continuous fetal heart monitoring, by performing real-time optimization of a transfer function from the sensed signals- to fetal and mother cardiac activity, The principles and. operation of a: system and method for Continuous Non-invasive Fetal/Mother Monitoring Using Fixed Location Sensors, according; to the present invention may be better understood with reference to the drawings and the accompanying description. Since sound propagates well in fluid, the amniotic fluid surrounding the developing embryo should be able to relay the vtbro-acoustie cardiac sound well (with minimal lass), notwithstanding the difficulty resulting from embryonic movemen Since there is little acoustic mismatch between the fetus and. the placenta and between the placenta and the belly surface, i is expected that the acoustic fetal cardiac waveforms will Suffer little loss even when the fetus is away front the sensors. On the negative side, the acoustic signal is sensitive to gastric activity, which produces sounds that interfere with the fetal cardiac sounds. External background sounds, mother cardiac sounds, as well as mother vocal sounds also interfere with the fetal cardiac sounds. The cardiac electrical activity, on the other hand, is not affected by the gastric activity sounds, or the other described sounds (except of course the mother's cardiac eieetrieal activity) and therefore., represents a modality that can provide independent cardiac information. A similar approach is made monitoring different bioparameters and/or usin a different -pair (or more) of^■sensors having different sensor modalities.

In a preferred ^' embodiment, the combined acoustic and electric signals are fused by applying Independent Components Analysis, They can be fused by other (nonlinear) optimization algorithm which attempts to separate between fetal and mother cardiac activity based on a different criteria of deviation from; Gaussian distribution, which is the criteria employed by ICA. In ICA, the fact that additive partially independent random variables become more Gaussian when added together is utilized. A linear transformation is. sought, which results in random variables, which re less Gaussian than the..distribution of the original random variables. ICA may be applied to a combination of two or more signals sensed by one or more sensors from one or more modalities. ICA. may also be used on processed sensors' data. Such processing includes but is not limited to-removing the mother's ECO by other means such as subtraction of a template pattern-. The larger the number of sensors, the larger the separation of signals into independent components is, and the better the cardiac common signal. -will become. We have found that fusing signals using their spectral representation increases robustness of the fusion.

Cardiac activity produces acoustic and electric signals. While these signals ar different in thei shape and frequency range, acoustic cardiac signals and electric cardiac signals are con-elated because they emanate from the same heart, if the heart of the fetus is. being monitored, or the same two hearts, if both the fetus's heart: and the mother's heart are monitored.

Electric and acoustic signals, for example electric and acoustic cardiac signals, propagate differently in the body: the acoustic signal propagates well in the amniotic fluid, with minimal, scattering. The electric signal is more scattered. The acoustic signals have lower SNR, while the electric signals have higher SNR since there is less noise interference in the electric signals,.

Thus, depending- on the location of the fetal heart with respect to the sensors, the ECG and acoustic signals may be sensed -with a different signal to noise ratios (SNR), For example, if the fetus is close to the electrodes, the ECG signal will be stronger. Jfthe fetus is far from the electrodes and there is a lot of fluid between the- fetus and the electrodes, the- acoustics signal maybe stronger. The .strength of the- signals also depends on the strength of the fetal and mother electrical, and acoustic activity .

Therefore, when fusing both together, the result, which includes a .component that is correlated with the fetal cardiac activity, will sometimes get the fetal cardiac activity component more from the EGG and sometimes more from the acoustic signal. By applying iCA to the electric signals and the acoustic signals together, one may obtain an output, that includes components that are more correlated with the fetal cardiac activity than with the acoustic activity and vice versa,

in one preferred embodiment, CA finds a matrix A which transfers a^' high- dimensional time series into a new high dimensional time series X i] ~ AY(^'t), where the different, time series components: Xit't) are more independent, namely: P(Xi,Xj) is roughly P(Xi P(Xj). In our case, the original acoustic components of the time series may include the combined cardiac acoustic sound of the mother and the fetus, together with the acoustic sound from the background noise and from gastric activity. The original components of the electric signals include the EGG component of the mother and the. fetus together with signals due to muscle activity. The underlying assumption is that the short-term cardiac activity- of the mother is independent from the fetal cardiac activity. Thus, ICA attempts to separate the mother cardiac activity components from the fetal .ones, and on the way, will separate background noise, gastric activity and muscle activity. Since the acoustic, and electric cardiac activity components of the mother are correlated - because they emanate from the same heart ~ and since the acoustic and electric cardiac activity components, of the fetus- are correlated— because they emanate from the: same fetal heart -- in a preferred embodiment when applying ICA to the combined acoustic and electric signal,- the result is hybrid components thai were fused from acoustic and electrical activity, but that are more robustly correlated to either the fetal or the mother cardiac activity (than acoustic components alone and than electric components alone).

