Multimodal stimulation system

Abstract

The present invention relates to un system (10) for electrical stimulation of a subject, comprising: a sensing device (1) configured to detect mechanical and thermal information and to transmit them to a control unit (2), such control unit (2) in which a first algorithm associated to a first neurophysiological model capable of reproducing the behaviour of the subject's mechanoreceptors, a second algorithm associated with a second neurophysiological model capable of reproducing the behaviour of the subject's nociceptors, and a third algorithm associated with a third neurophysiological model capable of reproducing the behaviour of the subject's thermoreceptors are stored; the control unit (2) being configured to determine generation frequencies associated to different types of electrical stimulation waves, which relate to the different pieces of information which can be detected by the sensing device (1), and to transmit them to an electrical stimulator (3); such electrical stimulator (3), configured to generate and send the stimulation waves to an electrical stimulation interface (4) configured to deliver them to a subject.

Description

[0001] MULTIMODAL STIMULATION SYSTEM

[0002] DESCRIPTION

[0003] Technical field of the invention

[0004] The present invention relates to un system for electrical stimulation of a body district of a subject, configured to collect a plurality of external mechanical and thermal stimuli, and to return a corresponding plurality of mechanical (either harmless or harmful or painful) and thermal (in terms of hot and cold) sensations and perceptions to the subject herself / himself, by activating the corresponding sensory fibers.

[0005] Background

[0006] The lack of somatic sensitivity makes the interaction with the external world very complex, under several points of view.

[0007] Such lack can be, for example, the consequence of several pathologies, therefore the communication with the sensory system is compromised. The main interface between the individual and the external world is the human hand, capable of performing sophisticated motions, manipulating objects and interacting with the environment. It is characterized by a complex mechanical structure and by a sophisticated sensory system, which allow to implement manipulation tasks through a bidirectional communication with brain. The interruption of such communication with brain, caused by the absence of a sensory return, leads to an alteration in the life quality, the capabilities to perform daily living activities (ADLs), interacting with the other people or performing work-related tasks. Another area in which the absence of somatic sensitivity constitutes a big limitation is the use of robotic systems in medical or working field, which often provide an interaction between man and machine. In such applications, both in clinical field, such as for example in the myoelectric prostheses and in the surgical or rehabilitation robotics, and in working field, such as for example in the tele-operated systems, it results important - if not crucial - to return to the user somatic sensations related to the environment therewith he / she is interacting.

[0008] In fact, the adoption of tele-operated systems, for example for surgical operations or to interact with virtual environments or environments dangerous for man, limits the physical interaction and the consequent sensory perception with the external environment. The presence of sensory feedback during teleoperations (for example through the use of aptic interfaces) allows the user to have information about the contact between the tele-operated system and the environment therewith he / she interacts. In the context of the tele-operated systems, the absence of somatic sensory return can have a significant impact on the interaction between the operator and the system which is controlled remotely. The somatic sensory return, in fact, results to be crucial to provide the operator a real time feedback on the force, on the resistance and on the physical interactions which occur between the tele-operated system and the surrounding environment.

[0009] The somato-topical return of the sensory information in the man-machine interaction systems, then, would allow to improve the user experience and the comprehension of the information by creating a more intuitive and natural connection between the user and the system, allowing a better control and a better comprehension of the on-going interaction.

[0010] Up to now, different approaches have been developed to try to restore a somato-topical feedback, based on the use of different devices depending upon the sensation to be elicited and not applicable in different domains. For example, the aptic gloves used in the working context cannot be used on subjects with hand amputation. Despite the scientific and technological progresses, the fact of providing one single solution usable in different domains and which guarantees realistic and reliable sensory feedback remains an open challenge up to now.

[0011] In particular, the solutions developed to the state of art use sensors capable of capturing stimuli coming from the interaction with the external environment. The acquired signal is the input in a coding strategy, which represents the fundamental principle of the sensory return to control the stimulation to elicit the wished sensation in the subject.

[0012] The most studied coding algorithms in literature are based on the linear modulation of the fundamental parameters of the stimulation wave. More innovative strategies are based on the development of biomimetic methods which exploit the modelling of the structures involved in the somato-sensory transduction, where the main field of application is the prosthetic one. The most relevant advantage of the biomimetic algorithms is that they allow to elicit in the subjects less artificial and more intuitive sensations, by increasing the acceptability of the device by the user. The scientific studies mainly concentrate on the recreation of sensations of touch, to return to the subjects the capability of discriminating the force of interaction with the object and the texture of the surface of the object itself.

[0013] Despite in the current state of art some studies exploit biomimetic approaches for returning somatic sensations, there is no single multimodal method capable of discriminating different types of stimuli and of coding the corresponding information, at the same time also returning temperature information, through a biomimetic approach to elicit increasingly natural sensations.

[0014] Even currently commercially available man-machine interaction solutions do not provide one single system, capable of returning the user the sensory information which is exchanged between the device and the external environment during somato-topical manipulation tasks.

[0015] At present, then, the use of specific devices is required, depending upon the sensation to be elicited and the application context. This involves an increase in the system complexity and in the cognitive load for the user, who will need to decode information of different type. Moreover, the research focuses on the introduction of sensory feedback, which has to return tactile information mainly related to the recognition of the force of interaction with the object and of the surface texture, and somato-topical coding approaches for returning temperature information through electrical stimulation have not yet been developed.

[0016] Summary of the invention

[0017] The technical problem placed and solved by the present invention is then to provide a multi-modal stimulation system, to allow to elicit mechanoception, nociception and thermoception somato-sensory sensations, in particular pressure, pain and hot / cold, caused by the interaction of the system itself with the external environment, by obviating the drawbacks mentioned above with reference to the known art.

[0018] Such problem is solved with the system for electrical stimulation defined in the independent claim 1 , and by the electrical stimulation method defined in the independent claim 9.

[0019] Additional features of the invention are set forth in the dependent claims.

[0020] The task of the present invention is to provide a system for electrical stimulation configured to return to a subject mechanoception, nociception and thermoception information by implementing coding strategies based on biomimetic neurophysiological models, which replicate the behaviour of the specialized sensory structures.

[0021] The provided system has the purpose of eliciting, through a multi-modal method, the whole spectrum of tactile and thermal sensations caused by the system interaction with the external environment, aiming at replicating the human somato-sensory system which consists of different types of skin receptors, capable of integrating different qualities of exteroceptive stimuli acting on the skin. In order to emulate the complex human somato-sensory system, the sensory return strategies are integrated in one single system capable of recognizing the different stimuli coming from the external environment and using corresponding coding algorithms, which vary depending upon the detected stimulus.

[0022] The present invention therefore aims at improving and making the manmachine interaction (in medical or working field) more effective through a more natural sensory return with respect to the one obtained by the devices already known to the state of art. An additional achieved advantage is that of providing a light system, allowing to reduce the load perceived by the user with respect to the already known devices and to provide a greater sense of immersion in the (real or remote) environment, by increasing the satisfaction thereof. For the applications in prosthetic field, this translates in an increase in the sense of embodiment of the prosthetic device which includes the proposed system.

