Methods and systems for the sensory substitution of smell in a subject

Abstract

The present disclosure relates to a method, and the associated system, for the sensory substitution of smell in a subject, the method comprising: detecting a volatile compound or mixture of volatile compounds (15) by means of an array of sensors to produce a plurality of digital measurement signals uniquely associated with the volatile compound or mixture of volatile compounds; generating, from the plurality of digital measurement signals, a multimodal stimulation instruction comprising at least a first stimulation signal for the trigeminal system and at least a second stimulation signal for the trigeminal system having different physical and / or chemical properties from the first stimulation signal; stimulating the trigeminal system of the subject by applying each of the first and second stimulation signals by means of a plurality of actuators.

Description

[0001] “Processes and systems for sensory substitution of smell in a subject”

[0002] Technical field of the invention

[0003] The present disclosure relates to methods and systems for sensory substitution of smell in a subject. More specifically, the present disclosure applies to the field of health and relates to methods and systems for sensory substitution of smell by stimulation of the trigeminal system.

[0004] State of the art

[0005] Olfactory deficits have a direct impact on people's quality of life; they reduce the pleasure of eating, influence the relationship with food, increase the risk of domestic accidents, significantly affect social life and can be a source of depression. Such a phenomenon is not uncommon since olfactory deficits affect an average of 20% of the population in Europe, depending on the type of deficit, namely hyposmia (diminished olfactory performance compared to the average of a population) or anosmia (total absence of smell), depending on the causes (traumatic, congenital, viral, etc.), depending on age and depending on the pathologies; this represents several million people in Europe. Given that olfactory loss has been identified as a symptom of COVID-19, this prevalence could increase if olfactory loss persists after recovery. The economic and social costs of these sensory losses are high.For all these reasons, restoring olfaction – completely or even partially – is a challenge for science and society.

[0006] Outside of the field of olfaction, current paradigms for restoring sensory functions focus primarily on stimulating defective sensory systems. For example, cochlear implants (based on direct electrical stimulation of the cochlea) allow hearing-impaired patients to achieve speech perception. For the vestibular system (balance), vestibular implants are inspired by the principle of the cochlear implant, and the restoration of vestibular function is promising. For vision, there are devices using electrical stimulation of anatomical sites of the visual pathway with restoration of limited visual perception.These technical solutions in the fields of vision, hearing and sense of balance are not suitable for smell because it is a much more complex chemosensory modality in which the stimulus (chemistry) - perception transfer function is not or only slightly described mathematically.

[0007] In the field of olfaction, current paradigms used to restore odor perception include medication, surgery, and training using repeated odor exposure protocols. See T. Hummel, et al. [Ref. 1]. Medical treatments with nasal steroid sprays and / or antihistamines are used to improve olfaction in patients with inflammatory nasal conditions. In addition, surgical treatments such as polypectomy have shown effects on the recovery of olfactory perception. Finally, repeated odor exposure may be a promising tool, but its effectiveness has not been demonstrated in all patients. In particular, these therapies have different effects depending on the etiologies.

[0008] A series of studies have also shown that electrical stimulation of central olfactory areas such as the olfactory bulb or the orbitofrontal cortex can induce an olfactory sensation in the electrically stimulated person. US Patent 9517342 [Ref. 2] describes an olfactory implant system configured to mimic a subject's sense of smell in such a way as to enable the identification of one or more odors. The system comprises a sensor array, a processor, a transmitter, a receiver-stimulator, and an array of implantable electrodes. Such an olfactory implant system generates "odor maps," also called "olfactory signatures," by detecting odorant molecules using an array of chemical sensors, and then transmits variable spatiotemporal electrical stimulations to electrodes positioned at different locations in the olfactory bulb or another part of the olfactory cortex.In response to these varying electrical stimuli, different forms of activity in the olfactory cortex are generated, mimicking a subject's sense of smell. Once trained, a subject should be able to correctly detect or identify one or more odors.

[0009] Implanting electrodes into regions of the olfactory bulb will likely evoke olfactory-like perceptions in patients (olfactory qualities, hedonic qualities, intensity), allowing them to regain odor perception. However, this is a major neurosurgery procedure involving opening the skull and a risk of acute and long-term complications. Although surgical progress is increasingly convincing and significant, in the long term, there is concern that the cost of the surgical procedure will be greater than the desired benefit (recovery of olfactory perception, improved quality of life).

[0010] Recently, a system has been proposed by Brooks et al [Ref. 3] which electrically stimulates the trigeminal nerve, which makes it possible to generate sensory perceptions such as to perceive the origin or intensity of an odor, while avoiding the need to implant electrodes in the olfactory bulb.

[0011] Specifically, the system described in [ref. 3] includes a device that is shaped like a nose ring and can be inserted into the user's nose. The device is configured to electrically stimulate the trigeminal nerve (5th cranial nerve). The trigeminal nerve is known for detecting mechanical or temperature stimuli. It is sensitive in particular to changes in pressure or temperature. The device is connected to external sensors (via Bluetooth communication) to be able to detect molecules in the environment and is configured to electrically stimulate the nasal septum in order to access the trigeminal nerve and create lateralized sensations. Thus, in a first study conducted on 10 individuals, the authors present electrical stimulations that vary in intensity and spatiality. The results suggest that participants can localize the location of the stimulation (right / left).According to the authors, the device thus described allows users to feel the intensity (low, medium, high) and the origin of an odor (right / left) without prior training.

[0012] Although trigeminal nerve stimulation is a very interesting alternative to a system for simulating an intrusive sense of smell as described in [Ref. 2], the system described in [Ref. 3] remains limited in the possibilities it offers to provide a subject with hyposmia or anosmia with a sensory substitution of smell. Indeed, it is not possible for a subject to discriminate between different odors or categories of odors, the system described in [Ref. 3] being limited to a quantitative (intensity) and spatial (left / right) detection of a volatile organic compound.

[0013] The present description proposes a system of sensory substitution of the sense of smell by stimulation of the trigeminal system which makes it possible in particular to discriminate between different odors or categories of odors.