Since we do not know a~priori which components will be better picked-up by the sensors, because that depends upon the location of the fetus relative to the particular fixed acoustic or electric signal sensors, applying ICA o the combined acoustic and electric signals may ensure that the separation into fetal and mother components is more likely to take place because the chance of any one of the modalities to produce the desired separation is hi gher than the chance of one of the modalities only.

ICA may be applied in a dynamic way, namely, to optimize the decomposition of the combined signal at discrete time intervals (which may be regular intervals or may be irregular intervals), for example, every 10 second, every 60 seconds, every few minutes. The duration at which to perform the optimi zation again, and the duration, of time on which the optimization is performed, e g, the last minute of recording, depends on the computational and power resources at hand. Accordingly, soon after the fetus moves, the parameters of said matrix resulting from the optimization change, to account for the fact that since the last optimization tire fetus is now located at different distances with respect to one or more (in some cases, all) sensors of the set of sensors.

While in one preferred embodiment, the ICA is performed on a time series of the signal, in a further preferred embodiment, spectral ICA is performed on the spectral representation of the signal. ICA ca be performed o any full or partial decomposition of the signal in particular orthogonal decomposition. It c an also be performed on any

1.0 sparse representation of the signal. Thus, for any version of the present invention, the type of TCA utilized is .not necessarily time series and is not necessarily spectral representation.— it can he either one including but not limited to said list, The specific representation may be chosen based on invariance features that are sought, for example, performing the analysi on a complex wavelet analysis representation of the signal improves the resilience to changes in the phase of the different signals with respect, to one another.

In one preferred embodiment depicted schematically in. FIG. 8, the present, invention may be^' characterized as-: a system 10 for continuous■ no in asive monitoring of one or more bioparanieters of -a fetus (see fetal hear 66 in FIG. 8) using multiple fixed location sensors. The one or more bioparameters may he cardiac bioparanieters. The one or more cardiac bioparameters may comprise heart rate variability. The one or more bioparameters may be: one or more bio aranieters measurin the brain or another orga of the fetus (or of the mother). In a preferred embodiment, the one or more bioparameters target a specific organ of the fetus (or mother) such that there is an inherent: partial dependence of the sensors of different modalities since the data btained from, the different modalit sensors still both emanate from the same organ of the same fetus (or person). The partial independence of the sensors of the different modalities, on the other hand, . may be generated by different reactions to noise or other errors either in transmission of the data to the sensor or other causes of differences in the way sensors of different modalities are affected.

System 10 may comprise (a) a holder uni 50 configured to hold adjacent to- a sensor area: 40 of body of a mother one or more sensors of a first sensor modality monitoring the one or more bioparameters (for example cardiac bioparameters) and one or more sensors of a second sensor -modality monitoring- the one or more bioparameters (for example: cardiac: bioparameters). Holder unit 50 may be wearable.

System 10 may further comprise one or more sensors 60 of the first sensor modality fixedly placed at one or more separate locations, for example m a sensing area 40 of the mother's abdomen, so as to repeatedly sense, at regular or irregular time intervals, the one or more cardiac bioparameters and produce at least, one first signal 61. The sensors 60 ma be fixed at th sensing area 40 by the holder unit 50,

System 10 may further comprise one or more sensors 70 of the second sensor modality fixedly placed at one or more separate l cations, for e ample in the sensing area 40 of the mother's body, so as to repeatedly sense, at regular (continual) or irregular time intervals, the one or more cardiac bioparameters of the -fetus (that ma be sensed by the at least one sensor of the first sensor modality) so as to produce at least one second signal 71. The at least on second signal 71 may be partially independent of the at least one first signal 61. The sensors 60 may be fixed at the: sensing area 40 by the holder unit 50. System 10 may include surround speakers situated on each sensor (or on fewer than all sensors ) of the one or more sensors of the first sensor modality and/or on each sensor (o OR fewer than all sensors) of the one or more sensors of the second sensor modality.