[0023] By still considering as application field the prosthetic one, the integration of sensors to detect force and temperature in a prosthesis allows the transduction of external stimuli acting on the prosthetic device following the interaction of the subject using the prosthesis with the environment. The external stimuli are processed by an algorithm capable of discriminating the piece of information and selecting the correct neurophysiological model: mechanoreceptor for tactile stimuli, nociceptor for pain stimuli or thermoreceptor for temperature stimuli. Subsequently, the model output is exploited to modulate the parameters of a stimulation wave which, once sent to the subject through neural interfaces and an electrical stimulator, will allow her / him to perceive the wished somatic sensation.

[0024] The proposed solution provides that one single system is capable of eliciting in the user mechanical and thermal sensations. The system mainly comprises: i) a sensory system required to capture the interactions with outside; ii) a control unit which processes the information transduced by the sensory system through the coding algorithms to determine the parameters of the electrical stimulation wave apt to reproduce the response of the user’s receptors; iii) an electrical stimulator capable of generating the stimulation wave to be sent to the user; iv) a neural interface therethrough the wished stimulation current is to be sent to the subject.

[0025] In other terms, the external stimuli applied on the sensory system of the invention are transduced and processed to be sent on the subject thanks to the electrical stimulation. In this way, the user is capable of perceiving the stimulus deriving from the interaction with the environment, by regaining artificially the sensory return.

[0026] As anticipated, in order to reproduce the sensory system of man, the stimulation system first of all comprises a sensory device which transduces force and temperature information. The system control unit is capable of discriminating the different stimuli linked to the interaction with the environment and detecting the correct stimulation method, by applying the correct coding algorithm among the ones stored in the control unit, which are associated to the reproduction of thermal (hot or cold), mechanical and harmful (pain) sensations.

[0027] In the present invention, in order to return temperature information, a coding strategy has been developed to elicit hot and cold thermal sensations by electrical stimulation. The coding algorithms modulate autonomously their activity based on the temperature detected by the sensory system. Depending on whether the contact takes place with a colder or hotter object than the skin temperature, the control unit is programmed to construct an electrical stimulation waveform corresponding to a heating or cooling phenomenon. The reproduced thermal sensations cover a temperature range within the one that elicits harmless sensations.

[0028] It is reiterated that the coding strategies used in the system of the invention implement a biomimetic approach, capable of imitating the behaviour of the specialized structures and / or of the fibres composing the afferent pathway and which respond to the external stimuli exerted on the skin. For example, in order to elicit the mechanical pressure sensation, the control unit is programmed to apply an algorithm based on the Izhikevich neuron model (described in Izhikevich, Eugene M. "Simple model of spiking neurons." IEEE Transactions on neural networks 14.6 (2003): 1569-1572), which reproduces the response of the afferent fibres associated with mechanoreceptors and nociceptors, responsible for coding in the human being the touch and pain information. In particular, in order to describe the behaviour of the thermoceptors, the control unit is programmed to apply an algorithm according to the Olivares model (described in: Olivares, Erick, et al. "TRPM8-dependent dynamic response in a mathematical model of cold thermoreceptor" PloS one 10.10 (2015): e0139314).

[0029] The use of models which describe the response of the fibres and skin receptors coding the sensory information aims at eliciting natural and easily interpretable sensations. Therefore, the present invention exploits a biomimetic coding strategy capable of reproducing the response of structures constituting the afferent way, such as mechanoreceptors, nociceptors and thermoceptors to stimuli of touch, pain and temperature, respectively. The pain stimuli can be traced back to pressure stimuli beyond harmful threshold values, known for the specific subject using the system.

[0030] The introduction of one single system of somatotopic sensory return in the manmachine interaction systems allows to keep the user involved during motor activities, to improve the grip performance, to improve the comprehension of the on-going interaction by also reducing the cognitive load necessary for the user to carry out the required operations, such as gripping and manipulation of objects.

[0031] The restitution of thermal sensations in the subjects allows them to detect potential sources of heat or cold that are harmful to the integrity of the system itself and / or to the user, or generally the temperature of the object itself.

[0032] The implementation of one single multi-modal system allows the user to obtain more information from contact with an object, useful to handle the grip and manipulation of the same by integrating several sensory pieces of information, thus by improving the sensory-motor coordination.

[0033] The proposed system can be used in different clinical fields, which can extend for example from the prosthetic field to the return of sensory information for the rehabilitation and assistance of subjects with sensory deficits. Moreover, the proposed system could be used in domestic environments, in which subjects who have undergone an amputation can use bidirectional prosthetic devices to carry out daily living activities.

[0034] Still, the invention is suitable to non-medical applications, such as the industrial robotics, with application for aptic interfaces for teleoperation, to ease for example the execution of procedures requiring the integration of sensory information, by consequently decreasing the cognitive effort by the operators.

[0035] Other advantages, features and use modes of the present invention will result evident from the following detailed description of some embodiments, shown by way of example and not for limitative purposes.

[0036] Brief description of figures

[0037] The figures of the enclosed drawings will be referred to by pure way of example and not for limitative purposes, wherein:

[0038] ■ Figure 1 shows a schematic representation of a preferred embodiment of a stimulation system according to the present invention;

[0039] ■ Figure 2 shows a schematic representation of a preferred embodiment of a stimulation system according to the present invention, integrated in a prosthetic device worn by a user;

[0040] ■ Figure 3 shows a schematic representation of a preferred embodiment of a stimulation method according to the present invention;

[0041] ■ Figure 4 shows a preferred embodiment of a stimulation wave delivered by means of a system according to the present invention;

[0042] ■ Figures 5a to 5c show objects used for a first, a second and a third example of mechanical stimulation of a system according to the present invention, respectively.

[0043] ■ Figures 6a to 6c show the temporal trend of the pressure stress applied to the force detection sensors included in the system of the invention for the first, second and third stimulation example, respectively; ■ Figures 7a to 7c show the temporal trend of the membrane potential associated to the first, second and third stimulation example, respectively;

[0044] ■ Figures 8a to 8e show the temperature trend over time in a fourth, fifth, sixth, seventh and eighth thermal stimulation example, respectively;

[0045] ■ Figures 9a to 9e show the temporal trend of the membrane potential associated to the fourth, fifth, sixth, seventh and eighth stimulation example, respectively;

[0046] ■ Figures 10 to 13 show the trend of responses, in terms of perception of intensity, pain, cold and hot, respectively, on a scale from 0 to 10, shown by the subjects using the system according to the present invention tested by administering the stimuli associated to the first, second, third, fourth, fifth, sixth, seventh and eighth stimulation example; and

[0047] ■ Figures 14 to 17 show the percentage of correct recognitions of the elicited sensation obtained by the subjects using the system according to the present invention and tested by administering the stimuli according to the first, second, third, fourth, fifth, sixth, seventh and eighth stimulation example.

[0048] Detailed description of preferred embodiments

[0049] Various embodiments and variants of the invention will be described hereinafter and this with reference to the above-mentioned Figures.

[0050] Analogous components are designated in the different Figures with the same numeral reference.

[0051] The different embodiments and variants described hereinafter are likely be used in combination, if compatible.

[0052] The present invention relates to a multimodal electrical stimulation system for returning somato-sensory information to a user subject during the user-system interaction, in which the returned information is mechanical and / or thermal depending upon the mechanical external stimuli and / or detected by the system, with the purpose of allowing mechanoreception, nociception and thermoception by the user subject.