[0014] Summary of the invention

[0015] In this description, the term "comprise" means the same as "include", "contain", and is inclusive or open and does not exclude other elements not described or shown. Furthermore, in this description, the term "approximately" or "substantially" means the same as "having a margin less than and / or more than 10%, for example 5%", of the respective value.

[0016] The present description relates, according to a first aspect, to a method for sensory substitution of the sense of smell in a subject, comprising: the detection of a volatile compound or a mixture of volatile compounds by means of a network of sensors to produce a plurality of digital measurement signals uniquely associated with said volatile compound or mixture of volatile compounds; the generation, from said plurality of digital measurement signals, of a multimodal stimulation setpoint, the multimodal stimulation setpoint comprising at least a first trigeminal system stimulation signal and at least a second trigeminal system stimulation signal, of a physical and / or chemical nature different from that of the first stimulation signal; the stimulation of the subject's trigeminal system by applying each of said at least a first and second stimulation signals by means of a plurality of actuators.

[0017] In this description, a volatile compound is generally understood to be a chemical compound that is in a gaseous state in the atmosphere; it may notably be a volatile organic compound, or VOC, which comprises at least one carbon atom and one hydrogen atom. An odorous volatile compound, or odorous mixture of volatile compounds, is a compound, or mixture of compounds, known to generate an olfactory perception.

[0018] In this description, we will speak of "odor detection" for the detection by a subject of such an odorous volatile compound (or mixture of odorous volatile compounds), and of "odor category detection" for the detection of olfactory qualities associated with odorous volatile compounds or mixtures of odorous volatile compounds.

[0019] The trigeminal system is understood as the set of nervous structures that receive information directly or indirectly from the 3 branches of the trigeminal nerve. The trigeminal nerve is the fifth cranial nerve; it provides sensory innervation to the regions of the head, nose, mouth, and face. It is divided into three branches: the ophthalmic branch, the maxillary branch, and the mandibular branch, connected to the brain and the central nervous system by different nerve relays.

[0020] The physical and / or chemical nature of each of said first and second stimulation signals is, for example, and in a non-limiting manner, one or a combination of the following stimulation natures, or equivalently belongs to a class or a combination of classes of physical and / or chemical nature among the following: electrical, thermal, mechanical, in particular acoustic, electromagnetic, in particular optical, chemical, electrochemical. As is apparent from the description, the nature or combination of stimulation nature is more particularly distinct between said first and second stimulation signals. Equivalently, said class or combination of classes is preferably distinct between said first and second stimulation signals. This is then referred to as multimodal stimulation, for example bimodal stimulation.This is therefore distinguished from a system implementing simple electrical stimulation, by varying parameters such as intensity or phase.

[0021] According to one example, the first stimulation signal and the at least one second stimulation signal belong to classes of physical and / or chemical nature that are different from each other. According to one example, one of said first and second stimulation signals comprises or is an electrical stimulation.

[0022] According to one example, one of said first and second stimulation signals comprises or is mechanical stimulation.

[0023] In one example, the first stimulation signal comprises or is electrical stimulation and the second stimulation signal comprises or is mechanical stimulation.

[0024] The multimodal instruction may include a third stimulation signal of the trigeminal system. This is then referred to as trimodal stimulation. Said nature or combination of these types of stimulation is preferably distinct between each stimulation signal. According to one example, the third stimulation signal includes or is a thermal stimulation. The method that is the subject of the present description makes it possible, thanks to multimodal stimulation, to confer on the subject sensory perceptions that depend on the volatile compound or mixture of volatile compounds detected. The sensory perceptions, for example, “cool”, “hot”, “irritating”, “tingling”, “itching”, etc. make it possible to confer on the subject information on the volatile compound itself or information on an olfactory quality linked to the volatile compound, for example “pleasant”, “unpleasant”, “fruity”, “floral”, “woody”, “edible” etc.The trigeminal sensory perceptions thus induced replace the olfactory perceptions normally generated by means of the olfactory system to allow the subject to distinguish different volatile compounds or mixtures of volatile compounds, or to distinguish olfactory qualities linked to these volatile compounds. A perceptual discrimination of odors or categories of odors is then possible, which was not possible with the system described in [Ref. 3].

[0025] According to one or more exemplary embodiments, said at least one first signal of the first stimulation and / or said at least one second signal of the first stimulation is variable as a function of time. Thus, said at least one first signal and / or said at least one second signal may comprise one or more pulses, with one or more pulse repetition frequencies and / or one or more pulse durations. By configuring the pulse frequency and / or duration, for example, a greater variety of multimodal stimulations may be introduced. It may also be possible to provide the subject with a greater or lesser intensity of sensory perception, allowing quantitative detection of sensory perceptions by the subject.From the above, it is understood that said at least one first signal and / or said at least one second signal may comprise one or more pulses, having one or more pulse repetition frequencies and / or one or more pulse durations and / or a non-zero and non-constant intensity over time, over at least a portion of said stimulation signal and preferably over at least 50%, preferably 70%, preferably 90%, of the duration of the stimulation signal. In particular, the processing unit and the plurality of actuators may be configured to implement these characteristics.

[0026] According to one or more exemplary embodiments, the stimulation of the trigeminal system of the subject is carried out simultaneously by said at least one first stimulation signal and said at least one second first stimulation signal.

[0027] According to one or more exemplary embodiments, the stimulation of the trigeminal system of the subject is carried out in a first step by said at least one first stimulation signal and in a second step, by said at least one second first stimulation signal. As is clear from the description, the first step is distinct from the second step.

[0028] By providing simultaneous or temporally shifted stimulations of the trigeminal system, different perceptions are also generated for the subject. In particular, the intensity and / or quality of perception may be felt differently.

[0029] According to one or more exemplary embodiments, the generation of the multimodal stimulation instruction comprises the application of a transformation law. Such a transformation law may be pre-established for a set of subjects or may be determined specifically for a given subject. Such a transformation law may also be pre-established and then scalable, for example it may be adapted for each subject.