^'The system and method of the present invention, in other preferred embodiments, utilize sensors that are of more than two different sensor modalities, for example three different sensor modalities, four different sensor modalities, five different sensor modalities (or more), in one preferred embodiment, the system or meihod utilizes one or- more sensors of a first modality and one or more sensors of a second modality for in other cases also one or more sensors of a third modality or in other cases also one or more sensors of a fourth modality for even more)), wherein the modalities are selected from acoustic, ECG, ultrasound, Doppler, ultrasound/Ttoppler (which may be described as high frequency acoustic signals with sensors using Doppler technology) and FN1RS (which stands for functional near infra red spectroscopy).

As shown in FIG. 8, system 1-0 may also include one or more processors 80, accompanied by program instructions or software 83, configured to receive the first and second signals, generate a combined signal 83 from a combination of the at least one first signal 61 and the at least one second signal 71 and process the combined signal to produce an output indicating the one or more cardiac bioparameters, System 10 may include a display 30 for displaying output of the measured, bioparameter. The display 30, in some preferred embodiments, may be integrated with the holder unit 50 or with the one or more processors 80 (see FIG. 8). in other preferred embodiments, display 30 may be integrated into a mobile communication device 20 (see FIG, 3),

Although FIG, 8 does not show a physical connection between holde unit 50 and the one or more processors 80 (since arrows 61, 71 are intended to depict wireless signals), the one or more processors 80 may be located remotely to the holder unit 50 or may be located on holder unit 50.. Furthermore, wires for transmission of signals 61 ,. 71 may also: be added where wired transmission is used instead of wireless transmission.

The at least one first signal 61 may have_. a noise, ^"N^'T, caused by one or more noisy background events. The at least one second signal 71 may have a noise N₂, caused by one or more noisy background events such that die noises Ni and ₂ are partially independent. For example, the background noise from within the fetus or mother may affect the first signal 61 differently than it. affects, the second, signal 71. For example, the at least one first signal may be an acoustic signal that propagates relatively well in the amniotic fluid, with minimal scattering. The at least one second signal 71 may be art electric signal that is more scattered than the at least one first signal 61 in the amniotic fluid. On the other hand, the at least one first, signal being acoustic signals may have lower signal to noise ratio (SNR), while the.- at least one second signal, bein electric signals, may have a higher signal to noise since there is less noise interference in the electric signals.

In case the fust signal (or the at least one first signal ) is an acoustic signal and the second signal (or the at least one second signal) is an ECG signal, the overall noise affecting these two signals of different modalities may be partiall independent because for example they may have independent and dependent components. The dependent components may be that both the first and second signals have noise derived from the heart of the mother. The independent component( s) ma be that the acoustic signal may have noise derived from gastrointestinal activity of the mother but do not have electric: noise in the muscle of the mother, whereas the ECG signal may have noise affecting i t from the electric noise i the muscle and does not have the gastrointestinal activity of the mother. The acoustic signal may also have noise caused by an external event, for example noise of people speaking outside the mother, which does not affect the ECG. signal.

In on preferred embodiment, the one or more sensors 60 of the first sensor modality are acoustic sensors that monitor cardiac activity of the fetus and the one or more sensors 70 of the second sensor modalit are electric signal sensors, for example ECG sensors, that monitor the cardiac activity of the fetus. In other preferred

embodiments, the one or more sensors of the first sensor modality 60 and second sensor modality 70 are chosen from among the following sensor modalities: acoustics sensors, ECG or other electric signal sensors, ultrasound, Doppler, ulirasound/Doppier sensors, FNI S or other infra: red sensors, other sensor modalities, "Uitrasouad Dop !er" means ultrasound sensors and sensors using Doppler technology.

Although of different modalities, the one or more processors of system 1 may combine the at least one first signal.61 with the at least one second signal 62. For example,, a combined signal 83 may he generated by the one or more processors 80 by adding a magnitude of the at least one first signal at a given point in time to a magnitude of^" the at least one second signal at the given point in time.

