WO2026068872A1 - Smart textile garment, system and method for functional assessment and monitoring of the lumbar spine using the smart textile garment - Google Patents

Abstract

The present invention relates to a smart textile garment, a system and a method for functional assessment and monitoring of the lumbar spine, based on the acquisition of bioelectrical signals and data concerning a user's postures and movements, which, in combination with artificial intelligence techniques, allows the sensations perceived by the user to be objectively assessed and classified, eliminating the subjective component in this operation. For this purpose, the garment of the invention incorporates a plurality of textile electrodes for acquiring various bioelectrical signals, such as ECG (HRV), EMG and/or EDA, which are processed by control electronics linked to a movement and posture acquisition system and to processing software that uses machine learning methods to associate patterns of movement and posture with bioelectrical signals indicative of injuries causing lumbar pain.

Description

SMART TEXTILE GARMENT, SYSTEM AND PROCEDURE FOR FUNCTIONAL ASSESSMENT AND MONITORING OF THE LUMBAR SPINE USING SAID SMART TEXTILE GARMENT

TECHNICAL SECTOR

The present invention relates to a smart garment or textile accessory capable of monitoring movements, postures, and physiological signals for application in the prevention, detection, and treatment of injuries causing lumbar spine pain. The invention also relates to the procedure for functional assessment and monitoring of the lumbar spine using a system that includes the smart garment, which enables the objective detection of injuries causing lower back pain.

To this end, the garment of the invention incorporates a plurality of textile electrodes for the acquisition of different bioelectrical signals, such as ECG (HRV), EMG and/or EDA, which are acquired by the textile electrodes and processed by a control electronics linked to a movement and posture acquisition system, as well as to a processing software that employs machine learning methods to associate movement and posture patterns with bioelectrical signals, indicative of injuries causing lower back pain.

The object of the invention is therefore to provide a smart textile garment and a functional assessment and monitoring procedure of the lumbar spine that offers the necessary data to provide an objective perspective to the diagnosis and monitoring of lumbar injuries, by quantifying or grading the pain experienced by the user, eliminating the subjective component associated with the perception of pain or discomfort by the user himself.

BACKGROUND OF THE INVENTION

The detection, diagnosis, and monitoring of injuries, such as lumbar injuries, is sometimes complex, since, on the one hand, it must be carried out based on specific measurements outside of the user's daily activity, and on the other hand, it has an associated subjective component. related to the user's perception of pain or sensations.

On the other hand, biometric or bioelectrical signals commonly used for monitoring or detecting injuries—such as ECG signals (electrocardiogram, which allows for the calculation of the HRV (Heart Rate Variability) parameter), EMG (electromyography), or EDA (Electrodermal Activity)—exhibit high variability. Therefore, the information they provide must be carefully analyzed and interpreted to be of practical use. This variability is linked, among other factors, to the quality of the signal captured by the electrodes responsible for acquiring the bioelectrical signals.

On the other hand, without a correlation between the aforementioned bioelectrical signals and the movement, position or posture of the user in which the signals are captured, it is not possible to achieve an objective assessment of the condition of the lumbar spine.

In this regard, US patent document US10398339B2 is known, which discloses a device for evaluating low back pain comprising a plurality of EMG sensors based on skin-adhered electrodes, in combination with IMU sensors. The document cites machine learning techniques for evaluating and preventing low back pain. However, the device in US10398339B2 does not provide an accurate and objective assessment, as it relies on a single biometric signal, EMG, and uses skin-adhered electrodes. This not only reduces the device's operability by preventing continuous use over time but also diminishes the accuracy of the acquired signals, since the electrodes are not integrated into the device itself.