[0030] According to one or more exemplary embodiments, the generation of the multimodal stimulation instruction comprises a step of preprocessing said plurality of digital measurement signals to generate a plurality of preprocessed digital signals. In the examples where the generation of the multimodal stimulation instruction comprises the application of a transformation law, said transformation law can then be applied to said plurality of preprocessed digital signals.

[0031] According to one or more exemplary embodiments, the preprocessing step comprises a projection step into a P-dimensional space, where P < M, with M the number of digital measurement signals of said plurality of digital measurement signals and P greater than or equal to 2. For example, P is between 2 and 4. A projection into a lower-dimensional space then makes it easier to generate the multimodal setpoint, for example by applying a transformation law. A projection step can be carried out by any known mathematical method, for example, but not exclusively, by a method comprising a principal component analysis. This is particularly advantageous for the differentiation of a large number of VOCs and the associated olfactory qualities, since many digital measurement signals can be obtained for each VOC or mixture of VOCs. The signature to be analyzed then becomes simpler to associate with the appropriate stimulation.Synergistically with complex multimodal stimulation generation, using two stimulations of different classes or combinations of classes as indicated above, a subject's differentiation of a large number of VOCs and their olfactory qualities can be improved.

[0032] According to one or more exemplary embodiments, the sensors of the sensor network comprise receivers associated with transducers configured to deliver digital signals from physical quantities emitted by the receivers in contact with a volatile compound or mixture of volatile compounds. Such sensors are, for example, the sensors of a detection device known as an “electronic nose”; an electronic nose, known in the state, is a device for detecting and analyzing volatile compounds or mixtures of volatile compounds, and more particularly odorous volatile compounds. Electronic noses are, for example, described in the journal article [Ref. 4], See also published patent application FR 3120445 [Ref. 5], published patent application WO2018 / 158458 [Ref. 6] or published patent application W02022 / 053690 [Ref.7], Such sensors have the advantage of being non-specific, i.e. they are not configured for the detection of a single volatile compound. Such sensors provide for a volatile compound or a mixture of volatile compounds, a unique olfactory signature, i.e. a vector of given dimension, specific to the volatile compound or mixture of volatile compounds.

[0033] The multimodal stimulation instruction is then determined from said olfactory signature, for example, but not necessarily, by means of a transformation law.

[0034] According to one or more exemplary embodiments, the method according to the first aspect further comprises a learning step allowing the subject to associate the multimodal stimulation with said volatile compound or mixture of volatile compounds or with an olfactory quality associated with said volatile compound or mixture of volatile compounds.

[0035] The present description relates, according to a second aspect, to a system for sensory substitution of the sense of smell in a subject, comprising: a network of sensors configured for the detection of a volatile compound or a mixture of volatile compounds in order to produce a plurality of digital measurement signals uniquely associated with said volatile compound or the mixture of volatile compounds; a processing unit configured to generate, from said plurality of digital measurement signals, a multimodal stimulation setpoint comprising at least a first stimulation signal of the trigeminal system and at least a second stimulation signal of the trigeminal system, of a physical and / or chemical nature different from that of the first stimulation signal; a plurality of actuators configured to stimulate the trigeminal system of the subject with said at least a first stimulation signal and with said at least a second stimulation signal.According to one or more exemplary embodiments, the sensors of the sensor network comprise receivers associated with transducers configured to deliver digital signals from physical quantities emitted by the receivers in contact with a volatile compound. Such sensors are advantageous because they are non-specific; they are used in particular in electronic noses known from the state of the art.

[0036] However, other types of sensors known to those skilled in the art may be used, which are specific to certain volatile compounds. These are sensors configured for the detection of a specific volatile compound, for example a gas; these are, for example, sensors configured for the detection of carbon monoxide, carbon dioxide, methane, etc.

[0037] According to one or more exemplary embodiments, the sensor network comprises at least 2 sensors, for example between 2 and approximately 300 sensors, advantageously between 2 and approximately 150 sensors, advantageously between approximately 30 and approximately 150 sensors.

[0038] According to one example, the sensor network comprises at least 5, and preferably at least 10, and more preferably at least 15 physicochemically different sensors. More particularly, the receptors may be physicochemically different between these sensors. By "physicochemically different" is meant that the sensors, where applicable the receptors, have different physicochemical properties, so that their interaction with a volatile organic compound differs between them. Thus, a larger number of VOCs can be differentiated by the system. A larger number of different data and sensor signatures can be obtained. Synergistically with a larger range of sensory perceptions that can be generated by the system and in the method, the differentiation by the subject of a large number of VOCs and their olfactory qualities can be improved.Note that a type of sensor can be present in several copies, for example in at least two replicas, or equivalently in at least two copies, on the sensor network.

[0039] According to one or more exemplary embodiments, said at least one first actuator and said at least one second actuator are chosen from: a thermal actuator, for example a thermoelectric actuator (see for example [Ref. 8]), an electric actuator (see for example [Ref. 9]), a mechanical actuator, for example an electromechanical actuator, for example a piezoelectric transducer (see for example [Ref. 10]), a chemical actuator, comprising for example a piezoelectric micro-dispenser (for example a piezoelectric micro-dispenser or “Piezoelectric Micro-Dispensers”, from the company Physik Instrumente©). Of course, several actuators may be integrated into a single multimodal device.

[0040] According to one or more exemplary embodiments, said at least one first stimulation signal and / or said at least one second stimulation signal is variable as a function of time.

[0041] According to one or more exemplary embodiments, the generation of the multimodal stimulation instruction by the processing unit comprises a projection step into a P-dimensional space, where P is greater than or equal to 2 and strictly less than the number of digital measurement signals of said plurality of digital measurement signals.

[0042] In one example, the actuators are not implanted.