The combined signal may not resemble either the at least one first signal or the at least one second signal. The combined signal 83 may correlate with the one or more cardiac bioparanieters ^' more than either the at least one first signal or the at least one second signal.

One or more processor 82 may output a numeric or graphic indication, of the one or more cardiac bioparameters for example using display 30. The output produced and/or displayed by system 10 may be based in part on which sensor correlates best with the combined signal.

In a preferred embodiment, one .or more processors 80 are configured to process the combined signal 83 by performing independent component analysis (ICA)^'on the combined signal 83. 1CA may be applied to the combined signal dynamically at discrete time intervals so as to optimize- a- decomposition, of the .combined ..signal. System 10 may be configured to be^■ambulatory and. operable by a mother at home without medical personnel. For example, system 10 may incorporate a transmitter tor communicating the output of the one or more processors to a remote serv er via at least one of low emission Bluetooth. (Bluetoo th Low Energy) for communication of an: output of the processor to a remote server. System 10 may also include automatic gain control (AGO) to extend the sensors' maximum allowable variations in the signal amplitude.

Although system 10 has been described in terms of two sensor modalities, system 10 may utilize more than two sensor modalities.. For example, system 10 may include one or more sensors of a third sensor modality (not shown) fixedly placed at one or more separate locations so as to repeatedly sense, at regular (continual) or irregular time intervals, the one or more cardiac bioparameters of the fetus sensed by the at least one sensor of the first sensor modality and the at least one sensor of the second sensor modality so as to produce at. least one third signal that is partially independent of the at leas one first signal and of the at least one second signal. In one preferred embodiment, the present invention is a. system for continuous non-invasive (passive) monitoring of a fetus using multiple fixed location sensors, comprising:

(a) one or more sensors from a first modality, for example one or more acoustic sensors, preferably a plurality of acoustic sensors at multiple fixed locations for sensing, for example repeatedly sensing for example each minute or other interval, one or more .bioparameters. of the fetus and/or mother;

(b) one or more sensors from a second modality, for example electric signal sensors such as ECG sensors, .preferably a plurality of electric signal sensors at multiple fixed location^'s for sensing, for example repeatedl sensing for example each minute o other interval, the one or. more bioparameters of the- fetus .and or mother;

(c) a wearable holder unit housing the one or more acoustic sensors and the one -or more electric signal sensors such that the holder unit fits to the changing shape of the mother as the^■ pregnancy develops, in a manner that one or more sensors of each modality are adjacent to the mother's body or otherwise poised to performing their sensing function, in a preferred embodiment adjacent to her belly or abdomen. As described i FIGS, 3-4, the holder unit ma he a state of the art pregnancy belt that is wearable and washable. It may have a unique textile with embedded wiring. The belt may be irritatton-free, The holder unit may include other components such as low emission Bluetooth (Bluetooth Low Energy) and automatic gain control (AGC) for communication of an output of the processor to a remote Server/. As shown in FIG. 4, in one preferred embodiment, surround, speakers and other stimulation or sensor devices may be situated on each, sensor for improved sound transmission or sensing quality.

(d) one or more processors 80, accompanied by compatible program instructions or software 82, for processin the input data recei ed from the sensors, The processin may include performing independent component analysis (ICA) on. the combined sensor's from the first and second modalities and for outputting a numeric or graphic indication of the measured parameter. The ICA performed on the combination of sensors from both modalities may produce hybrid components (components from multiple modalities: such as acoustic and electric signal) that were fused together from; the combined input that included acoustic and electrical activity. This output should be particularly correlated to monitored bioparameter of the fetus (and mother). The one or more processors-may be located remotely and/or may be located locally, such as in or on or attached to or adjacent to or near the holder unit 50.

The device and/or system may include transmitters or other suitable hardware and/or software for transmitting the .signals sensed:. by the -sensors to the one or more processors.

The system may also include a feature wherein the one or more processors are programmed or configured to repeatedly calculate the monitored bioparameter over time in order to optimize the robustness of the result. This is performed as via ICA or other optimization program based on different optimizatio goal. As the fetus' heart (or brain or other organ or body part being monitored ) moves closer or farther away from sensors of one modality, for example acoustic sensors, and farther or closer to sensors of a second modality, for example electric signals sensors, the one or processors will output revised results that vary over time.