[0053] According to a preferred embodiment schematically represented in Figure 1 , the stimulation system 10 mainly comprises a sensing device 1 comprising means to detect one external somato-sensory stimulus, which by way of example in the remainder of the discussion will be considered caused by the interaction with, or better manipulation of, an object. The device 1 is configured to acquire mechanical and thermal information by suitable detection means or sensors better described hereinafter.

[0054] The system 10 further comprises a control unit 2 with processes, through ad hoc developed coding algorithms, the information acquired by the sensing device 1 with the purpose of determining the features of the stimulation way more suitable to elicit in the subject using the system 10 a specific sensation, corresponding to the detected somato-sensory stimulus.

[0055] Still, the system 10 includes an electrical stimulator 3 capable of generating the stimulation wave to be sent to the user, and a neural interface or electrical stimulation interface 4, which comprises means configured to deliver to the subject the electrical stimulation determined by the stimulator 3.

[0056] According to a preferred variant, the sensing device 1 comprises at least one force sensor 12 having a plurality of operating areas. The sensor 12 is configured to detect the temporal trend of the intensity of a force F applied at one or more of said operating areas, the pressure P generated by such force at each one of the operating areas and the number N of operating areas on which the force F is applied. Such features are important to transduce both the force level and the contact area of the sensor 12 with the manipulated object. Moreover, the device 1 comprises at least one temperature sensor 13, configured to detect the temperature trend over time Tobj at the operating areas of the force sensor 12, as well as preferably to detect the temperature of the user’s skin. The data detected by means of the device 1 correspond for example to force and temperature data recorded during the tasks of gripping and manipulating the already mentioned object by the user subject.

[0057] The force and temperature sensors 12, 13 can be of different type, and preferably be suitable to measure force ranges typically applied during the manipulation in tasks which provide the interaction with objects of everyday life, as well as harmless temperature range. Moreover, the sensors 12, 13 preferably have sizes which make them easily integrable in different environments or terminal organs of manipulators.

[0058] The sensing device 1 is further configured to transmit the so detected force and temperature data to the control unit 2, wherein such data are processed, preferably after having been stored. In particular, the sensing device 1 is configured to be connected to, or integrated in, an element which can be worn by the user subject, for example a prosthesis.

[0059] The control unit 2 which receives the data detected by the device 1 can be implemented by means of a PC or a integrated control board. The control unit 2 is preferably arranged for the real-time processing of the data acquired through the device 1 , which are sent as input to the coding algorithms (shown in detail in the subsequent section) stored in the control unit 2. The latter, then, manages the information and generates the commands and stimulation parameters to be sent to the electrical stimulator 3.

[0060] In the control unit 2 the algorithms implementing the physiological models of the behaviour of the receptors of the specific subject using the system, in particular the mechanoreceptors, the nociceptors and the thermoreceptors, are stored.

[0061] In particular, for the specific subject who will be the system user, the mathematical modes are defined which reproduce the behaviour of the receptors of the pressure, pain and temperature tactile stimuli.

[0062] Such algorithms reproduce: a first neurophysiological model capable of reproducing the behaviour of the subject’s mechanoreceptors implementing a mathematical relationship between the temporal trend of a force F and the frequency of generation of a membrane potential of the aforementioned mechanoreceptors; a second neurophysiological model capable of reproducing the behaviour of the subject’s nociceptors implementing a mathematical relationship between the temporal trend of a force F and the frequency of generation of a membrane potential of the aforementioned nociceptors; and a third neurophysiological model capable of reproducing the behaviour of the subject’s thermoreceptors implementing a mathematical relationship between the trend of a temperature over time Tobj and the frequency of generation of a membrane potential of the aforementioned thermoreceptors.

[0063] Moreover, in the control unit 2 the characteristic parameters of the subject who will be the system user are stored, obtained by means of test or training sessions in which the subject uses the system of the invention and is subjected to the application of different types of forces and temperature variations with the purpose of establishing the threshold values corresponding to the subject sensitivity in appreciating the proposed stimuli.

[0064] A first characteristic parameter is the maximum number Nth of operating areas of the force sensor 12 which are activated by the force F and which result in a pain stimulation for the subject. In other terms, such first characteristic parameter corresponds to the greater number Nth of operating areas of the force sensor 12 thereat the application of the force F is detected when the subject feels a pain stimulation. Such number Nth can be detected experimentally for the specific subject thereto the system is intended by means of a calibration phase, for example by applying the force F at first at all operating areas and progressively at an increasingly lower number of the latter, until when the subject recognizes a painful stimulation.

[0065] A second parameter is the minimum pressure p detected dal force sensor 12 in association to a pain stimulation for the subject when the force F is localized at a number of operating areas lower than such maximum number Nth of operating areas. Even this second parameter can be determined experimentally for the specific subject thereto the system is intended by means of a calibration phase, for example by applying a force F of increasingly greater intensity, starting from an initial force intensity corresponding to a not painful stimulation for the subject, at a number of operating areas lower than Nth, until the subject recognizes a pain stimulation.

[0066] It is noted that, alternatively to the consideration of the number of operating areas and the count of the latter, an evaluation of the force sensor area involved by the stimulation can be used and the percentage with respect to the total area of the sensor can be evaluated. A third parameter is the maximum amplitude PAmax of the current pulse that can be delivered to the subject, beyond which a muscle contraction is generated. Additional parameters are the duration PWmech of a current pulse for which the subject perceives a non-painful mechano-tactile sensation, the duration PWnoc of a current pulse for which the subject perceives a painful mechano-tactile sensation and the duration PWtemp of a current pulse for which the subject perceives a non-painful thermal sensation.

[0067] In front of the above-mentioned algorithms and parameters stored in memory, the control unit 2 is configured to process the data incoming from the sensing device 1 to determine the type of detected stress (in general terms: pressure, pain and / or hot / cold) and to determine a corresponding electrical stimulation wave capable of reproducing artificially in the sensory fibres of the user subject the response caused by such specific stimulus, modelled depending upon the response which has been detected in the subject during training.

[0068] In particular, if the number of operating areas activated by the force sensor 12 is equal or greater than the maximum number of operating areas Nth, the control unit 2 is configured to provide as input to the first algorithm data relating to the trend of the force over time F to obtain as output the frequency of generation of a first membrane potential associated with the mechanoreceptors.

[0069] In case the number of operating areas activated by the force sensor 12 is lower than the maximum number of operating areas Nth and the pressure detected at such activated operating areas is greater than or equal to the minimum pressure p, the control unit 2 is configured to provide as input to the second algorithm data relating to the trend of the force over time F to obtain as output the frequency of generation of a second membrane potential associated with the nociceptors.

[0070] Additionally, the control unit 2 is configured to provide as input to the third algorithm the data relating to the temperature trend over time Tobj and the temperature of the subject’s skin, to obtain as output the frequency of generation of a third membrane potential associated with the thermoreceptors.