[0043] Brief description of the figures

[0044] Other advantages and characteristics of the technique presented above will appear on reading the detailed description below, made with reference to the figures in which:

[0045] Fig. 1A is a figure schematically illustrating an implementation of an example of a smell sensory substitution system according to the present disclosure, in a subject;

[0046] Fig. 1 B, a diagram illustrating the trigeminal system in humans;

[0047] Fig. 2 is a diagram showing the various elements of an example of a sensory substitution system for smell according to the present description;

[0048] Fig. 3A, a figure illustrating a plurality of measurement signals determined as a function of time, in ua;

[0049] Fig. 3B is a figure illustrating an odor map showing two odor signatures of two volatile compounds measured by a sensor array in an example of an odor sensory substitution system according to the present disclosure;

[0050] Fig. 4, a diagram illustrating steps for generating a multimodal stimulation instruction in a method of sensory substitution of smell, according to an exemplary embodiment;

[0051] Fig. 5, diagrams illustrating examples of sensory perceptions, associated with olfactory qualities corresponding to different volatile compounds, in an implementation of an example of a method for sensory substitution of smell in a subject according to the present description; Fig. 6, an example of a transformation matrix used in a method for sensory substitution of smell in a subject according to the present description;

[0052] Fig. 7 is a diagram illustrating an example of a sensory odor substitution system according to the present disclosure implemented on a subject;

[0053] Fig. 8A, a diagram illustrating an example of the installation of an actuator for implementing a method according to the present description;

[0054] Fig. 8B is a schematic showing in more detail the actuator illustrated in Fig. 8A;

[0055] Fig. 8C is a diagram illustrating an example of an actuator for a subject's sense of smell sensory substitution system according to the present disclosure;

[0056] Fig. 9, a graph illustrating the effect of electrical stimulation intensity on a subject's perception;

[0057] Fig. 10 is a graph illustrating subjects' perception associated with unimodal, bimodal, and trimodal stimuli.

[0058] In the various embodiments which will be described with reference to the figures, similar or identical elements bear the same references.

[0059] Detailed description of the invention

[0060] In the following detailed description, only certain embodiments are described in detail to ensure clarity of the disclosure, but these examples are not intended to limit the general scope of the principles emerging from this description.

[0061] The various embodiments and aspects described in the present description may be combined or simplified in multiple ways. In particular, the steps of the various methods may be repeated, interchanged, executed in parallel, unless otherwise specified.

[0062] Fig. 1A schematically illustrates an implementation of an example of a sensory substitution system 200 of the sense of smell according to the present disclosure, in a subject 10 and Fig. 1B represents a diagram illustrating the trigeminal system in humans.

[0063] As illustrated in Fig. 1A, the method for sensory substitution of the sense of smell according to the present description comprises the detection of a volatile compound or a mixture of volatile compounds 15 by means of a sensor network of a system 200 for sensory substitution of the sense of smell which will be described in more detail later by means of Fig. 2. The sensor network produces, for a volatile compound or mixture of volatile compounds 15, a plurality of digital measurement signals from which a multimodal stimulation setpoint is generated comprising at least a first stimulation signal of the trigeminal system and at least a second stimulation signal of the trigeminal system, of a physical and / or chemical nature different from that of the first stimulation signal. The trigeminal system of the subject is then stimulated by actuators of the system 200 which allow the application of the stimulation signals.This results in subject 10 having sensory perceptions associated with olfactory qualities 18.

[0064] As illustrated in Fig. 1 B, the trigeminal system 11 is the set of nerve structures that receive information directly or indirectly from the 3 branches of the trigeminal nerve. The trigeminal nerve is the fifth cranial nerve; it provides sensory innervation to the regions of the head, nose, mouth, and face. It is divided into three branches: the ophthalmic branch, the maxillary branch, and the mandibular branch, connected to the brain and central nervous system by nerve relays at the brainstem. The trigeminal system is involved in a diverse range of sensory functions (e.g., pain, irritation, sensations of hot and cold, tingling, etc.).The trigeminal system is an integral part of the chemosensory system that is activated, directly and indirectly, by odorants, i.e. volatile compounds capable of interacting with the receptors of the olfactory system to generate an olfactory perception.

[0065] It appears that olfaction, which results from the excitation of the olfactory nerve 12 and allows us to perceive odors, and trigeminal sensory perceptions are closely linked, due to anatomical and functional connections in the nasal cavity and in the brain. For example, the olfactory perception or smell of a mint candy is generated by the olfactory system and its freshness is generated by the trigeminal system. These sensations of smell and freshness are integrated in the brain to allow us to recognize and appreciate the mint candy.

[0066] By means of the method which is the subject of the present description, a person who has partially or totally lost their sense of smell, and who is equipped with such a system for sensory substitution of the sense of smell which is the subject of the present description, will be able to detect volatile compounds, and more particularly but not exclusively, odorous volatile compounds, by applying stimulation signals to the trigeminal system. The patient may be trained in advance to enable them to associate volatile compounds or mixtures of volatile compounds detected by the system 200 with sensory perceptions which result from the application of stimulation signals to the trigeminal system.Discrimination, memorization, and recognition of these different sensory perceptions can then be carried out, thus giving the patient the possibility of making the link between a volatile compound or a mixture of volatile compounds, or more generally an olfactory quality associated with volatile compounds or mixtures of volatile compounds, and an activation of neurons linked to learning.

[0067] Fig. 2 schematically illustrates a diagram representing the various elements of an example of a system 200 for sensory substitution of smell according to the present description.

[0068] The olfactory sensory substitution system 200 includes a sensor array 210 configured to detect a volatile compound or mixture of volatile compounds.

[0069] The sensors of the sensor network 210 are for example sensors of an electronic nose, and are described for example in [Ref. 5], [Ref. 6] or [ref. 7]. The sensors of the sensor network comprise in these examples receivers associated with transducers configured to deliver digital signals from physical quantities emitted by the receivers in contact with a volatile compound. The sensors thus make it possible to “photograph” a volatile compound or a mixture of volatile compounds by generating a recognition print or olfactory signature associated with this volatile compound or mixture of volatile compounds.