FIG. 10 depicts four ECG graphs and demonstrates the result that can be obtained after ICA decomposition. The ECG of the mother is strongly sensed in ail sensors and it can be observed in all four channels depicted in FIG:. 10. The top right graph of FfG. 10 includes a strong noise component that, after the ICA optimization, lias been pushed mainly to a single channel (i.e. that top right of FIG,.10). The to left and bottom left graphs in FIG. 10 represent two variations, one stronger .and one weaker, of the mother's ECG. The fetal ECG has been pushed to the right bottom channel. Accordingly, it is likely that the location of the heart of the baby is -closest to the area where the channel depicted at the right bottom graph is located. For example, in on version the system Includes eight sensors, including four acoustic sensors and four ECG sensors. As shown in FIG. 7 at a first point in time ("A") the heart 66 of the fetus is located closer to acoustic sensors 60 and at. a second point in ^■time ("¾"} fetal heart 66 is closer to electric signal sensors ^'70.

The present invention may also be described as a device if the one or more processors are local to and connected to the holder unit and/or sensors, if the one or more processors are remotely located, on the other hand, the present invention -may 'be described as a system that may include a device that may interact remotely with other el ements of the system.

As shown in FIG. 9. the presen invention, i a further preferred embodiment, is a first method 100 of continuou non-invasive (passive) monitoring of a fetus using multiple fixed location sensors. Method 1:00 may include a ste 1 } 0 of applying wearable holder unit to a pregnant mother's body in such a manner ^'that (a) one or more sensors from a first modality are applied to sense, for example repeatedly sense for example each minute or other interval, a mother' s body (such as her abdomen) over time, for example one or more acoustic sensors adj cent the skin, preferably a plurality of acoustic sensors at multiple locations,, to monitor one or more bioparameters of a fetus and/or mother; (h) and that one or more sensors from a second modalit are applied t sense, for example repeatedly sense for example eac minute or othe interval, the mother's body (such as her abdomen) over time, for example electric signal sensors such as ECG sensors adjacent the skin, preferably a plurality of electric signal sensors at multiple locations, t monitor the one or more bioparameters of the fetus and/or mother. Method 100 may- further include a ste 120 of the one or more sensors from the first and second modalities sending the sensed data signals to one or more processors as input data.

Method 1 0 may further include a. step 130 of the one or more processors, accompanied by compatible program instructions or software (for example special purpose software or program, instructions that transform the one or more processors into special-purpose processors), processing the input signals received from the sensors. The processing may include performing independent components -analysis (iCA) on the combined sensors from the first and second modalities and outputting a result.

Outputting: the result may Include outputting a numeric or graphic indication of the measured one or more hk>pararneter(s).

The one or more proe-essors may he located remotely and/or may he located locally, sneh. as in or on. or attached to -.or .-adjacent to or near the holder unit.

The present invention, in a still further preferred embodiment, is a second method 200 of continuous non-invasive (passi ve) monitoring of a fetus using .multiple fixed location sensors that may comprise:

(A) continuously or continually monitoring a fetus repeatedly at each time interval (for example, every second, every 15 seconds, every half minute, every three- quarters of a minute, every minute, ever minute and a quarter, every minute and a half, -every minute and three-quarters,, or every two minute or every 3 or 4 or 5 minutes) suc that;

(i) at a first point in time during the time interval using one or two or three or •four or more sensors from each of a first modali ty and a second modality (for example four- acoustic sensors and four electric signal sensors) to repeatedly measure a cardiac or other bioparameter of the fetus, (the sensors may be in a wearable hoider uni placed ^'on the mother.), wherein the signals from the acoustic sensors are of a first: strength (i .e. relatively strong) and the signals from the electric signal sensors are of a second strengt (i.e. relatively weak):

(ii) at a second point in time during the time interval using one or two or three -or four or more sensors from each- of a first modality and a second modality (for example four acoustic sensors and four electric signal sensors) to repeatedly measure a cardiac or other bioparameter of the fetus, (the sensors may be in a wearable hoider unit placed on the mother), wherein the signals from the acoustic sensors are relatively weak and the signals from the electric signal sensors are relatively strong (with respect, to their corresponding background' noise level);

(c) the one or more sensors from the first and second .modalities sending the sensed data, signals to one or more processors as input data;

the one or more processors, accompanied by compatible program instructions or software (for example special purpose software or program instructions that transform, the one or more processors into special-purpose processors), processing the input signals received from the sensors. The processing ma include performing independent components analysis (TCA) on the combined sensors from the first and second modalities and outputttng a result, which may include a numeric or graphic indication of the m ea si! red ar meter s ) .

