WO2009013363A1 - System for estimating the titration of breastfeeding - Google Patents

Abstract

The invention relates to a system for estimating the titration of breastfeeding, that is the volume of milk ingested by a newborn during breastfeeding. Said system establishes a fractal model of the human breast with the aim of estimating the total volume of milk stored therein. The system according to the invention uses a system for emitting and receiving ultrasounds for capturing an image before the child starts to breastfeed, and another after. Said images are pre-processed with both DCT and DWT before the estimation. The method for estimating the titration is based on a group of neuronal networks with the aim of reducing error between the actual value and the estimated value by reducing the bias (systematic error) and the variance of the estimating system using an average of the responses of the group of neuronal networks. An algorithm is also used to reduce said systematic error on the basis of image segmentation techniques combined with said fractal system.

Description

 System for estimating the volumetry of breastfeeding.

FIELD OF THE INVENTION

The present invention develops a system, including a digital ultrasonic scanning device, for estimating the volume of milk ingested by a newborn during the breastfeeding process, being configured as a handheld ultrasonic ultrasound device. . The system provides the final measurement through the ultrasound information obtained by its operating procedures.

BACKGROUND

The benefits of breastfeeding, both for women and for newborns, are well known to the medical community and this is evidenced by numerous publications in the field of pediatrics. In fact, both the World Health Organization (WHO) and the United Nations Children's Fund (UNICEF) recommend that all mothers breastfeed their newborns exclusively during the first six months of lifetime. In the recommendations published in the ^λ Healthy People 2010 'report, the US Department of Health sets the goal of reaching a 75% prevalence of breastfeeding during postpartum along with a 50% prevalence during the first six months of life

Although the current prevalence during the post- Childbirth is relatively high (around 70% in the US, 80% in Australia, 90% in Germany, etc.), less than half of newborns receive breastfeeding at six months of age. Although it is difficult to identify the direct causes of early termination of breastfeeding, the latest published pediatric studies suggest that there are numerous difficulties arising from the process of breastfeeding during the first weeks of postpartum, which seems to be which have a great correlation with this premature discontinuity.

These problems during the first 4 weeks of life are quite common. A third of mothers report perceiving at least one or more problems related to breastfeeding, such as lack of confidence in their ability to breastfeed, perception of not having enough milk, sucking problems by the newborn, etc. In turn, other studies have been published that raise the rate of women suffering from these problems up to 38%.

The patent WO2006054287, develops a system that integrates various devices integrated in a bra, with openings that allow breast exposure to breastfeed the newborn. Inside said bra are various ultrasound emitting / receiving probes, which use the Doppler Effect to calculate the flow of milk and from it to estimate the volume of milk taken by the newborn.

However, such a system does not turn out to be a means suitable for the estimation of the volume of milk taken by the newborn, since it is a highly invasive and disturbing system of intimacy between the mother and the newborn by forcing to breastfeed with a truly uncomfortable garment to be able to estimate volumetric

The US5827191 and WO2006003655 patents develop a system for measuring the volume of milk ingested by a newborn based on an elastic nipple covering through which, in order to measure it, the flow of milk taken.

Both systems have the disadvantage of establishing a physical barrier between the mother and the baby, disturbing not only the intimacy between the two, but also the breastfeeding process itself, since it hinders both the sucking of the infant and the production of milk derived from the stimulus direct nipple by the child.

However, the question of establishing an autonomous system for the estimation of the volume of breastfeeding that is precise, comfortable, non-invasive, easy to use and that should not be in contact with the mother or the newborn during the whole period persists The breastfeeding process Specifically, there is a need to establish a system that helps solve all the problems described above.

That is why, one of the main objectives of the present invention is to solve the problem derived from the process of breastfeeding of newborns during the first six months of life providing real-time data to new mothers on the volume of milk ingested by their newborn.