[0070] A known detection technique for obtaining, in use, a recognition fingerprint is for example a technique based on Mach-Zehnder interferometry (better known by the acronym MZI for "Mach-Zehnder Interferometry"), and described for example in [Ref. 7]. This technique makes it possible to detect a local change in optical index which characterizes the interaction of an analyte present in a gaseous sample with each receptor of the electronic nose. More precisely, the analyte interacts by adsorption / desorption with the receptors located in several distinct sensitive sites of a functionalized measurement surface. A measurement signal associated with each of the sensitive sites is detected in real time, representative of a variation in a refractive index due to an interaction of the fluid sample with the sensitive site thanks to a detectable phase shift between a reference arm of the interferometer and a detection arm on which each sensitive site is arranged.

[0071] Of course, other techniques are known to obtain a recognition fingerprint of a volatile compound or mixture of volatile compounds. For example, techniques based on surface plasmon resonance imaging (better known by the acronym SPR for "Surface Plasmon Resonance"). More precisely, the receivers of the electronic nose are configured to measure any change in refractive index due to an interaction of the fluid sample with any sensitive site thanks to a plasmonic effect. A surface reflectivity is measured by means of a camera (the transducer) which converts the photons into digital signals. In another variant, the sensitive sites are for example arranged on a matrix of vibrating nano or micro electromechanical membranes (NEMS or MEMS) in order to measure variations in the resonance frequency of the membranes.

[0072] In use, the sensors of the sensor array 210 produce a plurality of digital measurement signals Sm(t) uniquely associated with a volatile compound or mixture of volatile compounds.

[0073] An example of a plurality of digital measurement signals Sm(t) is shown as an example in Fig. 3A.

[0074] The digital measurement signals Sm(t) are in this example optical signals representative of a temporal variation of the local refractive index due to the interactions of the analyte with the receptors.

[0075] The digital measurement signals Sm(t) are sent to a processing unit 220 (Fig. 2) in order to produce a multimodal stimulation instruction comprising at least a first trigeminal system stimulation signal and at least a second trigeminal system stimulation signal, of a physical and / or chemical nature different from that of the first stimulation signal.

[0076] A plurality of actuators 230 (Fig. 2) are configured to stimulate the trigeminal system of the subject with the at least one first stimulation signal and with the at least one second stimulation signal. Actuators will be described in more detail later and may include electrical, thermal, mechanical, chemical, etc. actuators.

[0077] As illustrated in Fig. 2, the processing unit may comprise, in exemplary embodiments, a preprocessing module 222 and a multimodal stimulation calculation module 223.

[0078] The preprocessing module 222 receives the digital measurement signals Sm(t) produced by the sensor network.

[0079] Steps for processing the digital measurement signals Sm(t) produced by the sensor network are for example described in [ref. 7]. In exemplary embodiments, the preprocessing module 222 may comprise calibration and / or drift correction algorithms for correcting the signals Sm(t) from the sensor network. The preprocessing module 222 may also comprise a normalization algorithm, for example to obtain for each plurality of digital measurement signals an olfactory signature as shown in Fig. 3B.

[0080] Examples of olfactory signatures 321, 322 are shown for illustrative purposes in Fig. 3B with a radar map type representation.

[0081] In Fig. 3B, the numbers on the circle represent the identification numbers of the sensors used. The points on the radar map show a normalized signal value for each identified sensor.

[0082] The preprocessing module 222 may also comprise an algorithm aimed at reducing the dimension of the olfactory signature. This is, for example, a projection algorithm in a P-dimensional space, P advantageously between 2 and 4.

[0083] Thus, each olfactory signature obtained from the digital measurement signals Sm(t) can be a raw signature, i.e. a vector whose M components are representative of the response intensities of the M sensitive sites of the sensor network, a normalized signature, i.e. a vector whose M components are those of the raw signature but normalized using a normalization operation as described for example in [Ref. 7], a simplified signature obtained from a principal component analysis of the raw or normalized signature, i.e. a simplified vector, the P <N composantes étant par exemple les projections de la signature brute ou normalisée dans le référentiel orthonormé défini par les P premiers axes principaux sur lesquels les projections dépassent un seuil prédéfini. Dans la suite de la description, on parlera simplement de signature olfactive pour la signature brute, la signature normalisée ou la signature réduite.

[0084] A calculation module 226 (Fig. 2) for multimodal stimulation makes it possible to calculate the multimodal stimulation setpoint, for example, from the olfactory signature. Such an olfactory signature is specific to a volatile compound or a combination of volatile compounds and can therefore be advantageously used to deduce a multimodal stimulation setpoint.

[0085] As illustrated in Fig. 4, a multimodal stimulation instruction 403 may for example, but not necessarily, be determined from the plurality of digital measurement signals by means of a transformation law, represented in this example by a transformation matrix 402. Of course, when a series of processing steps are applied to the plurality of digital measurement signals, as described above, the transformation law may be applied to the signals resulting from these processing steps.

[0086] As illustrated in Fig. 4, the multimodal stimulation instruction 403 may comprise several stimulation signals, in this example thermal, electrical, mechanical stimulation signals, which will be transmitted to corresponding actuators 231, 232, 233, for example a heating electrode, an electrode for emitting an electrical signal, an actuator generating a vibration.

[0087] By way of illustration, Fig. 5 shows diagrams illustrating examples of multimodal stimulation instructions associated with different olfactory perceptions, in an implementation of an example of a method for sensory substitution of smell in a subject according to the present description.

[0088] In this example, an electronic nose as described for example in [Ref. 4] is used to produce, for a given volatile compound, an olfactory signature.

[0089] Thus, in a first example, we seek to substitute a sensory perception in a subject who suffers from hyposmia or anosmia in connection with a first odorant 501, for example mint. The electronic nose makes it possible to establish an olfactory signature 503 of the compound R-Carvone referenced 502 in Fig. 5 and a major compound of mint. As explained previously, for example by means of a transformation law, we establish, from the olfactory signature, a multimodal stimulation instruction making it possible to generate sensory perceptions 504 in the subject, in this example “warm” and “little irritating”. A subject may for example associate such sensory perceptions with an olfactory quality 505 “pleasant and edible”.