The present invention may also be described as a method 300 of monitoring one or more bioparameters of a fetus, for example one or more bioparameters that include a ^•cardiac bioparameter. Method 300 may have a step.310 of applying a wearable holder unit 50 to an abdomen of a pregnant . woman/female so that two or more sensor of a first modality and two or more sensors of a second modality are configured to at least continually sense one or more bioparameters, for example cardiac bioparameters, of the fetus, thereby generating a first signal from sensors of the first .modality and a second signal from sensors of the second modality. In case the one or more bioparameters refers to one or more cardiac bioparameters, the two or more sensors of the first modality and the two or more sensors of the second modality may targe a heart of the fetus to generate the first signal and the second signal. I other preferred embodiments, a single sensor of a first modality and a single sensor of a second modality are configured to at least continually- sense one or more bioparameters.

in systems and methods of the present invention, the term "continually sense" one or more bi oparamet ers refers to continually (at regular or -predetermined intervals) or ■continuously (all the time) sensing the one or more bioparameters and any combination thereof. In one preferred embodiment, the sensing is continuous and a wearer (i.e.

pregnant mother) wears the holder unit 50 during which time the sensing by the. sensors takes pl ace while the wearer goes about doing other activities, even for example sleep, in another preferred embodiment, the sensing is continual and the wearer wears the holder unit 50 while going about other activities and the sensors are configured to sense at predetermined^' intervals, which may be regular intervals. For example, the sensors may be programmed to sense every minute over the course of a full 24 hour period.

Furthermore, any combination of continuous and continual is also encompassed by "continually sense". "Continuous" sensing is also subject- to the ability of the sensors. For example, even if a heart rate is continuous and the sensors are worn and in position to sense and tamed on, the sensors may not be able to sense literally continuously without a break or even every miUiontb of a second but. may need to sense at. lengthier intervals, for example every second or every minute. Of course, the present invention is not limited to a case where the wearer does other aeti vi ties. Furthermore_., .although the present invention is designed to be operable at home by a non-medical user, use in a doctor's office, hospital or other medical setting is also encompassed by the present invention_*

In a preferred embodiment, the first and second signals have a degree of independence from one another as a result of being derived from sensors of different modalities, even though they may be sensing the same heart or other orga or object or person. For example, generation of the first signal may be distorted, in a different amount; than generation of the second signal as a result of. t least one of (i) the medium of transmission of the. first and second signals and (ii) background noise.

Method 300 may also have a step 320 in which one or more processors thai are configured to recei ve the first signal and the second signal continually combine^' the first signal and the second signal to- form a combined signal.

Method 300 may further comprise a step 330 in which the one or more processors apply independent component analysis (!C A) to the combined signal to separate on! independent components of the combined signal so as to generate a remaining signal containing dependent components of the .combined si gnal.

Method 300 may include a step 340 of outputting an ongoing measurement of the cardiac bioparameter based on .the remaining signal. ( putting the result may include O^'ut utting a numeric or graphic indication of the monitored or measured one or more bioparameter(s).

Any of the methods of the present invention may also include a step of repeatedly sensing a fetal bioparameter during a time interval including sensing the fetal bioparameter at a first point of the time interval when a sensed fetal organ is nearer to sensors of a first modality than to sensors of a second modality, and including sensing the fetal bioparameter^'. at a second point of the time interval when the fetal organ is nearer to the sensors of the second modality,

Structural components mentioned in connection with the system or device or method of the present invention may be incorporated into any of the methods of the present invention and steps or features of any of the methods of the present inventio may be incorporated into the system or device of the present invention.

In any version of the present invention, the holder unit may communicate, for example wirelessly, to a mobile application. As shown in FIG. 3, the mobile .application may display output of the processing relating to monitored bioparameier{s) of the fetus and/or mother o computer screen or a mobile device such as a mobile telephone.