[0071] In summary, the algorithms for determining the parameters of the stimulation wave are based on the use of the variable outgoing from the respective neurophysiological model, such as the frequency of generating the action potentials, to modulate the stimulation frequency of a wave which is preferably square, biphasic and symmetrical, as it will be described later. The remaining parameters of the stimulation wave, such as the current amplitude and the pulse duration, are subject-specific and predetermined during the subject training phase.

[0072] The control unit 2 is further configured to transmit the so-determined frequencies of generation of the first, second and third membrane potential to the electrical stimulator 3, which in turn is configured to generate stimulation waves, having a modulated frequency depending on that of the respective membrane potential, to be sent to the stimulation interface 4.

[0073] The electrical stimulator 3 can be implemented by means of any general device capable of delivering the stimulation waves programmable by interfacing with the control unit 2. The stimulator 3 preferably has to be controlled under current, so as to guarantee the safe delivery on the subject of a known current, independently from the impedance of the subject tissue on which the stimulation is applied. Preferred features of the electrical stimulator 3 are: the presence of two stimulation channels; stimulation currents, for each channel, included in a range of at least 0.05mA - 5mA with a resolution of 0.01 mA; programmability; stimulation frequencies which can be delivered in a range of at least 1 Hz - 500Hz.

[0074] In particular, the electrical stimulator 3 is configured to generate three types of stimulation waves, each one associated to a sensory stimulus to be reproduced, which can be associated to pressure, pain and temperature. The period of the stimulation waves consists of three main parameters: frequency (Pulse Frequency, PF), duration (Pulse Width, PW) and amplitude (Pulse Amplitude, PA). In particular, the electrical stimulator 3 is configured to generate first square, biphasic and symmetrical electrical stimulation waves designed to elicit in the subject a non-painful mechano-tactile sensation, whose frequency PF is modulated by the frequency of generation of the first membrane potential. The stimulation wave has a pulse amplitude PA equal to the maximum amplitude PAmax and a pulse duration PW equal to the first duration PWmech.

[0075] The electrical stimulator 3 is further configured to generate second square, biphasic and symmetrical electrical stimulation waves designed to elicit in a subject a painful mechano-tactile sensation, whose frequency PF is modulated by the frequency of generation of the second membrane potential. The stimulation wave has a pulse amplitude PA equal to the maximum amplitude PAmax and a pulse duration PW equal to the second duration PWnoc.

[0076] Still, the electrical stimulator 3 is configured to generate third square, biphasic and symmetrical electrical stimulation waves designed to elicit in a subject a non-painful thermal sensation, whose frequency PF is modulated by the frequency of generation of the third membrane potential. The stimulation wave has a pulse amplitude PA equal to the maximum amplitude PAmax and a pulse duration PW equal to the third duration PWtemp.

[0077] The control unit 2 is further configured to transmit the frequencies of generation of the first, second and third membrane potential to an electrical stimulator 3. The electrical stimulator 3 is configured to generate and send the stimulation waves to the stimulation interface 4. Such stimulation interface 4, in turn, is configured to receive the first, second and third electrical stimulation waves and to deliver them to the subject using the system 10. According to preferred variants of the invention, the stimulation interface 4 comprises one or more skin electrodes, configured to be applied at a skin surface of the user subject; alternatively, or additionally, the interface 4 comprises one or more electrodes which can be implanted in the user subject.

[0078] According to a preferred embodiment of the invention in prosthetic field, an interface 4 can be used configured for the electrical stimulation of the peripheral nervous system in not invasive way, that is by transcutaneous electrical nerve stimulation (TENS). Within such implementation the affixing, as neural interface, of self-adhesive surface electrodes on the subject skin at the median and ulnar nerves, which innervate the palmar side of the hand, can be provided. In particular, for each nerve the use of a pair of electrodes, an active electrode and a return electrode, is provided. Figure 2 shows an exemplification of such embodiment variant, comprising: a prosthesis of the hand 100, which integrates the sensing device 1 comprising the force and temperature sensors; a control unit and an electrical stimulator for generating the stimulation waves according to what already described (not represented); and two pairs of not invasive electrodes for the electrical stimulation, applied on the skin of the subject’s stump at the (median and ulnar) nerves which innervate the palm of the hand.

[0079] Therefore, it will be understood that the present invention also relates to a method for the generation of an electrical stimulus suitable for eliciting the perception of mechanical and thermal stimuli by a subject.

[0080] The proposed method comprises a first phase for detecting mechanical and thermal data by means of: at least one force sensor 12 comprising a plurality of operating areas, configured to detect the temporal trend of the intensity of a force F applied at one or more of such operating areas, the pressure P generated by such force at each one of said operating areas and the number N of operating areas on which the force F is applied; and at least one temperature sensor 13, configured to detect the temperature trend over time Tobj at the above-mentioned operating areas.

[0081] Still, it is necessary to determine: a first algorithm associated to a first neurophysiological model capable of reproducing the behaviour of the subject’s mechanoreceptors, wherein such first algorithm implements a mathematical relationship between the temporal trend of a force F and the frequency of generation of a membrane potential of the aforementioned mechanoreceptors; a second algorithm associated with a second neurophysiological model capable of reproducing the behaviour of the subject’s nociceptors, wherein such second algorithm implements a mathematical relationship between the temporal trend of a force F and the frequency of generation of a membrane potential of the aforementioned nociceptors; and a third algorithm associated with a third neurophysiological model capable of reproducing the behaviour of the subject’s thermoreceptors, wherein such third algorithm implements a mathematical relationship between the trend of a temperature over time Tobj and the frequency of generation of a membrane potential of the aforementioned thermoreceptors.

[0082] Still, for implementing the proposed method it is necessary to determining the following characteristic parameters: the maximum number of operating areas Nth of the force sensor 12 which are activated by the force F in association to a pain stimulation for the subject; the minimum pressure p detected by said force sensor 12 in association to a pain stimulation for the subject when the force F is localized at a number of operating areas lower than the maximum number of operating areas Nth; the maximum amplitude PAmax of a current pulse that can be delivered to the subject, beyond which muscle contraction is generated; a first duration PWmech of a current pulse for which the subject perceives a nonpainful mechano-tactile sensation; a second duration PWnoc of a current pulse for which the subject perceives a painful mechano-tactile sensation; a third duration PWtemp of a current pulse for which the subject perceives a nonpainful thermal sensation.

[0083] The method then includes to provide as input to the first algorithm data relating to the trend of the force over time F to obtain as output the frequency of generation of a first membrane potential associated with the mechanoreceptors, if the number of operating areas activated by the force sensor 12 is equal or greater than the maximum number of operating areas Nth of the force sensor 12. Or, if the number of activated operating areas of the force sensor 12 is less than the maximum number of operating areas Nth of the force sensor 12 and the pressure detected at such activated operating areas is greater than or equal to the minimum pressure p, the method includes to provide as input to the second algorithm data relating to the trend of the force over time F, to obtain as output the frequency of generation of a second membrane potential associated with the nociceptors. Moreover, the method provides to provide as input to the third algorithm data relating to the temperature trend over time Tobj, to obtain as output the frequency of generation of a third membrane potential associated with the thermoreceptors.