Also known is patent document US2023363711A1, which describes a garment-like device for monitoring and/or treating patients with general injuries. It specifies the use of ultrasonic energy for communication between the different modules (monitoring and/or treatment, monitoring, and regulation), as well as the possibility of incorporating sensors for motion (accelerometers), EMG, heart rate, oxygen saturation, temperature, and other signal combinations. The document discloses the use of machine learning algorithms for signal monitoring. However, the device disclosed in US2023363711A1 is not capable of effectively measuring lumbar curvature or the EDA signal. combine it with the other signals used. The Machine Learning algorithms mentioned are used only for monitoring the signals, so the device does not allow for the objective identification of pain and its intensity using the Artificial Intelligence model.

Thus, the applicant of the present invention is not aware of any device or system that allows for the precise measurement of all the necessary biomedical and biomechanical parameters (ECG (HRV), EMG, EDA) through the integration of sensor electrodes into the textile itself, nor does it allow for combining these measurements with the monitoring of postures and movements using at least two IMU sensors, which, together with the use of software and artificial intelligence, provides a functional and objective assessment of injuries or the condition of the lumbar spine.

For all the above reasons, the applicant of this patent identifies the need to develop a smart textile garment and a procedure for detecting and monitoring injuries that allows for an objective assessment of the patient's pain, based on the processing of all the parameters acquired by the sensors integrated into the garment, since the system has been previously trained with AI learning techniques, in order to prevent, detect and be able to carry out the monitoring of the treatment of injuries, especially lumbar or back injuries.

DESCRIPTION OF THE INVENTION

The smart textile garment, system, and procedure of the present invention provide a tool capable of functional assessment and continuous monitoring of the lumbar spine in users with injuries or diseases causing back pain. It is comfortable and easy to use, while providing objective and reliable information thanks to textile electrodes integrated into the garment itself, which capture various bioelectrical signals or medical data from the user. The combination of this data with posture and movement data, which the garment of the invention also collects, allows for an objective assessment and the monitoring of subsequent treatment, enabling adjustments to said treatment based on the recorded results, if necessary. The invention involves the use of Artificial Intelligence and Big Data tools, which allows for a deeper study and understanding of the mechanisms and causes involved in lower back pain, in order to establish strategies for its prevention and possible treatments through the analysis of this information.

Thus, the garment of the present invention allows the collection and monitoring of the following physiological signals:

Electrodermal activity (EDA): Electrodermal activity (EDA) is one of the ideal indicators for understanding the sympathetic nervous system. EDA can be measured using skin conductance data, as it is directly proportional to sweat secretion. This makes skin conductance an ideal measure for assessing sympathetic nervous system activation. The EDA signal exhibits sharp changes or spikes associated with a reaction to a stimulus (pain). These spikes are commonly known as the skin conductance response (SCR). EDA and SCR are measured in the same units, typically microsiemens (pS). The EDA signal spectrum ranges from 0.045 to 0.15 Hz, although it can increase to 0.37 Hz during intense exercise. The parameters that define the SCR are:

- Latency, that is, the SCR peak appears between 1 and 5 seconds after the stimulus that causes it.

- Amplitude; for a variation in the EDA signal to be considered SCR, the amplitude must be at least 0.05pS or 0.04pS.

- Recovery time.

- Surface Electromyography (sEMG): Electromyography (EMG) is a diagnostic procedure used to assess the health of muscles and the nerve cells that control them (motor neurons). Surface electromyography allows for the measurement of muscle tension without invasive monitoring methods. Electromyography results can reveal nerve dysfunction, muscle dysfunction, or problems with the transmission of nerve signals to muscles. This technology aims to detect neuromuscular disorders by assessing the health and performance of the muscles. sEMG allows for the analysis of various physiological changes. caused in different muscles, which makes it easier to diagnose the possible injury and monitor it.

Electrocardiogram (ECG): Heart rate variability (HRV) can be calculated from the ECG signal. HRV is defined as the variation in the time between RR intervals on the electrocardiogram and reflects the activity of the autonomic nervous system on cardiac function. Similar to what occurs with EDA, heart rate is an indicator of autonomic nervous system (ANS) activation, which can reflect changes in mood or even pain. Based on analyzed scientific articles (M. Riquelme et al. 2020, S. Gomez et al. 2022, M. García et al. 2021), it has been shown that HRV can be a clear indicator of chronic pain. Furthermore, when acute pain occurs, the activity of the autonomic parasympathetic nervous system also increases; therefore, a decrease in HRV can reflect this type of pain while the heart rate increases.