[0090] In a second example, we seek to substitute a sensory perception in the subject of a second odorant 511, for example jasmine. The electronic nose makes it possible to establish an olfactory signature 513 of the benzyl acetate compound referenced 512 in Fig. 5 and a compound present in jasmine. As explained previously, we establish, from the olfactory signature, a multimodal stimulation instruction making it possible to generate sensory perceptions 514 in the subject, in this example “hot” and “moderately irritating”. A subject could for example associate such sensory perceptions with an olfactory quality 515 “pleasant and inedible”.

[0091] In a third example, we seek to substitute a sensory perception in the subject of a third odorant 521, in this example the goat. The electronic nose makes it possible to establish an olfactory signature 523 of the hexanoic acid compound 522, a characteristic compound of this odorant. As explained previously, we establish, from the olfactory signature, a multimodal stimulation instruction making it possible to generate sensory perceptions 524 in the subject “moderately hot” and “irritating”. A subject could for example associate such sensory perceptions with an olfactory quality 515 “unpleasant and inedible”.

[0092] In exemplary embodiments, a subject training step may allow the subject to associate each of the sensory perceptions 504, 514, 524, with the olfactory qualities 505, 515, 525, or even with the odorants themselves.

[0093] Furthermore, in exemplary embodiments, steps for defining multimodal stimulation may be provided to better correspond to specific sensory perceptions for a subject. Thus, in the case of using a transformation law, this may be adapted to the subject.

[0094] Still for illustration purposes, Fig. 6 represents an example of a transformation matrix that could be used in a method of sensory substitution of smell in a subject, according to the present description. The transformation law can for example associate digital measurement signals uniquely associated with a VOC or mixture of VOCs, with stimulation signals chosen from classes of different physical or chemical nature, to generate the multimodal stimulation setpoint. The transformation law can associate digital measurement signals uniquely associated with a VOC or mixture of VOCs with a degree of intensity of a stimulation signal and / or a defined variation of the stimulation signal, and in particular according to the modalities set out in the description.

[0095] In this example, we seek to apply two stimulation signals to the trigeminal system. The first stimulation signal is a thermal excitation signal and the second stimulation signal is an electrical excitation signal comprising electrical pulses generated with a variable frequency f.

[0096] As explained previously, an olfactory signature of a given volatile compound or mixture of volatile compounds, obtained by means of an electronic nose as described previously, contains M-dimensional information corresponding to the M receptors of the electronic nose. For example, M = 64.

[0097] In this exemplary embodiment, the processing unit (220, Fig. 2) is configured to reduce the dimension of the signature, by projection into a P-dimensional space, for example P = 2. A projection operation into a 2D space is carried out in a known manner, for example by means of a principal component analysis. Other dimensionality reductions are possible, for example by means of machine learning for example.

[0098] Thus, a signature with M=64 points becomes a vector with two coordinates X and Y calculated by the model. The model imposes for example that X and Y are between 0 and 1.

[0099] In an exemplary embodiment, to determine the electrical excitation signal, the following steps can be carried out.

[0100] During a calibration step (optional), carried out on the subject, we detect a minimum amplitude that is required to trigger a sensory perception in the person. In this example, we will propose an electrical stimulation with fixed amplitude, that is to say that there is no variation in the amplitude of the stimulation depending on the value of X. During this same calibration step, we define with the person a minimum frequency detected by the person (called Fstim_mini), at the stimulation amplitude defined during the calibration.

[0101] In this example, we choose to stimulate at the frequency Fstim_mini with the amplitude fixed for X = 0. We also choose to sweep the stimulation frequency up to a maximum value equal to 2 times the minimum stimulation frequency, the maximum value obtained for X = 1.

[0102] With the assumptions described above, for X = 0, we want a frequency f = Fstim_mini and for X = 1, we want a frequency f = 2x Fstim_mini.

[0103] In the general case, we therefore obtain a frequency f: f(X) = (X +1)* Fstim_mini which corresponds to the coefficient 601 of the matrix illustrated in Fig. 6.

[0104] In an exemplary embodiment, to determine the thermal excitation signal, the following steps can be carried out.

[0105] During a calibration step (optional), carried out on the person, a minimum temperature is detected which is required to trigger a sensory perception in the person.

[0106] In this example, we will propose a fixed duration of thermal stimulation (for example 1 second). In other words, there is no variation in the duration during which the heating is applied depending on the value of Y.

[0107] In this example, we apply, in operation, a thermal stimulation, noted SDthermique, for the predetermined fixed duration. We choose that the value of the temperature T is equal to SDthermique for Y = 0. We choose that the temperature T is equal to SDthermique +5°C for Y = 1, for safety reasons (we do not want to exceed a certain threshold). In the general case, we therefore obtain a temperature T:

[0108] T(Y) = 5*Y + SDthermique, which corresponds to the coefficient 602 of the matrix illustrated in Fig. 6.

[0109] It is thus possible, thanks to the modal stimulation instruction determined from a plurality of digital measurement signals, to offer the subject a sensory substitution of smell that is much more refined than that proposed in the state of the art, since it becomes possible to generate in the subject a plurality of sensory perceptions associated with the stimulation signals of the trigeminal system, as has been explained by means of the examples above.

[0110] Fig. 7 represents a diagram illustrating an example of a system 200 for sensory substitution of smell according to the present description, implemented on a subject 10.

[0111] The olfactory sensory substitution system 200 comprises a support 280 configured to be worn by the subject, for example by means of branches which are fixed behind the ears. The sensor network 210 configured for the detection of a volatile compound or a mixture of volatile compounds is, in this example, arranged on the support, close to the subject's nose. The processing unit configured to produce a multimodal stimulation instruction from the plurality of measurement signals produced by the sensors, is for example located remotely, that is to say it is not worn by the subject. The processing unit communicates with the sensor network for example by means of a Bluetooth module 260, for example a module implementing a short-range, low-power wireless Bluetooth technology or BLE, according to the acronym "Bluetooth Low Energy".The Bluetooth module 260 is connected to the sensor network 210 by electrical wires 250 which run in the branches of the support 280 for example.

