Non-invasive method and system for characterising and certifying cognitive activities

1 . A method for characterising and certifying cognitive activities in a non-invasive manner by means of a characterisation and certification system, comprising:

a detection module comprising at least one gaseous component measuring element configured for generating at least one signal indicating a temporal evolution of at least one detected gaseous component;

a characterisation module configured for characterising the at least one signal generated by the detection module based on an individual or combined sequential structure of the temporal evolution thereof and for determining whether said at least one signal corresponds to a development of a specific cognitive activity; and

a machine learning algorithm that is trained with a method comprising acquiring the at least one detected gaseous component over time in the form of the at least one signal and classifying known activities associated with the acquired at least one signal as a result label, wherein the known activities are associated with one or more cognitive activities;

wherein the method comprises the following steps:

a) generating at least one signal indicating the temporal evolution of at least one gaseous component corresponding to the specific cognitive activity detected by means of the at least one measuring element of the detection module during a predetermined time period;

b) receiving the generated signal by means of the characterisation module;

c) characterising the generated signal based on the sequential structure of events in the temporal evolution thereof, by means of the characterisation module; and

d) determining whether the generated signal corresponds to the development of the specific cognitive activity based on a result of the characterisation;

wherein step (c) comprises identifying at least one temporal event in the signal, and wherein the determination of step (d) is performed based on the sequentiality of the identified temporal events; and

wherein steps (c) and (d) are performed using the trained machine learning algorithm, which uses the sequential structure of the temporal evolution of the cognitive activity;

wherein the method further comprises classifying the at least one signal into at least one subcategory of cognitive activities by means of the characterisation module; and

wherein the classification of the signal is performed by means of the machine learning algorithm previously trained with training signals which are classified into at least one predefined subcategory of cognitive activity, and wherein the result label is fed back into the machine learning algorithm to improve precision thereof.

2 . The method according to claim 1 , wherein the system comprises a conditioning module configured for conditioning the signal generated by the detection module; and wherein the method further comprises, between steps (a) and (b), the steps of receiving and conditioning the generated signal by means of the conditioning module, and wherein step (c) is performed on the generated and conditioned signal.

3 . The method according to claim 2 , wherein the step of conditioning the generated signal comprises filtering and/or sampling said generated signal maintaining the sequential structure of the temporal evolution of the cognitive activity.

4 . The method according to claim 1 , wherein the cognitive activity belongs to one of the following categories:

effective cognitive activity or cognitive activity that fulfils pre-established objectives; or

cognitive activity associated with an emotion which causes the release of substances detectable by the device; or

cognitive activity with a specific level of attention; or

cognitive activity typical of a work routine; or

cognitive activity typical of a school routine; or

cognitive activity typical of a leisure routine; or

cognitive activity typical of an examination or test; or

cognitive activity typical of a job interview; or

cognitive activity typical of a specific health condition of an individual; or

a combination of any of the above.

5 . The method according to claim 1 , wherein classification of the signal is performed based on the sequentiality of the identified temporal events.

6 . The method according to claim 1 , wherein the temporal events detected in the signal comprise maximums, minimums, slopes, the surpassing of thresholds, and/or sequential sets of the foregoing which define a specific temporal structure.

7 . The method according to claim 1 , wherein the at least one gaseous component measuring element is an olfactory sensor, said olfactory sensor being configured for detecting at least one substance emitted by an organism during a cognitive activity.

8 . The method according to claim 7 , wherein the olfactory sensor is of any of the following types: chemoresistive, chemocapacitive, potentiometric, gravimetric, optical, acoustic, thermal, polymer, amperometric, chromatographic, spectrometric, or field effect sensor.

9 . The method according to claim 1 , wherein:

the detection module further comprises at least one environmental condition detecting element for detecting environmental conditions;

the method further comprises a step of obtaining measurements of at least one magnitude by means of the at least one environmental condition detecting element and of identifying temporal events of the signal which are associated with said at least one magnitude; and

wherein said identified temporal events are used as additional context information during the step of characterising the signal.

10 . The method according to claim 1 , wherein:

the detection module further comprises at least one external event recording element;

the method further comprises a step of identifying temporal events of the signal which are associated with the presence of external events; and

wherein said identified temporal events are used as additional context information during the step of characterising the signal.

11 . The method according to claim 1 , wherein step (c) of the method comprises comparing the generated signal with at least one reference signal of the cognitive activity.

12 . The method according to claim 1 , wherein the steps of the method are repeated periodically, where the repetition period is a predefined value, in order to monitor the detected and/or classified cognitive activities.

13 . A non-invasive system for characterising and certifying cognitive activities, comprising:

a detection module comprising at least one gaseous component measuring element configured for generating at least one signal indicating a temporal evolution of at least one detected gaseous component;

a characterisation module configured for characterising the at least one signal generated by the detection module based on an individual or combined sequential structure of the temporal evolution thereof, for determining whether said at least one signal corresponds to a development of a specific cognitive activity;

a machine learning algorithm that is trained with a method comprising acquiring the at least one detected gaseous component over time in the form of the at least one signal and classifying known activities associated with the acquired at least one signal as a result label, wherein the known activities are associated with one or more cognitive activities;

wherein the characterisation module is configured to:

a) receive the generated signal by means of the characterisation module;

b) characterise the generated signal based on the sequential structure of events in the temporal evolution thereof, by means of the characterisation module; and

c) determine whether the generated signal corresponds to the development of a specific cognitive activity based on the result of the characterisation;

wherein step (b) comprises identifying at least one temporal event in the generated signal, and wherein the determination of step (c) is performed based on the sequentiality of the identified temporal events; and

wherein steps (b) and (c) are performed using the trained machine learning algorithm, which uses the sequential structure of the temporal evolution of the cognitive activity;

wherein the method further comprises classifying the at least one signal into at least one subcategory of cognitive activities by means of the characterisation module; and

wherein the classification of the signal is performed by means of the machine learning algorithm previously trained with training signals which are classified into at least one predefined subcategory of cognitive activity, and wherein the result label is fed back into the machine learning algorithm to improve precision thereof.