According to a first aspect of the present invention, the smart textile garment of the invention has at least the following elements:

- a flexible textile body with textile electrodes for capturing bioelectrical signals that are embroidered onto the textile body itself. The textile body can be of any nature and construction, including both woven and elastic fabrics. Flexible bodies of other natures suitable for garment making, such as cellulosic or polymeric substrates, or a combination thereof, should be considered equivalent solutions;

- an electronic system for acquiring data on a user's movements and postures, such that said electronic system comprises at least two inertial measurement units (IMUs), each equipped with a 3-axis accelerometer and a 3-axis gyroscope. Advantageously, the use of at least two of these IMUs allows for the detection of spinal curvature, which improves the system's accuracy; and control electronics capable of acquiring and processing the signals. bioelectrical signals captured by textile electrodes and inertial sensors. Preferably, the control electronics include an analog-to-digital converter integrated circuit with at least 8 input channels;

- a battery to power the electronic components of the assembly;

- means of transmitting bioelectrical signals and processed movement and posture data, which allow data communication between the smart textile garment and a computer or server, for subsequent management and visualization of the same;

Optionally, the garment features at least one PPG sensor (short for Photoplethysmography). The PPG sensor would allow for dual validation of heart rate measurement or blood pressure monitoring. The textile electrodes have an advantageous configuration that results in higher quality signals, minimizing noise and inaccuracies. To achieve this, the textile electrodes, embroidered onto the garment, have a three-dimensional configuration, including a primary conductive thread that serves as the connection between the electrode and the control electronics via wiring. The control electronics process the voltage signals captured by the electrodes and transform them into the bioelectrical signals of ECG, EMG, and/or EDA. Each textile electrode has at least two stitches of a second conductive thread, with a three-dimensional foam piece, between 3 mm and 6 mm thick, sandwiched between these stitches. Preferably this foam will be made of ethylene vinyl acetate (EVA).

Among other non-limiting embodiments of the present invention, the first connecting conductor is a polyamide thread with a silver coating of greater thickness than the second conductor thread, which integrates the embroidery passes that form the textile electrode. Preferably, both conductor threads are continuous silver-coated polyamide 6.6 threads, such that the first conductor thread has 8 plies (210 denier), while the second conductor thread has 3 plies (100 denier), where a ply in both cases contains 34 filaments.

The configuration described for the textile electrodes gives them a certain volume, on the order of 5 to 8 mm in thickness, which further ensures a proper embroidery process. This three-dimensional configuration is achieved with the foam piece. The interleaved design is particularly advantageous compared to the flat or two-dimensional electrodes commonly used for this type of application, as it ensures close and continuous contact with the user's skin, resulting in optimal signal quality acquired by the electrode.

As detailed above, the textile electrode has at least two passes or layers of embroidered conductive thread with different orientations, ensuring a homogeneous distribution of conductivity. Thus, in a preferred embodiment, the electrodes have six passes or layers: one embroidered layer oriented at +45° with respect to the horizontal axis of the textile body, followed by a layer embroidered at -45° and a layer embroidered horizontally. The three-dimensional foam is then placed over these three passes of embroidered conductive thread, and on top of the foam, three additional passes of embroidered conductive thread are placed following the same sequence of embroidery orientations described above.

In order to improve the conductivity of the electrode, in an alternative embodiment it has 8 embroidery passes: one embroidery pass in the +45° direction, one pass at -45°, one pass in the vertical direction and one pass in the horizontal direction, then the three-dimensional foam, and then another four embroidery passes following the order discussed for the first four passes.

Preferably, the textile electrode has a maximum of 10 layers, as a higher number could cause excessive friction of the embroidery needle with the other embroidered layers, which could lead to breakage of the conductive thread.