The mobile device, and/or the holder unit (or one or more sensors of the holder unit) may be in communication,, for example wireless, with a .server in cyberspace. Data storage of output generated by the one or more processors of the present invention may utilize cloud storage or other suitable storage.

As a result of the present invention, it has now become practical to. continuously monitor pregnant women and their fetuses on a. large scale. Many more women can. pregnant women can participate in this and the number of data points over time for each woman- and fetus is greatly enlarged in some eases by one or more orders of magnitude. As a result, the continuous monitoring of the pregnant mother ami of her fetus with many parameters now justifies the .application of "big data- analysis¹' collected, stored and processed on servers in cyberspace.

Reasonable prediction of health conditions based, on the many monitored parameters is believed to be possible by comparing the data coHected from -a particular pregnant woman and her fetus against baseline reference data for pregnant woman and fetuses having particular -demographic, arid/or geographic or other statistical properties. The present invention may for example-allow one or more processors to output the information so thai a medical practitioner can reasonably conclude and advise a pregnant to make changes in her activity. Fo example, a woman whose activit level shows that -she did not move for X hours and whose heart, rate and/or heart rate variability or other cardiac activity shows certain data, may be told that statistical data shows that other pregnant women who moved that little and who had that cardiac data had a higher rate of sufferin from, a particular pregnancy stress-related condition within a week and that therefore she should take a day off from, work and/or make other changes in her activity level or stress level. This example is of course non-exclusive and non-limiting.

Single Modali ty Embodiments

Although the drawings and descriptio have described the presen invention, as utilising sensors from multiple modalities, in another preferred embodiment of the present invention, sound components only from acoustic signals are decomposed by the ICA. In this case we separate between- ^'the- cardiac activity of the mother, the fetus, gastric activit and background noise, in one version, acoustic sensors adj cent the lower abdomen of the mother sense fetal, cardiac activity and acoustic se sors adjacen the upper abdomen of the mother sense the mother's cardiac activity. Cardiac activity may inciude heart beat, specific heart sounds (i.e. SI through S4) etc. The signal to noise ratio may be improved using ICA as dependent data from more than one sensor is gathered: together, and non-dependent data: {representing other phenomena than heart: activity) are rejected to a different ICA channel.

In a still further preferred embodiment, electric components from electric signals alone such as from an ECG are decomposed by .ICA. We separate between the cardiac activity of the mother, electric activity of the fetus and background electric signals from the muscle activity of the mother.

Although heretofore the present invention has been described mainly i term of monitoring the fetus, similar monitoring; of the mother's vital signs may be performed as well

While the invention has been described with respect to a limited number of embodiments, it will be appreciated that many variations, modifications and other applications of the inventio may be made. Therefore, the claimed invention as recited in the claims that follow is not limited to the embodiments described herein.

Claims

WHAT IS CLAIMED IS:

1. A system for continuous non-invasive monitoring of one or more bioparameters.of a fetus using. multiple fixed location sensors, comprising;

a holder unit configured to bold adjacent a body of a mother one or more sensors of a first sensor modality monitoring the one or more bioparameters and one or more sensors of a second sensor modality monitoring the one or more bioparameters,

one or more sensors of the first sensor modality fixedly placed at one o more separate locations so as to repeatedly sense, at regular or irregular time intervals, the one or more bioparameters and produce at least one first signal;

one or more sensors: of the second sensor modality fixedly placed a one or more separate locations: so as to repeatedly sense, at regular or irregular time intervals, the- one or more bioparameters of the Fetus sensed by the at least one sensor of the first sensor modalit and so-. as to produce at least one second signal, that is partially independent. -of the at least one first signal; and

one or more processors, accompanied by program, instructions or software, configured to receive the first and second signals, generate a combined signai from a combination of the at least one first signal and the at least one second signal and process the combined signal to produce an output indicating the one or more bioparameters.

2, The system of claim 1 , wherein the at least one first signal has noise, Nj, caused by one or more noisy background events , wherein the. at least one second signal has a noise, V, caused by one or more noisy background events and wherein the noises ^"Nj and N₂ are partially independent

3. The system of claim 1, wherein one noisy background event from within the fetus or mother affects the at leas one first signal differently than said one noisy background event the at least one second signal,

4. The system of claim 1 , wherein the one or more sensors of the first sensor modaiity are acoustie sensors that monitor cardiac activity of the fetus and the one or more sensors of the second sensor modality are electric signal sensor that monitor the cardiac activity of the fetus.