[0112] The olfactory sensory substitution system 200 also includes in this example a plurality of actuators 230 configured to stimulate the trigeminal system of the subject with trigeminal system stimulation signals established by the processing unit and transmitted by means of the Bluetooth module 260 and electrical wires.

[0113] A battery for operating the system 200 is, for example, arranged in one of the side branches of the support 280.

[0114] Of course, it will be possible to envisage, in exemplary embodiments, a completely miniaturized system 200 in which all of the elements can be arranged in the subject's nose. Fig. 8A illustrates in more detail an example of the installation of an actuator for implementing a method according to the present description and Fig. 8B is a diagram showing an example of an actuator as illustrated in Fig. 8A.

[0115] In the example of Fig. 8A, multimodal actuators 830 are arranged on the edges of a clip 835 configured to be positioned in the nose, so as to pinch the nasal wall. The multimodal actuators 830 are then in contact with the outer inner walls of the nose. The clip can be held on the nasal wall for example by magnets (not shown).

[0116] As seen in Fig. 8B, the multimodal actuators 830 may be connected to wires 850, which run through the branches of the support 280 (Fig. 7).

[0117] Fig. 8C is a diagram illustrating an example of a multimodal actuator 830 for a smell sensory substitution system according to the present disclosure.

[0118] The multimodal actuator 830 comprises in this example a bimodal electrode (electrical and thermal) with heating elements 831 and electrodes 832 for the application of an electrical signal.

[0119] Several specific examples of implementation are now described for illustrative purposes.

[0120] Example of a connection between an artificial sensor system and a multimodal stimulator.

[0121] According to this example, the system combines an electronic nose and a multimodal stimulator that integrates electrical, mechanical, and thermal stimulation means. The system makes it possible to present a VOC or a mixture of VOCs to the sensors 210 of the electronic nose and to convert the chemical information into digital information in the form of digital measurement signals, as illustrated in Figure 2. These signals may in particular form a digital signature associated with the VOC or mixture of VOCs in a unique way, as illustrated in Figure 3B. Based on these measurement signals, information is sent to the stimulator in order to offer the human subject unimodal (electrical or thermal or mechanical), bimodal (electrical-mechanical, electrical-thermal, mechanical-thermal), or trimodal (electrical-mechanical-thermal) stimulation. The stimulation is sent into the nasal cavity and perceived by the human subject, as illustrated in Figure 1.

[0122] Example of a test of the effect of electrical stimulation amplitude on volunteers.

[0123] An experiment conducted on 69 human subjects in which a very low intensity electrical stimulation (903: around the detection threshold) and a low intensity electrical stimulation (904: twice the threshold) are delivered into the subject's nasal cavity (site: anterior septum). The results are illustrated in Figure 9 with the 902 counts as a function of the applied intensity 901. Different perceptions are induced by patterns of the same electrical stimulation presented at different amplitudes: the stronger stimulation induces a higher perceived intensity.

[0124] Example of a test with unimodal, bimodal and trimodal stimulations on volunteers.

[0125] In an experiment conducted on 6 volunteer participants, we stimulated their nasal cavity (anterior septum) using 7 different types of stimulation: unimodal (electrical E or thermal T or mechanical M), bimodal (electrical-mechanical EM, electrical-thermal TE, mechanical-thermal TM) and trimodal (electrical-thermal-mechanical TEM). Participants had to describe their perception in a multidimensional way according to the dimensions of tingling, vibration and heat. The overall hedonic value of the stimulation had to be evaluated as well. Figure 10 illustrates the results obtained with the 902 sample according to the type of perception (tingling: not hatched, vibration: vertical hatching, heat: horizontal hatching). The results showed that unimodal stimulations only evoked sensations related to their nature: tingling for electric, vibration for mechanical and heat for thermal.On the other hand, when these stimulations are combined, the perceptions are more complex: prickly and vibrating sensations for electrical-mechanical, prickly and hot for electrical-thermal, and vibrating and hot for mechanical-thermal. Finally, trimodal stimulation induced all 3 types of sensations (prickly, vibrating and hot). The hedonic valence of the stimulations was also different: when the stimulation included a thermal component, it was most often unpleasant, whereas mechanical stimulation alone (unimodal) was perceived as more pleasant.

[0126] Examples of stimuli for chemical, electrochemical and stimulation modes

[0127] For chemical stimulation, piezoelectric microdispensers, for example, can allow the localized deposition of one or more chemical species that can simulate cold (menthol), heat (capsaicin), stinging sensations (weak organic acid, CO2), or an alkaline sensation (diluted ammonia). Other types of microdispensers, other than piezoelectric, can be considered. It is preferable to take into account a fairly long afterglow in this case of chemical stimulation. These chemical stimuli can also be deposited in the nasal cavity using an air flow. For electromagnetic stimulation, an infrared diode can allow local heating of a tissue and has the advantage of having less inertia than a resistive heating system.Finally, for electrochemical stimulation, by choosing the material of the electrodes, we can carry out the hydrolysis of water and obtain gas bubbles which will have an effect close to mechanical stimulation.

[0128] Although described through a number of detailed exemplary embodiments, the systems and methods for sensory substitution of smell include various variations, modifications and improvements which will be apparent to those skilled in the art, it being understood that these various variations, modifications and improvements are within the scope of the invention, as defined by the following claims.