Another of the objectives and advantages obtained with the ease and portability of the system of the present invention refers to the increase in the quality of life and comfort of new mothers and newborns since minimal contact is required for the realization of the estimation (requires only two pre and post suction scans), being another of the main objectives of the present invention to extend the breastfeeding process beyond the first six months of life.

BRIEF EXPLANATION OF THE INVENTION

In accordance with the background of the invention detailed above, the system for estimating volumetry in breastfeeding uses a contact ultrasonic device using the physical Back-Scatter principle with which a first scan of the breast of the breast is performed. mother providing a first set of parameters which may include, the total volume of the breast, the estimated number of lobes and lobules (functional units of the mammary gland), the number of lactophores-channels, acoustic properties of the tissue, etc.

With the data set obtained above, a fractal model is obtained through which the volume of milk stored (total capacity) in the lactophores and in the whole of the mammary gland (especially the lobules). This model also allows an estimation of the milk production rate once the volumetric reserve previously calculated with the model is consumed.

Once the previous measurements have been carried out together with the first estimate of the volumetry and once the process of breastfeeding the newborn with a breast is finished, a second scan is carried out by contact with the maternal breast with the ultrasonic sensor. With this second scan you get the same data set detailed above.

With the available data set, a function / system is inferred that, from an input of breast images, before and after breastfeeding is able to provide the total volume of milk consumed. To do this, this estimate is modeled with a classification process based on different categories (input data).

For this, a preprocessing of the images is carried out prior to the estimation phase. The system presented in this invention can use the discrete cosine transform (Discrete Cosinus Transform / DCT) or the autocorrelation matrix values of the input images. The preferred implementation of this system performs said preprocessing by means of Wavelet transforms (Discrete Wavelet Transform /

DWT), since it captures the specificities of the Input ultrasound (ie the images of the two scans performed with the ultrasonic sensor) at different levels of resolution, and the DCT (Discrete Cosinus Transform).

Once the preprocessing of the incoming ultrasound is performed, the volume of milk ingested by the newborn is estimated through the implementation of a committee of neural networks of the type ^ Radial Basis Functions', which calculate a function that returns the volume ingested from Preprocessed images. In order to reduce the variance of the previous estimate, the neural networks committee is constituted by a set of networks, which is trained from a sampling of the input image, so that by averaging the outputs of said networks the variance of the estimate is compensated and, at the same time, the input of each neural network has a lower dimension. This committee of neural networks can be combined with image processing techniques and, in particular, with the segmentation of preprocessed images to calculate the degree of collapse of the breast tree (ie of all breast tissue) and, as of this degree For compression, estimate the volume of milk ingested (this estimate is direct since the total volume of milk available from the fractal model of the breast is known).

BRIEF EXPLANATION OF THE DRAWINGS

Figure 1 shows a representation of the part bottom of the system for estimating breastfeeding where the ultrasonic emitting device is detailed.

Figure 2 shows a representation of the profile of the system for the estimation of breastfeeding where it is appreciated that it is a non-invasive handheld device.

Figure 3 shows a superior representation of the system for the estimation of breastfeeding where the screen where the data of the readings of the volumetry and the temporal evolution of the same are shown.

Figure 4 shows the high level architecture of the global system for ultrasonic capture and the implementation of estimation algorithms. System components include an ultrasonic emitter, a transmission / reception switch, analog / digital converters FPGA or DSP devices and interconnection elements between plates.

Figure 5 shows the minimum, medium and maximum volume of milk stored in a breast (BSC) obtained from the application of the fractal model of the present invention.

Figure 6 shows the bank of filters that implement the DWT (Discrete Wavelet Transform) of the present invention. The high-pass and low-pass components of the filter bank are detailed as well as the resolution levels for the calculation of the wavelet transformation. Figure 7 shows the block diagram for the estimator based committee of neural networks of the Radial Basis Functions type, which estimate the volumetry of breastfeeding from the samples of preprocessed images (preprocessed ultrasound).