[0084] Subsequently, such frequencies of generation of the first, second and third membrane potential are provided as input to an electrical stimulator 3, which is configured to generate: first square, biphasic and symmetrical electrical stimulation waves designed to elicit in the subject a non-painful mechano-tactile sensation, whose frequency PF is modulated by the frequency of generation of the first membrane potential, wherein the stimulation wave has a pulse amplitude PA equal to said maximum amplitude PAmax and a pulse duration (PW) equal to the first duration PWmech; second square, biphasic and symmetrical electrical stimulation waves designed to elicit in the subject a painful mechano-tactile sensation, whose frequency PF is modulated by the frequency of generation of the second membrane potential, wherein the stimulation wave has a pulse amplitude PA equal to the maximum amplitude PAmax and a pulse duration PW equal to the second duration PWn0C; third square, biphasic and symmetrical electrical stimulation waves designed to elicit in the subject a non-painful thermal sensation, whose frequency PF is modulated by the frequency of generation of the third membrane potential, wherein the stimulation wave has a pulse amplitude PA equal to the maximum amplitude PAmax and a pulse duration PW equal to the third duration PWtemp.

[0085] The electrical stimulator 3 sends the above-mentioned first, second and third electrical stimulation waves to a stimulation interface 4, which in turn delivers them to the subject.

[0086] With reference to Figure 3, it illustrates the salient phases of a preferred embodiment of the biomimetic method to elicit somato-topic sensations of touch, pain and temperature according to the present invention.

[0087] The method comprises a first phase for characterizing the sensations for estimating the stimulation parameters which, in turn, comprises the sub-phases for positioning correctly the neural interface and determining the subject-specific stimulation. The neural interface has to be positioned on the neural tissue of interest of the subject in order to elicit the wished somatic sensation. In particular, the optimum positioning of the interface is determined by evaluating the selectivity of the elicited area and the quality of sensation reported by the subject upon varying the stimulation parameters.

[0088] After the optimum positioning of the neural interface, the subject is administered to a protocol for characterizing the sensations, in which the fundamental parameters of the stimulation wave are varied in specific ranges.

[0089] With reference to the considered preferred implementation, Figure 4 shows a symmetric biphasic square wave and the related stimulation parameters, thereamong: the pulse amplitude PA; the pulse duration PW; the period T; the pulse frequency PF; the overall duration of the stimulus L; the last pulse n.

[0090] The purpose of administering to the subject the protocol for characterizing the sensations is the determination of its characteristic parameters, such as:

[0091] PAmax: threshold value of the maximum current amplitude that can be delivered to the subject, beyond which muscle contraction is generated;

[0092] PWmech: PW value for which the subject perceives a distinguishable and comfortable mechano-tactile sensation and for which he / has has evaluated the perceived intensity greater than or equal to Ith (threshold value indicating a moderate intensity) and the perceived pain less than Pth (threshold value designated a mild pain);

[0093] PWnoc: a distinguishable mechano-tactile sensation for which the subject has evaluated the perceived intensity greater than or equal to Ith (threshold value designating a moderate intensity) and the perceived pain greater than or equal to Pth (threshold value indicating a mild pain);

[0094] PWtemP: PW value for which the subject perceives a distinguishable and comfortable sensation and for which he / she has evaluated the perceived intensity greater than or equal to Ith (threshold value indicating a moderate intensity) and the perceived pain less than or equal to Pth (threshold value indicating a mild pain). The so-determined values are used in the coding algorithms to construct the stimulation wave suitable to elicit the wished somatic sensation.

[0095] The second phase of the proposed method relates to the return of sensory information and comprises sub-phases related to: transduction of the somatosensory (mechano-tactile and / or thermal) information; identification and resolution of the neurophysiological model which describes the behaviour of the skin receptors involved in the response to such somato-sensory information; implementation of the algorithm for coding the somato-sensory information; delivery of the stimulation wave generated by the mechano-tactile or thermal coding algorithm to the user subject through the interface.

[0096] As anticipated, the sensing device allows to detect the mechanical and thermal information in the interaction environment, in particular the trend of the force over time F and the contact area of the sensor with the object manipulated by the force sensor, and the trend of temperature of the manipulated object Tobj over time by means of the temperature sensor.

[0097] This information is given as input to the neurophysiological model selected to describe the behaviour of the specialized structures composing the afferent pathway, coding the tactile or thermal somato-sensory information.

[0098] The afferent structures responsible for the sensory transduction on the skin are the sensory skin receptors. In particular, the mechanoreceptors and the nociceptors respond respectively to the harmless mechanical stimuli and nociceptive agents on the skin.

[0099] According to preferred variants of the invention, the neurophysiological model adopted to describe the behaviour of the mechanoreceptors and of the nociceptors is the Izhikevich neurone model. The Izhikevich model is a simple biologically plausible neuron model such as the Hodgkin-Huxley model, but computationally more efficient, such as the Integrate-and-Fire (IF) model.