On the other hand, the smart textile garment of the invention can be configured as a vest, short-sleeved shirt, or long-sleeved garment. When the textile garment is configured as a long-sleeved garment, such as a long-sleeved shirt or a jacket, at least one of its sleeves has an extension designed to partially cover the palm of the hand. At least one EDA measuring electrode is located in this extension. Advantageously, the placement of the EDA measuring electrodes in the palm area of the user's hand provides better conductivity compared to other areas of the body, since the conductivity of the skin is combined with the conductivity of the hand's perspiration, which has a large number of sweat glands. Regarding the other bioelectrical signals to be acquired by the system, the electrodes responsible for acquiring the ECG signal will preferably be placed in conjunction with the user's chest, while the electrodes responsible for acquiring the EMG signal will be placed in conjunction with the back muscles.

A second aspect of the present invention relates to the functional assessment and monitoring system for the lumbar spine. This system comprises the described smart garment and an electronic device, such as a computer, smartphone, or tablet, equipped with communication capabilities for the smart garment, data reception and storage capabilities, a processor, and software that receives bioelectrical signal data captured by the electrodes of the smart garment. The system employs Big Data and artificial intelligence techniques thanks to the novel software included therein. This software allows for the labeling of bioelectrical signal data and data on movements and postures from the inertial electronic system. This labeling is used to train an artificial intelligence model with a multi-layered machine learning architecture. The software then compares the new data acquired by the smart garment with the trained model to perform the functional assessment and monitoring of a user's lumbar spine. In this way, the combination of different physiological signals, along with the monitoring of the user's movement and posture, and the use of AI, enables the system to objectively perform functional assessment and monitoring of the lumbar spine, allowing for the prevention, detection, and treatment of injuries causing lower back pain, by objectively classifying the patient's sensations based on a scale.

A third aspect of the present invention relates to the procedure for functional assessment and monitoring of the lumbar spine using the described system. For this purpose, the user places the garment snugly on their torso, the system's electronics are switched on, and it is verified that the ECG cardiac signal is being captured correctly as a method of system validation. The user then performs a series of guided physical activities, preferably movements that induce flexion, torsion, and extension of the lumbar region, allowing the measurement of the patient's exertion and sensations in a wide variety of situations.

Thus, when the patient feels pain, they will indicate it to the professional accompanying them during the test so that the existence of the pain, as well as its severity, can be indicated via an application installed on the electronic device equipped with the software. For example, on a scale of 'severe pain', 'mild pain', no pain reported. Alternatively, the user can indicate the perceived pain in the application.

The procedure for functional assessment and monitoring of the lumbar spine comprises the following stages:

- Capture of signals by the textile electrodes and of data relating to movements and postures by the electronic system and input of user pain perception data into the software;

- conditioning and processing of the signals by the control electronics of the smart textile garment for their transformation into bioelectrical signals of ECG, EMG and/or EDA;

- transmission of data packets of bioelectrical signals and data relating to movements and postures to the electronic device, such as a smartphone;

- Labeling of data packets by the software installed on the electronic device, taking into account the user's perceptions, according to the range of pain present at that moment, which is indicated in the software (computer application) installed on the electronic device. This labeling is what will allow supervised machine learning by the system;

- Data is stored in a server database and trained by the software using a predefined multi-layered machine learning artificial intelligence model. The model's weights and parameters are then adjusted to identify and classify the new data sets. The software indicates the accuracy level of the trained model.

- downloading the trained models by the software and receiving new sets of data (bioelectric signals and posture and movement data) from the user in real time on the electronic device;

- Preprocessing of the new data by the software, performing inference from the trained model to obtain the approximation for each of the labels with which the model was trained. Preferably, this preprocessing consists of averaging and deletion of outliers; and communication to the user or professional of the results obtained with the artificial intelligence model on the functional state of the lumbar spine through an application installed on the electronic device.