[0129] References

[0130] Ref. 1: T. Hummel, et al “Position paper on olfactory dysfunction”, Rhinology Supplement 25, 1 - 30 (2017)

[0131] Ref 2: US 9517342

[0132] Réf 3: J. Brooks et al, “Stereo-Smell via Electrical Trigeminal Stimulation”, Pre-Print of CHI (2021)

[0133] Réf. 4: Karakaya et al, “Electronic Nose and Its Applications: A Survey”, International Journal of Automation and Computing, 17(2), 179-209 (2020)

[0134] Réf. 5 : FR 3120445

[0135] Réf. 6 : WO2018 / 158458

[0136] Réf. 7: WC2022 / 053690

[0137] Réf. 8: Skinner et al, “Variation in thermally induced taste response across thermal tasters”, Physiology & behavior, 188, 67-78 (2018)

[0138] Réf. 9: Scheibe et al, “Investigation of the topographical differences in somatosensory sensitivity of the human nasal mucosa”, Rhinology, 50(3), 290-293 (2012)

[0139] Réf. 10: Channer et al, “Middle Ear Implants: Historical and futuristic perspective”, Journal of Otology, Vol.6 No.2, 10 - 18 (2011)

Claims

CLAIMS 1. System (200) for sensory substitution of smell in a subject (10), comprising: a sensor network (210) configured for the detection of a volatile compound or a mixture of volatile compounds in order to produce a plurality of digital measurement signals uniquely associated with said volatile compound or the mixture of volatile compounds; a processing unit (220) configured to generate, from said plurality of digital measurement signals, a multimodal stimulation setpoint comprising at least a first trigeminal system stimulation signal and at least a second trigeminal system stimulation signal;a plurality of actuators (230) configured to stimulate the trigeminal system of the subject with said at least one first stimulation signal and with said at least one second stimulation signal, characterized in that the at least one second stimulation signal of the trigeminal system is of a physical and / or chemical nature different from that of the at least one first stimulation signal, the physical and / or chemical nature of each of said first and second stimulation signals belongs to a class or a combination of classes of physical and / or chemical nature among the following: electrical, thermal, mechanical, electromagnetic, chemical, electrochemical, said class or combination of said classes being distinct between said first and second stimulation signals.; 2. Sensory substitution system for smell according to claim 1, in which the sensors (210) of the sensor network comprise receivers associated with transducers configured to deliver digital signals from physical quantities emitted by the receivers in contact with the volatile compound or mixture of volatile compounds.

3. A sensory odor substitution system according to any one of the preceding claims, wherein the physical and / or chemical nature of the at least one first stimulation signal and the at least one second stimulation signal stimulation is a physical and / or chemical nature belonging to a class of physical and / or chemical nature chosen from the following: electrical, thermal, mechanical, chemical, electrochemical, electromagnetic, the first stimulation signal and the at least one second stimulation signal belonging to different classes.

4. A sensory odor substitution system according to any preceding claim, wherein one of said first and second stimulation signals comprises or is electrical stimulation and the other of said first and second stimulation signals comprises or is mechanical stimulation.

5. A sensory odor substitution system according to any one of the preceding claims, wherein the processing unit (220) and the plurality of actuators (230) are configured such that said at least one first stimulation signal and / or said at least one second stimulation signal is variable as a function of time, said at least one first signal and / or said at least one second signal comprising one or more pulses having: one or more pulse repetition frequencies, and / or one or more pulse durations, and / or a non-zero intensity that is not constant over time, on at least a portion of said stimulation signal.

6. A sensory odor substitution system according to any preceding claim, wherein the plurality of actuators (230) are configured to stimulate the trigeminal system of the subject with said at least one first stimulation signal at a first time and with said at least one second stimulation signal at a second time.

7. Sensory substitution system for smell according to any one of the preceding claims, in which the generation of the multimodal stimulation instruction by the processing unit comprises the application of a transformation law.

8. Sensory substitution system for smell according to any one of the preceding claims, in which the generation of the multimodal stimulation instruction by the processing unit comprises a step of projection into a P-dimensional space, where P is greater than or equal to 2 and strictly less than the number (M) of digital measurement signals of said plurality of digital measurement signals.

9. Sensory substitution system for smell according to any one of the preceding claims, in which the sensor network comprises at least 5, and preferably at least 10, and more preferably at least 15 physicochemically different sensors.

10. A method of sensory substitution of smell in a subject, comprising: detecting a volatile compound or mixture of volatile compounds (15) by means of a sensor network (210) to produce a plurality of digital measurement signals (S m(t)) uniquely associated with said volatile compound or mixture of volatile compounds; generating, from said plurality of digital measurement signals, a multimodal stimulation setpoint, the multimodal stimulation setpoint comprising at least a first trigeminal system stimulation signal and at least a second trigeminal system stimulation signal, of a physical and / or chemical nature different from that of the first stimulation signal, and in which the physical and / or chemical nature of said first and second stimulation signals belongs to a class or a combination of classes of physical and / or chemical nature among the following: electrical, thermal, mechanical, electromagnetic, chemical, electrochemical, said class or combination of said classes being distinct between said first and second stimulation signals;stimulating the trigeminal system of the subject by applying each of said at least one first and second stimulation signals by means of a plurality of actuators (230).; 11. Method for sensory substitution of smell according to the preceding claim, in which said at least one first stimulation signal and / or said at least one second stimulation signal is variable as a function of time, said at least one first signal and / or said at least one second signal comprising one or more pulses having: one or more pulse repetition frequencies, and / or one or more pulse durations, and / or a non-zero and non-constant intensity over time, on at least part of said stimulation signal.

12. Method for sensory substitution of smell according to any one of the two preceding claims, in which said at least one first stimulation signal and said at least one second stimulation signal are applied in a time-shifted manner.

13. Method for sensory substitution of smell according to any one of the three preceding claims, in which the generation of the multimodal stimulation instruction comprises the application of a transformation law.

14. Method for sensory substitution of smell according to any one of the four preceding claims, in which the generation of the multimodal stimulation instruction comprises a step of projection into a P-dimensional space, where P is greater than or equal to 2 and strictly less than the number (M) of sensors in the sensor network.

15. A method of sensory substitution of smell according to any one of the five preceding claims, further comprising: a learning step allowing the subject to associate the multimodal stimulation with said volatile compound or mixture of volatile compounds or with an olfactory quality associated with said volatile compound or mixture of volatile compounds.

Images