[0100] The behaviour of the mechanoreceptors is described by the Izhikevich neuron model by combining the features of the slowing adapting (Slowly Adapting, SA) and rapidly adapting (Rapidly Adapting, RA) receptors, respectively, at regular and rapid frequency. The model of the nociceptors exploits the dynamics of the dell’algoritmo di codifica delle informazioni somato-sensoriali; erogazione dell’onda di stimolazione generata dall'algoritmo di codifica meccano-tattile o termica al soggetto utilizzatore tramite l'interfaccia Come anticipato, i rilevare le informazioni meccaniche e te particolare l‘andamento della forza nel te on l’oggetto manipolato mediante il sens emperatura dell’oggetto manipolato Tobjn atura. Queste informazioni sono date in ingresso al modello neurofisiologico scelto per descrivere il comportamento delle struttu ate che compongono la via afferente, che codificano le informazioni somato-sensoriali tattili o termiche. Le strutture afferenti responsabili della trasduzione sensoriale sulla pelle sono i recettori cutanei sensoriali. In particolare, i meccanorecettori e i nocicettori rispondono rispettivamente agli stimoli meccanici innocui e nocicettivi agenti sulla cute. In accordo a varianti preferite dell’invenzione, il modello neurofisiologico adottato per descrivere il comportamento dei meccanorecettori e dei nocicettori il modello di neurone di Izhikevich. Il modello di Izhikevich è un semplice modello di neurone biologicamente plausibile come il modello di Hodgkin- Huxley, ma computazionalmente più efficiente come il modello Integrate-and- Fire (IF). Il comportamento dei meccanorecettori viene descritto dal modello di neurone di Izhikevich combinando le caratteristiche dei recettori a lento adattamento (Slowly Adapting, SA) e a rapido adattamento (Rapidly Adapting, RA), rispettivamente, a frequenza regolare e rapida. Il modello dei nocicettori sfrutta la dinamica dei neuroni Izhikevich RA per imitare il comportamento delle terminazioni nervose libere. Le equazioni fondamentali del modello implementate sono riportate di seguito: dove è il potenziale di membrana del neurone e la variabile di recupero.Variando il valore dei quattro parametri ( , , , ), il modello riproduce ilcomportamento di diversi tipi di neuroni. Il valore associato al parametro determina la selezione del modello di neurone da riprodurre, ossia RS per il modello di meccanorecettore o FS per il modello di nocicettore. P rappresenta la pressione in ingresso esercitata durante il contatto con un oggetto. Essa varia in base all'area di contatto dell’oggetto sul sensore presente all’interfaccia di interazione e alla forza applicata F. Pertanto, l’informazione tattile di contatto può essere innocua o nocicettiva, ed essere riprodotta rispettivamente dal modello dei meccanorecettori o nocicettori. Il parametro rappresenta un fattore di scala della grandezza in ingresso. La selezione del modello recettoriale da utilizzare è descritta dallo pseudocodice riportato più avanti, dove ^^ è il numero di aree del sensore di forza attive e nth è un valore soglia del numero di aree attivate. Questa soglia definisce il grado di localizzazione della pressione applicata. ρ è un valore soglia di pressione, per cui una pressione localizzata al di sopra di questa soglia determina l’attivazione del modello di nocicezione. In altre parole, un oggetto appuntito può essere manipolato senza dolore se la pressione sulla pelle non supera la soglia del dolore. Il metodo proposto meccanorecettori e dei nocicettori quando a contatto con diversi oggetti aventi superfici con raggi di curvatura differenti, ad esempio con oggetti dalla superficie piatta, oppure appuntita, a contatto con l’interfaccia di interazione del sistema (cioè, con il dispositivo di rilevazione). Di seguito è riportato lo pseudocodice implementato: I recettori cut e informazioni di temperatura son . no in termorecettori del freddo e termorecettori del caldo, e sono tipicamente posizionati a diverse profondità della pelle. In accordo a varianti preferite, per descrivere l’andamento del potenziale di questi recettori quando sottoposti a degli stimoli termici (sia statici sia dinamici, mostrando quindi una varia zione della frequenza di attivazione in seguito a raffreddamento o riscaldamento e distinguendo tra aumento e diminuzione della temperatura) si adotta un modello neurofisiologico dei termorecettori del freddo. In particolare, per descriver egato il modello di Olivares, che descrive la risposta n intervallo di temperatura sufficientemente ampio (10 o il comportamento dei soli termorecettori del freddo. L’equazione fondamentale si basa sul modello di Huber & Braun dei recettori del freddo, a cui è stato aggiunto il termine di corrente relativo al canale TRPM8. In accordo a tale implementazione, il canale TRPM8 è considerato l'entità molecolare chiave per il rilevamento delle temper edde nel sistema dell’invenzione. te IM8 è stata aggiunta per tenere conto della risposta dinamica nelle terminazioni nervose sensibili al freddo. Di seguito, si riporta la suddetta equazione fondamentale: dove Cm è la capacità di membrana; Id, Ir, Isd e Isr sono rispettivamente le correnti di depolarizzazione, ripolarizzazione, depolarizzazione lenta e ripolarizzazione lenta; IM8 è la corrente mediata dal canale TRPM8; Il corrisponde ad una corrente di perdita aspecifica e Iwn è un termine di rumore. Le correnti sono calcolate come segue the initial temperatures of the skin and the object, as expressed by the following formula: where Tsis the interface temperature, k is the thermal conductivity, p is the density, c is the specific heat, is the initial temperature of the object read by the somato-sensory device, is the initial temperature of skin and

[0101] (kpc)1 / 2is the contact coefficient.

[0102] The following equation describes depending upon time and skin depth: where x is the position inside the skin in which the temperature (x of the cold thermoreceptor is equal to 160 pm) is calculated, Tsis the interface temperature, t is time and aSkin is the thermal diffusivity of the skin.

[0103] Once determined the above-mentioned parameters, the implementation of the algorithm for coding the detected somato-sensory information is provided.

[0104] In case of mechano-tactile stimulation, the output of the neurophysiological model describing the behaviour of the mechanoreceptors and of the nociceptors is the membrane potential. The membrane potential is capable of capturing the specific patterns of space-time pulses which are typical of the receptor behaviour for different stimuli. This piece of information is lost when only the neuron discharge frequency is taken into consideration. Consequently, the stimulation signals based on the discharge frequency only would probably generate electrical stimuli difficult to be discriminated for the subjects.

[0105] Therefore, in the biomimetic strategy proposed in the present invention, the membrane potential modulates the stimulation frequency of a square, biphasic and symmetric wave. In particular, every time an action potential occurs, a period of the stimulation waveform is generated.

[0106] As described above, the frequency PF of the wave to be delivered depends upon the frequency for generating the receptor action potentials. The other parameters are predetermined in the first phase of the method, wherein the somatic sensations are characterized upon varying the characteristic stimulation parameters. In particular, the pulse amplitude PA is fixed at the maximum stimulation threshold PAmax, whereas the duration of the pulse PW varies depending upon the selected receptor model: for the mechanoreceptor model it is fixed at a value indicated as PWmech, for the nociceptor model it is fixed at a value indicated as PWnoc.

[0107] At this point, a coding algorithm results to be defined based on the thermoreceptor membrane potential, consisting in a time sequence of action potentials. According to the implemented strategy, the output signal is directly converted in stimulation paradigm required to cause the wished sensory perception. Therefore, the stimulation current administered to the subject replicates the time trend of the thermoreceptor membrane potential in response to a specific thermal stimulus.

[0108] The coding algorithm works on the stimulation frequency by constructing a symmetric biphasic square wave whenever the thermoreceptor produces an action potential. The PW of the stimulation wave was fixed at PWtemp, whereas the specific PA of the subject is PAmax. As shown previously, such parameters are predetermined in the first method phase.

[0109] In the last method phase, the stimulation wave generated by the mechano- tactile or thermal coding algorithm is sent to the control unit and subsequently transmitted, through the interface, to the user subject.

[0110] According to the transduced information and the applied strategy, the subject perceives a force and / or temperature somatic sensation upon the contact of the sensing device with the object.

[0111] A variant of the present invention can provide the use of implantable stimulation means, that is an interface configured to be installed more or less permanently within the organism of the user subject of the system of the invention. For example, a plurality of stimulation electrodes can be installed at the nerves of the subject by surgical operation. The electrodes can be of intraneural (crossing the nerve transversally or longitudinally) or epineural (surrounding the nerve externally around the epineurium) type.

[0112] Another possible variant of the invention in the prosthetic application field is the use of neural interfaces for the central nerve system of the user, that is of electrodes configured to stimulate electrically the somato-sensory cortex, intended to receive the somato-sensory stimuli, such as (fine and coarse) touch, temperature, pain, pressure.

[0113] The biomimetic method for coding sensory information through the multimodal system of the invention was validated, in a preliminary study, on 15 healthy subjects through the TENS, by administering an experimental protocol for eliciting mechano-tactile, nociceptive or thermoceptive sensations. The validation was performed by using a system for the non invasive stimulation, of the type shown in Figure 2.

[0114] Different mechano-tactile and thermal stimuli were generated to test the coding strategies. The mechano-tactile stimuli were obtained in preliminary phase through the use of three real objects. They were connected to a traction machine which applies a constant force (equal to 8N) on a sensory device, consisting of five sensitive areas. In this way, the recorded signals were then simulated as inputs into the proposed method with the purpose of subjecting all subjects to the same stimuli.

[0115] The contact with three objects, each one shown in Figure 5a, 5b and 5c respectively, was simulated, having different extension and morphology of the surfaces of contact with the detecting device, which can be integrated in the prosthesis worn by the subject:

[0116] • the Object with a contact area equal to 38.48 mm2;

[0117] • the Object 2 with a contact area equal to 153.94 mm2; and

[0118] • the Object 3 with a contact area equal to 346.36 mm2.