The artificial intelligence model comprises at least one of the following: networks neural networks, Bayesian networks, decision trees, support vector machines, models based on genetic algorithms, regression.

Therefore, according to the procedure described, the bioelectric signals are labeled, firstly, based on the user's perceptions at the time of their acquisition in a specific position or movement, and subsequently, with the trained model it is possible to offer a prediction based on the new signals captured by the garment, allowing functional assessment of the user's lumbar spine and knowing whether or not they suffer pain, as well as the degree of pain, thus eliminating the subjective factor thanks to the use of artificial intelligence.

The artificial intelligence model comprises at least one of the following: neural networks, Bayesian networks, decision tree, support vector machines, models based on genetic algorithms, regression.

BRIEF DESCRIPTION OF THE DRAWINGS

To complement the description that follows and to aid in a better understanding of the characteristics of the invention, according to a preferred embodiment thereof, a set of drawings is included as an integral part of said description, in which, for illustrative and non-limiting purposes, the following has been represented:

Figure 1 shows an example of the embodiment of the textile garment of the invention, in which it is configured as a long-sleeved t-shirt with a central zipper closure and an extension in the form of a fingerless glove that covers the palm of the hand, as well as the pattern of the same, where the arrangement of the electrodes that collect the different bioelectric signals can be observed.

Figure 2 shows a detail of the sleeve extension of the garment in Figure 2, with the electrodes that measure the EDA signal embroidered on it, as well as the placement of said extension on the user's hand, the electrodes being covered by an additional portion of fabric and the cables that connect the textile electrodes to be visible. the control electronics.

Figures 3 to 6 show the embroidery procedure for the textile electrodes according to a preferred embodiment of the invention in which the textile electrodes have 6 embroidery passes with the second conductive thread; where in Figure 3 the initial embroidery step of the first conductive thread connecting to the wiring is shown, in Figure 4 the embroidery process of the first three passes at +45°, -45° and horizontal is shown, in Figure 5 the placement of the three-dimensional foam is shown, while in Figure 6 the embroidery of the second set of embroidery passes on the three-dimensional foam is shown, as well as an image of the final result of the embroidered textile electrode.

Figure 7 shows an example of an ECG signal measured with a two-dimensional flat embroidered textile electrode, where the time in seconds has been represented on the x-axis, so that five minimum division units of the x-axis are equivalent to 0.45 seconds, while the voltage in mV has been represented on the y-axis.

Figure 8 shows an example of an ECG signal measured with a textile electrode with volume as contemplated in the smart garment according to the present invention, where the time in seconds has been represented on the abscissa axis, such that five minimum division units of the abscissa axis are equivalent to 0.45 seconds, while the voltage in mV has been represented on the ordinate axis.

PREFERRED EMBODIMENT OF THE INVENTION

The smart textile garment of the present invention that participates in the system and procedure for functional assessment and monitoring of the lumbar spine claimed is configured from a textile body (1) on which the electrodes responsible for collecting the different bioelectrical signals are embroidered according to the object of the invention.

Figure 1 illustrates an example of the textile garment configured as a long-sleeved T-shirt with a central zipper closure, where the sleeves have an extension (5) in the form of a fingerless glove that covers the palm of the hand, so that the EDA signal acquisition electrodes (7), two in the illustrated example, are positioned in contact with the user's palm, and where the acquisition electrodes of The ECG signal electrodes (8) are located on the chest, while the EMG signal acquisition electrodes (6) are placed on the lumbar and abdominal areas. Figure 2 shows a detail of the sleeve extension (5) with the EDA signal acquisition electrodes (7) embroidered on the fabric, as well as an image of the extension (5) placed on the user's hand.

As detailed above, the textile electrodes embroidered onto the garment (1) allow for the acquisition of biomedical signals such as EMG, ECG, and/or EDA. The garment also incorporates an electronic system for acquiring movement and posture data. This information is processed in conjunction with the bioelectrical signal data to provide a functional assessment of the lumbar spine. This electronic system is also connected to the control electronics and includes three inertial measurement units (IMUs), each equipped with a 3-axis accelerometer and a 3-axis gyroscope.