Figure 6 shows the block and operation diagram for the estimation of breastfeeding volumetry combining the fractal model defined in the present invention with image segmentation techniques.

DETAILED EXPLANATION OF THE INVENTION

The present invention consists of a system for estimating the volume of milk ingested by a newborn during breastfeeding whose data is evaluated in real time by means of an ultrasound capture device (1) linked with different digital subsystems (7, 8 , 9) that implement, respectively, a fractal model (7) of the human breast, a committee of classifiers based on neural networks (8) of the RBF type and, finally, another subsystem (9) that calculates the degree of collapse of the breast tree . Although the preferred implementation of the system uses the digital fractal (7) and neural network (8) subsystems above, the breastfeeding estimator of the present invention can be integrated with the sub-system for measuring the collapse of the breast tree (9) to reduce the variance of the estimates made.

The device (1) consists of an ultrasonic transmitter and receiver (2), a transmission card responsible for generating the ultrasound pulses of the conformation of the ultrasound emission beam and the digital / analog conversion of said pulses and of the beam conformation ( ^λ beaforming ') of the emitter. In the preferred implementation of the system, said sending / receiving device (2) is implemented by a single DSP microcontroller device type FPGA.

It also consists of the device (2) of one or several receiving devices (5) (one for each channel according to the previous beam conformation) whose function is the reception of the pulses emitted by the sending card

(4), analog / digital conversion and ultrasound conformation, which will be used as input for the estimation of the volume of milk ingested. In the preferred implementation of the system, said receiving elements are implemented by a DSP device

/ FPGA for each channel. Depending on the degree of reliability desired in the final system, the number of ultrasound channels may vary between 8 and 64.

Finally, the device (1) consists of a connection panel between cards, which interconnects the sending card (4) and the receiving cards (5) (8-64 channels) with the controller card (6) which is responsible not only for control the receiving transmitter (2) but also on it the filters and detailed models are implemented then. In the preferred implementation of the system, said controller is implemented by means of an FPGA type DSP device.

For the calculation of the breast volume and interpretation of the images obtained by ultrasound the invention has the appropriate means of computing and processing information that allow it to apply different procedures according to different subsystems for processing the information received.

Subsystem (7) implements a fractal model described below. In the breast tree, the channels bifurcate dichotomously, reducing their length and diameter systematically. On the other hand, it is estimated that the breast tree ends in approximately 2 ¹⁰ lobules. Each of these lobules is divided into 4 generations of alveoli, which constitute the mammary gland. Considering the lower part of the breast tree (ie generations 5 to 10) and assuming that milk flows in the canal system following the Poseuille law (in fact it is a very good approximation for the area of the tree considered), structures can be deduced optimal by minimizing viscosity dissipation ( ^Λ viscous dissipation ') within the volume of the tree. In fact, the detailed analysis of this proposition by means of Lagrange multipliers suggests that the best tree structure is fractal with dimension 3. In this ideal tree, the successive segments will be homothetic with a constant size / radius ratio (h). In fact, this result can be considered as a particular case of the law. Hess-Murray, who models the blood tree.

Assuming that the branches that branch off are symmetrical

(This assumption turns out to be quite reasonable for the inner part of the chest), the ratio between the diameter and the length between the p-1 generation and the p generation is h _p . If we denote V as the volume of a given channel, the channel reduced by a factor h results in a volume multiplied by a factor h ³ for each generation. For after p generations, the sizes will have been reduced by a factor hixh∑x .. xh _p so that the volume of a tree of N + l generations

(indexed from 0 to N) can be written as follows:

V = V ₀ + ¿2 ^p (h _{l X} h ₂ x. Xh _p ) ³ V ₀ p = l

If we assume that the reduction factor is constant between generations, the volume can be rewritten as:

V = V ₀ (l + ¿(2h ³ ) ^p ) p = l

By rewriting the previous equation as a finite difference equation (EDF), the volume can be written as follows:

V (n + l) = V (n) + V _or r ^{n "1}

Donde r=2h³. De todas maneras, hasta este momento sólo se ha calculado el volumen de leche almacenada en el interior del árbol mamario por lo que no se ha considerado todavía la contribución al volumen total derivada de los lobulillos. Como se verá a continuación, dicho volumen es bastante mayor que el calculado anteriormente .The term Nth can be written as:

Where r = 2h ³ . However, until now only the volume of milk stored inside the breast tree has been calculated, so the contribution to the total volume derived from the lobules has not yet been considered. As will be seen below, this volume is much larger than the one calculated above.

As mentioned earlier, there is a one-to-one relationship between the number of lactophores channels in the last iteration (N) and the number of lobules. In other words, a breast tree with 2 ^N branches corresponds to 2 ^N lobules. Therefore, assuming a spherical shape for the lobules with an average radius h, the total chest capacity (BSC / ^λ Breast Storage Capacity ') can be written as follows:

V = V ₀ ( ^{1 ~ rN1} ) + -2 ^{N + 1} πh ³ 1-r 3

The value of h can be calculated from the ultrasound taken in the subsystem (3) since this value is the relationship between the length of the lactophore channel and its radius. In conclusion, the subsystem (4) is responsible for estimating this factor h and calculating the BSC, using the above equation from h and the number of terminal lactophores.

On the other hand, the fractal model presented above not only defines the BSC but also provides the theoretical basis for the subsystem (9) since during a milk intake, the amount of Breast tissue should be kept constant. In other words, this implies that any change within the chest during a milk intake will be due to a pressure drop (ie the structure of the tree of channels and lobules collapses during the 'Emptying of the chest). Therefore, the volume of milk consumed by the newborn can be estimated from the BSC and the degree of breast tissue collapse. Figure 5 shows the minimum, average and maximum BSC for a breast as a function of h.

After the first ultrasound scan

(ultrasound) from which the data detailed above (Basically BSC, h and number of lactophores channels) have been obtained, a second scan (ultrasound) is performed after the taking for automatic real-time estimation of the milk taken by the newborn

On the other hand, the subsystem inputs (8) consist of two images obtained from the ultrasound of the receiving transmitter (2) made before and after breastfeeding. The present subsystem consists of two distinct phases. One defines the preprocessing unit, which is designed with the purpose of obtaining invariant characteristics / data regarding translations, rotations and / or scaling (both in frequency and time) from the images / ultrasound. The second phase is based on ^ Machine Learning 'techniques and computes the total volume from the data obtained from the preprocessed images / ultrasound. The preprocessing phase of the images (10) aims to transform the input image into a vector, which captures the characteristics associated with the problem of estimating the volume of breastfeeding. This vector is invariant in relation to the rotation of the input images, changes in the angle of measurement (or taking of the ultrasound), common characteristics related to the production of milk that are independent of the person, etc.

In the preferred implementation of said image preprocessing phase (10) two different techniques are used that can be used independently or in conjunction with the aim of obtaining the data vector described above. The first technique uses the Discrete Cosine Transform (Discrete Cosinus Transform or DCT) and the second uses the Two-Dimensional Wavelet Transform (Discrete Wavelet Transform or DWT).

In the present invention the DCT transform has been selected because it computes a representation of the image as a combination of horizontal and vertical frequencies. Therefore, with this representation, the details of the image that have certain frequency characteristics will appear along a circle depending on the angle that said characteristic has with respect to the horizontal and vertical axes. Therefore, the representation of the input image in the transformed DCT domain will be invariant with respect to the rotations that can be made on the original image. Moreover, just a part of the image in the transformed domain will be relevant, so that the region of interest of the image in the transformed domain may be selected in order to reduce redundancy while, at the same time, all the information contained in the image is preserved original.