[0119] Figures 6a, 6b and 6c represent the mechanical stimuli for validating the coding strategy for the force information; in particular, each Figure shows the trend of the pressure detected at the force sensor, exerted respectively by the grip of the three objects. The pressure threshold p was fixed at 40kPa. Figures 7a, 7b and 7c show respectively the trend of the membrane potential associated to each one of the mechanical stimuli illustrated in Figures 6a, 6b and 6c, and then to each one of the three objects used in the test.

[0120] In this example, in order to simulate the contact between the three different objects with different parts of the fingertips of a prosthetic hand, a sensory device was considered consisting of five operating areas (or sensitive areas).

[0121] Based upon the involved contact surface, three cases occurred:

[0122] • the Object 1 activates all areas;

[0123] • the Object 2 activates three sensitive areas only;

[0124] • the Object 3 activates one single sensitive area.

[0125] Based upon the number of activated areas n and the pressure level applied thereon p, the corresponding receptor model is selected. The threshold nth was fixed at 3. From Figure 6c it can be noted that the Object 3 activates the nociception model (n<nth and P>p); therefore, a nociception sensation is associated thereto. On the subject a square, biphasic, symmetric stimulation wave is then applied, having PA fixed to the maximum current threshold of the subject PAmax, PW equal to the PWnoc of the subject and PF modulated by the frequency of the action potentials visible in the result of the nociception model Vm.

[0126] As far as the thermal stimuli are concerned, the contact with five types of stimulus was simulated, represented in Figures 8a to 8e, and therefore Figures 9a to 9e show the trend of the respective membrane potentials:

[0127] • static stimulus in the cold range, 20°C;

[0128] • static stimulus in the hot range, 35°C;

[0129] • dynamic stimulus of moderate cooling (contact with a cold object made of wood), starting from a basal skin temperature equal to 33°C;

[0130] • dynamic stimulus of intense cooling (contact with a cold object made of aluminium), still starting from a basal skin temperature of 33°C.

[0131] • stimulus of heating (contact with a hot object made of aluminium), starting from a skin adaptation temperature of 20°C.

[0132] The experimental protocol provided the following operating steps: Mapping phase: the threshold values and the parameters to be fixed specific for each subject (PAmax, PWmech, PWnoc, PWtemp) are determined;

[0133] Training phase: each mechano-tactile and thermal stimulus is administered to the subject for three times (Cold, Hot, Moderate cooling, Intense cooling, Heating, Object 1 , Object 2, Object 3). Then, the participant was asked to characterize the perceived sensation in terms of quality (tingling, vibration, pressure, and so on), naturalness, depth, elicited region, intensity, pain, cold and heat;

[0134] Test phase: for each (Static, Dynamic, Pain and Combined) protocol each stimulus was administered five times in random order. In the Combined protocol, the subject has to discriminate the sensation among the dynamic stimuli of temperature and the Object 3, that is to discriminate mechano-tactile stimuli from thermal stimuli. The order for administering each protocol was different and randomized for the subjects participating to the study.

[0135] Figures 10 to 13 show the evaluation of the different stimuli, in particular they represent the trend of the responses in terms of intensity, pain, cold and hot shown by the subjects to the stimuli administrated during the training phase, in a scale from 0 to 10. A result to be highlighted is related to the variation in perceived intensity among the mechano-tactile stimuli, therefore it increases from Object 1 to Object 3. As expected, upon a greater pressure applied on the sensor, the algorithm elicits in the subject a sensation of greater intensity. Moreover, for Object 3 the perceived pain obtained the higher median value with respect to the other stimuli, since it is the one thereto the coding for the nociceptive pressure sensation is associated. At last, the cold was reported only for the static and dynamic stimuli related to the sensations of cold temperature; whereas the hot sensation was reported more for the hot static stimulus at 35°C.

[0136] Figures 14 to 17 show the results obtained about the discrimination of the elicited sensations. The Success Rate (SR), that is the percentage of the correct responses on the total number of stimuli, was calculated for each administered protocol. The temperature protocols obtained a lower SR average value with respect to the other protocols. This can be due to the fact that the participants had difficulties in reminding the correct name of the stimuli despite the elicited sensations were well distinguishable. This result was probably influenced by the order of administering the protocols.

[0137] The protocol of the mechano-tactile sensations obtained a success rate in the correct detection of the stimuli (SR) equal to 84.44%; in particular, Object 3, which was also the one perceived as more intense and bothersome, resulted to be the easiest one to be distinguished: even in the Combined Protocol it reached an average value of correct discrimination percentage of 88%.

[0138] The present invention has been sofar described with reference to preferred embodiments. It is to be meant that other embodiments belonging to the same inventive core may exist, as defined by the protective scope of the herebelow reported claims.

Claims

CLAIMS1. A system (10) for electrical stimulation of a subject, comprising: a sensing device (1) configured to detect mechanical and thermal data, wherein said sensing device (1) comprises:• at least one force sensor (12) comprising a plurality of operating areas, configured to detect the temporal trend of the intensity of a force (F) applied at one or more of such operating areas, the pressure (P) generated by such force (F) at each one of said operating areas and the number (N) of operating areas on which the force (F) is applied; and• at least one temperature sensor (13), configured to detect the temperature trend over time (Tobj) at said operating areas; wherein said sensing device (1) is configured to transmit the data detected by said at least one force sensor (12) and at least one temperature sensor (13) to a control unit (2), said control unit (2), in which the following is stored: a first algorithm associated to a first neurophysiological model capable of reproducing the behaviour of the subject’s mechanoreceptors, said first algorithm implementing a mathematical relationship between the temporal trend of a force (F) and the frequency of generation of a membrane potential of the aforementioned mechanoreceptors; a second algorithm associated with a second neurophysiological model capable of reproducing the behaviour of the subject’s nociceptors, said second algorithm implementing a mathematical relationship between the temporal trend of a force (F) and the frequency of generation of a membrane potential of the aforementioned nociceptors; and a third algorithm associated with a third neurophysiological model capable of reproducing the behaviour of the subject’s thermoreceptors, said third algorithm implementing a mathematical relationship between the trend of a temperature over time (Tobj) and the frequency of generation of a membrane potential of theaforementioned thermoreceptors; and the following characteristic parameters:• a maximum number (Nth), equal to the largest number of operating areas of said force sensor (12) thereat the application of the force (F) is detected at a pain stimulation for the subject,• a minimum pressure (p), equal to the smallest pressure value detected by said force sensor (12) at a pain stimulation for the subject when the force (F) is localized at a number of operating areas lower than said maximum number (Nth),• maximum amplitude (PAmax) of a current pulse that can be delivered to the subject, beyond which muscle contraction is generated;• first duration (PWmech) of a current pulse for which the subject perceives a non-painful mechano-tactile sensation;• second duration (PWnoc) of a current pulse for which the subject perceives a painful mechano-tactile sensation;• third duration (PWtemp) of a current pulse for which the subject perceives a non-painful thermal sensation; the control unit (2) being configured: if the number of activated operating areas of said force sensor (12) is equal to or greater than said maximum number (Nth), to provide as input to said first algorithm data relating to the trend of the force (F) over time to obtain as output the frequency of generation of a first membrane potential associated with the mechanoreceptors; and if the number of activated operating areas of said force sensor (12) is less than said maximum number (Nth) and the pressure detected at such activated operating areas is greater than or equal to the minimum pressure (p), to provide as input to said second algorithm data relating to the trend of the force (F) over time to obtain as output the frequency of generation of a second membrane potential associated with the nociceptors, to provide as input to said third algorithm data relating to the temperature trend over time (Tobj) to obtain as output the frequency of generation of a thirdmembrane potential associated with the thermoreceptors, said control unit (2) being further configured to transmit said frequencies of generation of the first, second and third membrane potential to an electrical stimulator (3); said electrical stimulator (3), configured to generate and send to a stimulation interface (4): first square, biphasic and symmetrical electrical stimulation waves designed to elicit in the subject a non-painful mechano-tactile sensation, whose frequency (PF) is modulated by the frequency of generation of said first membrane potential, wherein the stimulation wave has a pulse amplitude (PA) equal to said maximum amplitude (PAmax) and a pulse duration (PW) equal to said first duration (PWmech); second square, biphasic and symmetrical electrical stimulation waves designed to elicit in the subject a painful mechano-tactile sensation, whose frequency (PF) is modulated by the frequency of generation of said second membrane potential, wherein the stimulation wave has a pulse amplitude (PA) equal to said maximum amplitude (PAmax) and a pulse duration (PW) equal to said second duration (PWnoc); third square, biphasic and symmetrical electrical stimulation waves designed to elicit in the subject a non-painful thermal sensation, whose frequency (PF) is modulated by the frequency of generation of said third membrane potential, wherein the stimulation wave has a pulse amplitude (PA) equal to said maximum amplitude (PAmax) and a pulse duration (PW) equal to said third duration (PWt em P)> said stimulation interface (4), configured to receive said first, second and third electrical stimulation waves and to deliver them to a subject.