According to a preferred embodiment, the control electronics are based on an integrated circuit from the family of multidirectional, simultaneous-sampling delta-sigma (AZ) analog-to-digital converters (ADCs) with 24-bit resolution, amplifiers with programmable amplification factor, an internal reference, and a built-in oscillator. This converter has 8 input channels, providing different possibilities for simultaneous signal acquisition as needed, as shown in Table 1, which includes a variant embodiment of the invention that incorporates a PPG sensor.

Table 1. Examples of signal acquisition with an 8-channel input integrated circuit.

Thus, Table 1 reveals different variants for the acquisition of bioelectric signals: - Acquisition of EMG signals from 4 muscle groups and, using a PPG sensor, simultaneous acquisition of HRV.

- Acquisition of 3 EMG signals from 3 muscle groups and the ECG signal and, using a PPG sensor, simultaneous acquisition of HRV.

- Acquisition of EMG signals from 3 muscle groups and the EDA signal and, using a PPG sensor, simultaneous acquisition of HRV.

- Acquisition of EMG signals from 2 muscle groups, the ECG signal, the EDA signal and, using a PPG sensor, simultaneous acquisition of HRV.

As detailed in the embodiment illustrated in Figure 6, the textile electrodes (12) have a three-dimensional structure with a certain volume, and include a first conductive connecting thread (2), and 6 embroidery passes with a second conductive thread (3), with a three-dimensional EVA foam piece (4) interposed between them, which has a thickness between 3 mm and 6 mm.

Thus, as shown in Figure 3, the first conductive connecting thread (2) is first embroidered onto the textile body (1) using an embroidery head (9), which determines the connection point (10) of the electrode to the system's control electronics. Next, the passes that make up the textile electrode are embroidered onto the selected area (11) on the textile body (1) with different orientations to achieve homogeneous conductivity. Figure 4 shows the embroidery process for the first three passes in sequential order: first, an embroidery pass with the second conductive thread (3) oriented at +45° to the horizontal axis, a second pass with an orientation of -45° to the horizontal axis, and a third pass with the thread (3) horizontally oriented.

The foam piece (4), which can be seen in figure 5, is placed over these first three passes, and then three more layers of embroidery are embroidered on it, creating a textile electrode with volume (12), as seen in figure 6.

Advantageously, the configuration of the textile electrodes of the garment of the invention offers better signal quality, as evidenced in the experiments shown in Figures 7 and 8. Specifically, Figure 7 shows the ECG signal acquired by a Flat dry electrode (without three-dimensional foam piece (4). This type of electrode provides poor contact with the user's skin, which results in ripple noise due to permanent electromagnetic interference, an effect that is amplified by a tendency to move caused by the user's own breathing, as seen in figure 7.

On the other hand, Figure 8 shows the ECG signal acquired by an electrode that includes the three-dimensional foam piece (4) according to the present invention. By including a foam (4) in the embroidered electrodes (12), the pressure exerted on its central point is much greater, preventing the electrode from slipping on the skin and increasing electrode-skin conductivity. Thus, as can be seen in Figure 8, the electrodes of the garment of the invention reduce or even eliminate external electromagnetic interference and increase the signal amplitude, making it easier to identify the characteristic waves of different bioelectrical signals. Additionally, a clear reduction in signal ripple is observed, reaching a point where it does not interfere with the identification of small-amplitude waves such as the P wave of the electrocardiogram.

Regarding the peak-to-peak voltage of the signal, at the electrode that includes the foam piece (4), it has increased from 0.5 mV to 3 mV; that is, the signal amplitude has multiplied by 6, which demonstrates the quality of the signals acquired with the garment of the present invention.