The second approach to preprocessing (10) of the input image is based on the DWT. The application of this preprocessing of the input images together with the DCT or independently since scattering properties ^Λ 'of the input ultrasound may be characterized by their spatiotemporal univocally properties is considered. Normally, these properties have been studied by the Discrete Two-Dimensional Fourier Transform (2D-DFT) or by the STFT (Short-Term Fourier Transform). It is a well-known result that both 2D-DFT and STFT cannot simultaneously represent the temporal and frequency properties at the same time, and with different degrees of resolution, of an input image.

The DWT presents the ideal framework for the problem of estimating the volume of breastfeeding since it provides a representation of the images in which the relevant information (presence of tubes, layers of tissue, etc.) of information related to the smooth change of the gray scales, artifacts due to the diffraction of the waves, the effect of scaling and rotation of the image because each measurement is performed under slightly different conditions, etc .

The DWT coefficients describe the correlation between the selected wavelet (in the preferred implementation of the present invention Haar and Daubechines wavelets have been selected) and the image at various scales / resolutions (ie the degree of similarity between the image and the wavelet for a time-discrete combination and position). In other words, the calculated coefficients provide the amplitudes of the wavelet series over a set of scales and translations, which must be added in order to obtain the original image. From this perspective, the DWT analysis can be understood as a search on the image of interest of characteristics that resemble the selected wavelet. This search is done on several scales and several wavelet sizes. For this reason, the Haar Wavelets have been selected to identify the lactophores channels while the Daubechines Wavelets are used both for the detection of lobules and for the elimination of artefacts derived from the absorption and scattering of the initial images.

The implementation of this analysis in the preprocessing (10) is carried out by means of a filter bank (figure 6). With this set of filters, the input image is divided into various spectral components ('sub-band coding') in order to obtain the high-pass, low-pass and pass-band components of the input signal. These components are obtained by applying filters h (high-pass) and g (step- low) . The application of filter h is analogous to the application of the selected wavelet to the input image while the application of filter g is analogous to the application of a scaling or smoothing function to the input signal.

The subsystem (8), apart from the preprocessing phase

(10) described above, also includes a module

(11) whose objective is to infer a general function from a set of available examples

(ultrasounds stored in a Database). From this general function and the two preprocessed images, the volume of ingested milk can be estimated.

Since the estimation of the dependence between the volume and the two ultrasounds available before and after breastfeeding is quite complicated, that the dimensionality of the entries is really high and the number of available examples is low, the use of a Neural network of the Radial Basis Function (RBF) type to solve the problem described above. The RBF neural network is based on the computation of centroids

(The most intuitive function that can be computed is an interpolation between values using a ^λ lookup table '). Said computation of centroids and interpolation assumes that the underlying function to be estimated is smooth (low variability) so that the effect of noise on the input data (ultrasound) translates into an uncertainty in the estimated volume, which will be around the centroid , but said effect will be limited since said volume of uncertainty is not too high in the system presented.

On the other hand, the criterion for evaluating the quality of the estimated function will be the Mean Square Error (MSE) applied to a control or validation sequence on a set of elements of the original Ultrasound Database. The performance of the RBF learning algorithms, which infer the function of volume from the available examples and inputs, can be improved by various techniques. Such techniques include the combination of classifiers and / or training of the classifiers with non-uniform distributions on the data, the coding of the output of the classifiers or the exploitation of the properties of the Kernels due to the high dimensionality of the data. These methods for improving estimator performance are analyzed by the expectation of the input / output data and by the relationship between the dimensionality of the input and the number of samples.

The error on the measurement of the module (11) can be decomposed as the sum of the squares of the bias of the estimate and its variance. By aggregating neural networks, the estimation error can be controlled by reducing both bias and variance. Therefore, if the set of neural networks is trained so that the output of each network is a random variable, the effect of computing the average of all outputs will result in a reduction of the variance and, therefore, of the error of estimation. This reduction will be a function of the number of Neural networks used in the system. In the preferred implementation of the system, depending on the degree of reliability required, between 1 and 100 neural networks can be used.