2. The system (10) according to the previous claim, wherein said stimulation interface (4) comprises one or more skin electrodes.

3. The system (10) according to one of the previous claims, wherein said stimulation interface (4) comprises one or more implantable electrodes.

4. The system (10) according to one of the previous claims, wherein said first and second neurophysiological models correspond to the Izhikevich neuron model.

5. The system (10) according to one of the previous claims, wherein said third neurophysiological model implements the following equation:where Cm is the membrane capacitance; Id, lr, Isd and lsrare the depolarization, repolarization, slow depolarization and slow repolarization currents, respectively; IMS is the current mediated by the molecular channel TRPM8; II corresponds to a non-specific leakage current; lWn corresponds to noise; and IM8 is a corrective term associated with the dynamic response of cold-sensitive nerve endings; wherein the currents are calculated as follows:wherein a is an activation term representing the probability of opening of the channels, g is the maximum conductance density, E is the reversal potential and p(T) is a temperature-dependent current correction factor.

6. The system (10) according to one of the previous claims, wherein said sensing device (1 ) is wearable, preferably shaped like a glove.

7. A prosthesis (100) comprising un system (10) for electrical stimulation of a subject according to one of the previous claims.

8. A method for the generation of an electrical stimulus suitable for eliciting the perception of mechanical and thermal stimuli by a subject, comprising the following phases: detecting mechanical and thermal data by using: at least one force sensor (12) comprising a plurality of operating areas,configured to detect the temporal trend of the intensity of a force (F) applied at one or more of such operating areas, the pressure (P) generated by such force at each one of said operating areas and the number (N) of operating areas on which the force (F) is applied; and• at least one temperature sensor (13), configured to detect the temperature trend over time (Tobj) at said operating areas; determining: a first algorithm associated to a first neurophysiological model capable of reproducing the behaviour of the subject’s mechanoreceptors, said first algorithm implementing a mathematical relationship between the temporal trend of a force (F) and the frequency of generation of a membrane potential of the aforementioned mechanoreceptors; a second algorithm associated with a second neurophysiological model capable of reproducing the behaviour of the subject’s nociceptors, said second algorithm implementing a mathematical relationship between the temporal trend of a force (F) and the frequency of generation of a membrane potential of the aforementioned nociceptors; and a third algorithm associated with a third neurophysiological model capable of reproducing the behaviour of the subject’s thermoreceptors, said third algorithm implementing a mathematical relationship between the trend of a temperature over time (Tobj) and the frequency of generation of a membrane potential of the aforementioned thermoreceptors; and the following characteristic parameters:• a maximum number (Nth), equal to the largest number of operating areas of said force sensor (12) thereat the application of the force (F) at a pain stimulation for the subject is detected,• a minimum pressure (p), equal to the smallest pressure value detected by said force sensor (12) at a pain stimulation for the subject when the force (F) is localized at a number of operating areas lower than said maximum number (Nth),• maximum amplitude (PAmax) of a current pulse that can be delivered to the subject, beyond which muscle contraction is generated;• first duration (PWmech) of a current pulse for which the subject perceives a non-painful mechano-tactile sensation;• second duration (PWnoc) of a current pulse for which the subject perceives a painful mechano-tactile sensation;• third duration (PWtemp) of a current pulse for which the subject perceives a non-painful thermal sensation; providing as input to said first algorithm data relating to the trend of the force (F) over time to obtain as output the frequency of generation of a first membrane potential associated with the mechanoreceptors, if the number of activated operating areas of said force sensor (12) is equal to or greater than said maximum number (Nth), or providing as input to said second algorithm data relating to the trend of the force (F) over time to obtain as output the frequency of generation of a second membrane potential associated with the nociceptors, if the number of activated operating areas of said force sensor (12) is less than said maximum number (Nth) and the pressure detected at such activated operating areas is greater than or equal to the minimum pressure (p); providing as input to said third algorithm data relating to the temperature trend over time (Tobj) to obtain as output the frequency of generation of a third membrane potential associated with the thermoreceptors, providing as input to an electrical stimulator (3) said frequencies of generation of the first, second and third membrane potential; generating by said electrical stimulator (3): first square, biphasic and symmetrical electrical stimulation waves designed to elicit in a subject a non-painful mechano-tactile sensation, whose frequency (PF) is modulated by the frequency of generation of said first membrane potential, wherein the stimulation wave has a pulse amplitude (PA) equal to said maximum amplitude (PAmax) and a pulse duration (PW) equal to said first duration (PWmech); second square, biphasic and symmetrical electrical stimulation waves designedto elicit in a subject a painful mechano-tactile sensation, whose frequency (PF) is modulated by the frequency of generation of said second membrane potential, wherein the stimulation wave has a pulse amplitude (PA) equal to said maximum amplitude (PAmax) and a pulse duration (PW) equal to said second duration (PWnoc); third square, biphasic and symmetrical electrical stimulation waves designed to elicit in a subject a non-painful thermal sensation, whose frequency (PF) is modulated by the frequency of generation of said third membrane potential, wherein the stimulation wave has a pulse amplitude (PA) equal to said maximum amplitude (PAmax) and a pulse duration (PW) equal to said third duration (PWt em p).

Images