Claims

R E I V I N D I C A T I O N E S ^1a . - Smart textile garment characterized in that it comprises: - a textile body (1) with textile electrodes for capturing bioelectric signals embroidered on the textile body (1), wherein the textile electrodes comprise a first connecting conductive thread (2), and at least 2 embroidery passes of a second conductive thread (3) with different orientations, such that a three-dimensional foam piece (4) with a thickness between 3 mm and 6 mm is interposed between the passes of the second conductive thread (3); - an electronic system for acquiring data on a user's movements and postures, such that said electronic system comprises at least 2 inertial sensors (IMU) each of which is provided with a 3-axis accelerometer and a 3-axis gyroscope; - a control electronics capable of acquiring and processing the bioelectric signals captured by the textile electrodes and inertial sensors; - a battery; and - means of transmitting the bioelectrical signals and the processed movement and posture data, the textile electrodes being associated with the control electronics for processing the captured voltage signals and transforming them into bioelectrical signals of at least one ECG signal, one EMG signal and/or one EDA signal. ^2a.- Smart textile garment, according to claim ^1a , characterized in that the control electronics include an analog-to-digital converter integrated circuit with at least 8 input channels. ^3a.- Smart textile garment, according to claims ^1a or ^2a characterized in that it has at least one PPG sensor. ^4a . - Smart textile garment, according to any one of the claims previous, characterized in that the first connecting conductor wire (2) and/or the second conductor wire (3) are made of polyamide 6.6 wires with a silver coating. 5 ^a .- Intelligent textile garment, according to claim 4 ^a characterized in that the first connecting conductor thread (2) has 8 strands of 34 filaments each. 6 ^a .- Intelligent textile garment, according to claim 4 ^a characterized in that the second conductive thread (3) has 3 strands of 34 filaments each. 7 ^a .- Smart textile garment, according to any of the previous claims, characterized in that it is configured as a vest, short-sleeved t-shirt or long-sleeved garment. ^8a.- Smart textile garment, according to claim ^7a , characterized in that, when the textile body (1) is configured as a long-sleeved garment, at least one of its sleeves has an extension (5) intended to partially cover the palm of the hand, so that the EDA measuring electrode (6) is arranged on said extension (5) in a manner coinciding with the area of the user's palm. ^9a.- Functional assessment and monitoring system of the lumbar spine characterized in that it comprises: an intelligent textile garment according to any of the preceding claims from ^1a to ^8a ; an electronic device such as a computer, smartphone or tablet provided with means of communication with the intelligent textile garment, means of receiving and storing data, a processor and software that receives the data of the bioelectrical signals captured by the intelligent textile garment, for its labeling and training of an artificial intelligence model with multi-layer machine learning architecture, such that the software compares the new data acquired by the intelligent textile garment with the trained model, for the functional assessment and monitoring of the lumbar spine of a user. ^10a . - Functional assessment and monitoring procedure of the lumbar spine using the system of claim 9a ^, characterized in that it comprises the steps of: Capture of signals by the textile electrodes and data relating to movements and postures by the electronic system and input of user pain perception data into the software; conditioning and processing of the signals by the control electronics of the smart textile garment for their transformation into bioelectrical signals of ECG, EMG and/or EDA; transmission of the data packets of the bioelectrical signals and data relating to movements and postures to the computer device; labeling of the data packets by the software according to the user's perceptions; storage of the data in a server database and training by the software of an artificial intelligence model with a predefined multilayer machine learning architecture; download of the trained models by the software and reception of new groups of user data in real time; processing of the new data by the software performing inference of the trained model to obtain the approximation to each of the labels with which the model was trained; and communication of results obtained with the artificial intelligence model on the functional state of the lumbar spine through an application installed on the computer device. ^11a . - A method according to claim ^10a , characterized in that the artificial intelligence model comprises at least one of the following: neural networks, Bayesian networks, decision tree, support vector machines, models based on genetic algorithms, regression. ^12a . - A method, according to claim ^10a or ^11a , comprising a preprocessing step of the input data provided to the artificial intelligence model by means of averaging and outlier removal operations.