The neural networks of the present invention are organized in a committee-like structure so that, given a common input, each network uses a sampled version of the common input and calculates the output of each of them in parallel. The effect of sampling the common input is that a different systematic error is introduced into each network. Therefore, when the output of a group of neural networks is added, since each network has a different bias, the estimation error decreases even more because the systematic error is now taken as a mean and the reduction of the variance described previously.

In the preferred implementation of the present invention, it is proposed to train a committee of RBF neural networks with sampled input images (or segmented images) so that each neural network has a different image as input so that when all the outputs of the Neural networks by calculating the average of all outputs of the neural networks, the systematic error will be less than that obtained with RBF trained with the entire image and not with sampled versions (or image segments) of it. Figure 7 shows the block diagram of this system.

Finally, in order to further reduce the error systematic of the estimation of the volumetry, the system (8) can be combined with the system (9) (figure 8) where the input images are segmented by image processing techniques, which include opening, smoothing, ^and overlay ', calculation of the perimeter / segmentation and estimation of the area of the previous perimeter. The ratio of areas calculated by this subsystem applied to the fractal model (BSC) described above provides a second estimate of the completely incorrect volume with which it has been calculated with the module (11) figure 7 so, if the average between the two is taken estimates, the systematic error decreases.

It is understood that in the present case there may be variable as many alterations of finish or shape do not modify the essence of the invention.