5. The- system of clai m 4, wherein the one or more sensors of the second sensor modality are ECG sensors.

6. The system of claim 1 , wherein the one or more sensors of the first sensor modality are one of (i) acoustie, (ii) ECG, (hi) ultrasound/Doppler and (i v) functional near infra red spectroscopy sensors and wherein the one: or more sensors of the second modality are another one of (i) acoustic, (it) ECG. (iii) ulirasoimd/Doppler and (iv) functional near infra red spectroscopy sensors,

7. The system of claim 1. wherein the combined signal is generated by the one or more processors by adding a magnitude of the at least one first signal at a given point in time to a magnitude of the at least one second signal at the given point in time.

8. The system of claim 1 , wherein the combined signal does not. resemble either the at least one first signal or the at least one second signal.

9. The system of claim I , wherein the combined signal correlates with the one or more bioparameters more than either the at least one first signal -or the at least one second signal.

10. The system of claim 1 , wherein the one or more processors are configured to process the combined signs! by performing independent component analysis (ICA) on the combined signal.

1 1. ^"The system of claim 10, wherein the ICA is applied to the combined signal dynamically at discrete time intervals to optimize a decomposition of the combined signal.

12. The system of claim 1 , wherein the one or more processors output a numeric or graphic indication of the one or more bioparameters,

13. The system of claim L wherein the output is based in part on which sensor eon-elates best with the combined signal.

1.4. The system of claim 1 , wherein the system, is configured to be -ambulatory and operable by a mother at home without medical personnel,

15. The system, of claim 1, a transmitter for communicating the output of the one or more processors to a remote server via at least one of low emission Bluetooth (Bluetooth Low Energy) and automatic gain control (AGC).

16. The system of claim L wherein the one or more bioparameters comprise one or more cardiac bioparameters.

17. The system of claim 16, wherein the one, or more cardiac bioparameters comprise a heart rate variability.

1 S. The system of claim 1, wherein the one or more bioparameters comprise one or more brain bioparameters.

19. The system of claim 1 , wherein the holder unit is wearable.

20, The syst©m_: of claim 1, further comprising surround speakers situated on each sensor of the one or more sensors of the first sensor rnodaiiry,

21 , The system of claim 1, wherein the one or more processors located remotely to the holder unit

22, The system of claim 1 , further comprising one or more sensors of a third sensor modality fixedly placed at one or more separate locations so^' as to continually sense, at repeating time intervals, the one or more cardiac bioparameters of th fetus sensed by the at least one sensor of the firs sensor modality and the at least one sensor of the second sensor modality so as to produce, at least one third signal that is partially independent of the Ml east one first signal and of the at least one second signal_*

23, A method of monitoring one or more bioparameters of a fetus, comprising;

applying a wearable holder una to an abdomen of a pregnant woman so that two or more sensors of a first modality and two or more sensors of a second modality are configured to continually sense one or more bioparameters of the fetus, thereby generating: a first signal from sensors of the first modality and a second signal from sensors of the second modality, wherein the first and second signals have a degree of independence from one another as a result of being derived from sensors of different modalities;

one or more processors that are configured to receive the first signal and the second signal continually-combining the first signal and. the second signal to form a .combined signal ;

the one or more processors applying independent component analysis to the combined signal to separate out independent components of the combined signal so as to generate a remaining signal containing dependent components of the combined signal; and

o tputtmg an ongoing measurement of the one or more bioparameters based on the remaining signal,

24. The method of claim 23y further comprising generation of the first signal being distorted in a different amount than generation of the second signal.

25. The method of claim 24, wherein generatio of the first signal is distorted in a different amouni than generation of the second signal as a resul t of a t least one of (i) a medium of transmission of the first and second signals and (ii) a background noise.

26. The method of claim 22, wherei the one or more bioparameters are one or more cardiac bioparameters and wherein the two or more sensors of the first modality and the two: or more sensors of the second modality target a heart: of the fetus to generate the first signal and- the second signal.