Claims

1.- SYSTEM FOR ESTIMATION OF BREASTFEEDING VOLUMETRY CHARACTERIZED by - incorporating a non-invasive hand-held device (1) for said estimation of the volume of milk ingested by a neonate, using an ultrasound receiver emitter (2) integrated into said handheld device (1) for capturing an ultrasound before breast feeding and another after breast feeding - have said device (1) with computing, data storage and processing means, suitable for the processing of the images obtained - have operating procedures intended to estimate the volume of milk ingested by the neonate in real time in measurements of cubic centimeters, milliliters, ounces, etc. - present the estimated volumetric reading on a screen (3) of said device (1). 2.- SYSTEM FOR ESTIMATION OF BREASTFEEDING VOLUMETRY, according to the previous claim, CHARACTERIZED because said operating procedures can simultaneously establish different subsystems that make independent estimates of the total volume of milk stored in the breast (BSC / Breast Storage Capacity). , said combination of subsystems intended to reduce the systematic error in the measurements of said total volume of milk consumed and increase its degree of precision. 3.- SYSTEM FOR ESTIMATION OF BREASTFEEDING VOLUMETRY, according to the previous claims, CHARACTERIZED because said device (1) uses a contact ultrasound sensor to capture ultrasounds before and after breastfeeding. 4.- SYSTEM FOR ESTIMATION OF BREASTFEEDING VOLUMETRY, according to the previous claims, CHARACTERIZED in that said ultrasound capture is carried out by means of a transmitting card (4) and an ultrasound receiving card (5). 5.- SYSTEM FOR ESTIMATION OF BREASTFEEDING VOLUMETRY, according to the previous claims, CHARACTERIZED because the system for capturing ultrasound is implemented in said handheld device (1) using DSP devices, and/or microcontrollers, for example made using FPGA chips. 6.- SYSTEM FOR ESTIMATION OF BREASTFEEDING VOLUMETRY, according to the previous claims, CHARACTERIZED because the system for capturing ultrasound (2) can be implemented with approximately 1 to 10 FPGA devices depending on the number of reception channels that desired (for example, in a range of 8 to 64) 1.- SYSTEM FOR ESTIMATION OF BREASTFEEDING VOLUMETRY, according to the previous claims, CHARACTERIZED because the system for the Ultrasound capture (2) has a coordinating card (6) for the formation of the images (ultrasounds) captured by the ultrasound emission (4) and reception (5) cards. 8.- SYSTEM FOR ESTIMATION OF BREASTFEEDING VOLUMETRY, according to the previous claims, CHARACTERIZED because the issuing (4) and receiving (5) cards are connected to the coordinating card (6) through a serial connection panel . 9.- SYSTEM FOR ESTIMATION OF BREASTFEEDING VOLUMETRY, according to the previous claims, CHARACTERIZED because said subsystems for estimating volumetry are stored in said coordinating card (6). 10.- SYSTEM FOR ESTIMATION OF BREASTFEEDING VOLUMETRY, according to the previous claims, CHARACTERIZED because the coordinator card is implemented with a DSP device, and/or microcontrollers, in turn made for example using FPGA chips. 11.- SYSTEM FOR ESTIMATION OF BREASTFEEDING VOLUMETRY, according to the previous claims, CHARACTERIZED because the images obtained are preprocessed before the estimation using the DCT transform (Discrete Cosinus Transform). 12.- SYSTEM FOR ESTIMATING THE VOLUMETRY OF THE BREASTFEEDING, according to the previous claims, CHARACTERIZED because the images obtained can also be preprocessed before estimation using the DWT transform (Discrete Wavelet Transform). 13.- SYSTEM FOR ESTIMATION OF BREASTFEEDING VOLUMETRY, according to the previous claims, CHARACTERIZED because the wavelets selected for preprocessing are the Haar and Daubechines wavelets. 14.- SYSTEM FOR ESTIMATING THE VOLUMETRY OF BREASTFEEDING, according to the previous claims, CHARACTERIZED because the procedure for estimating the volume of milk ingested is carried out from two preprocessed images (before and after breastfeeding) . 15.- SYSTEM FOR ESTIMATING THE VOLUMETRY OF BREASTFEEDING, according to the previous claims, CHARACTERIZED because said estimation of the volume of milk ingested is carried out by a subsystem with a fractal model (7) of the human breast. 16.- SYSTEM FOR ESTIMATING THE VOLUMETRY OF BREASTFEEDING, according to the previous claims, CHARACTERIZED because said estimation of the volume of milk ingested is carried out using neural networks. 17.- SYSTEM FOR ESTIMATION OF BREASTFEEDING VOLUMETRY, according to the previous claim, CHARACTERIZED because the previous neural networks are configured approximately as a committee of 1 to 100 classifiers. 18.- SYSTEM FOR ESTIMATION OF BREASTFEEDING VOLUMETRY, according to claims 16 and 17, CHARACTERIZED because to reduce the systematic error of the system, each element of the classifier committee has as input a sampled version of the preprocessed images. 19.- SYSTEM FOR ESTIMATION OF BREASTFEEDING VOLUMETRY, according to the previous claims, CHARACTERIZED because to reduce the systematic error of the system, image segmentation techniques can be used to estimate the degree of collapse of the mammary tree. 20.- SYSTEM FOR ESTIMATION OF BREASTFEEDING VOLUMETRY, according to the previous claim, CHARACTERIZED because the volume of milk ingested can be estimated by means of a subsystem (9) from the BSC and the degree of compression of the mammary tree. 21.- SYSTEM FOR ESTIMATION OF BREASTFEEDING VOLUMETRY, according to the previous claims, CHARACTERIZED because said device (1) works with batteries (12). 22.- SYSTEM FOR ESTIMATING THE BREASTFEEDING VOLUMETRY, according to the previous claims, CHARACTERIZED because the system generates a alarms in the event that an error occurs (batteries (12), ultrasound capture, etc.). 23.- SYSTEM FOR ESTIMATION OF BREASTFEEDING VOLUMETRY, according to the previous claims, CHARACTERIZED because the system can have the appropriate memory records and be configured to store data from various readings and give an evolution of the feedings over time. over time. 24.- SYSTEM FOR ESTIMATION OF BREASTFEEDING VOLUMETRY, according to the previous claims, CHARACTERIZED because the system can have the appropriate communication means and be configured to upload the readings to a computer for analysis.