14 . The system according to claim 13 , further comprising a conditioning module configured for conditioning the signal originating from a specific cognitive activity generated by the detection module.

15 . The system according to claim 13 , wherein the characterisation module of the system is further configured for classifying the at least one signal into at least one subcategory of cognitive activities.

16 . The system according to claim 13 , comprising contextualization elements of cognitive activity:

at least one environmental condition detecting element for detecting environmental conditions; and/or

at least one external event recording element.

17 . The system according to claim 13 , wherein the at least one gaseous component measuring element is an olfactory sensor, said olfactory sensor being configured for detecting at least one substance emitted by an organism during a cognitive activity.

18 . The system according to claim 17 , wherein the olfactory sensor is of any of the following types: chemoresistive, chemocapacitive, potentiometric, gravimetric, optical, acoustic, thermal, polymer, amperometric, chromatographic, spectrometric, or field effect sensor.

19 . A system for characterising and certifying cognitive activities in a non-invasive manner by means of a characterisation and certification system, comprising:

a detection module comprising at least one gaseous component measuring element configured for generating at least one signal indicating a temporal evolution of at least one detected gaseous component;

a characterisation module configured for characterising the at least one signal generated by the detection module based on an individual or combined sequential structure of the temporal evolution thereof and for determining whether said at least one signal corresponds to a development of a specific cognitive activity; and

a machine learning algorithm that is trained with a method comprising acquiring the at least one detected gaseous component over time in the form of the at least one signal and classifying known activities associated with the acquired at least one signal as a result label, wherein the known activities are associated with one or more cognitive activities;

a data processing system comprising means to:

a) receive the generated signal by means of the characterisation module;

b) characterise the generated signal based on the sequential structure of events in the temporal evolution thereof, by means of the characterisation module; and

c) determine whether the generated signal corresponds to the development of a specific cognitive activity based on the result of the characterisation;

wherein step (b) comprises identifying at least one temporal event in the generated signal, and wherein the determination of step (c) is performed based on the sequentiality of the identified temporal events; and

wherein steps (b) and (c) are performed using the trained machine learning algorithm, which uses the sequential structure of the temporal evolution of the cognitive activity;

wherein the method further comprises classifying the at least one signal into at least one subcategory of cognitive activities by means of the characterisation module; and

wherein the classification of the signal is performed by means of the machine learning algorithm previously trained with training signals which are classified into at least one predefined subcategory of cognitive activity, and wherein the result label is fed back into the machine learning algorithm to improve precision thereof.

20 . A system for characterising and certifying cognitive activities in a non-invasive manner by means of a characterisation and certification system, comprising:

a detection module comprising at least one gaseous component measuring element configured for generating at least one signal indicating a temporal evolution of at least one detected gaseous component;

a characterisation module configured for characterising the at least one signal generated by the detection module based on an individual or combined sequential structure of the temporal evolution thereof and for determining whether said at least one signal corresponds to a development of a specific cognitive activity; and

a machine learning algorithm that is trained with a method comprising acquiring the at least one detected gaseous component over time in the form of the at least one signal and classifying known activities associated with the acquired at least one signal as a result label, wherein the known activities are associated with one or more cognitive activities;

a tangible, non-transitory computer readable medium storing a computer program comprising instructions which, when the program is run by a computer, causes the computer to:

a) receive the generated signal by means of the characterisation module;

b) characterise the generated signal based on the sequential structure of events in the temporal evolution thereof, by means of the characterisation module; and

c) determine whether the generated signal corresponds to the development of a specific cognitive activity based on the result of the characterisation;

wherein step (b) comprises identifying at least one temporal event in the signal, and wherein the determination of step (c) is performed based on the sequentiality of the identified temporal events; and

wherein steps (b) and (c) are performed using the trained machine learning algorithm, which uses the sequential structure of the temporal evolution of the cognitive activity;

wherein the method further comprises classifying the at least one signal into at least one subcategory of cognitive activities by means of the characterisation module; and

wherein the classification of the signal is performed by means of the machine learning algorithm previously trained with training signals which are classified into at least one predefined subcategory of cognitive activity, and wherein the result label is fed back into the machine learning algorithm to improve precision thereof.

21 . A system for characterising and certifying cognitive activities in a non-invasive manner by means of a characterisation and certification system, comprising:

a detection module comprising at least one gaseous component measuring element configured for generating at least one signal indicating a temporal evolution of at least one detected gaseous component;

a characterisation module configured for characterising the at least one signal generated by the detection module based on an individual or combined sequential structure of the temporal evolution thereof and for determining whether said at least one signal corresponds to a development of a specific cognitive activity; and

a machine learning algorithm that is trained with a method comprising acquiring the at least one detected gaseous component over time in the form of the at least one signal and classifying known activities associated with the acquired at least one signal as a result label, wherein the known activities are associated with one or more cognitive activities;

a tangible, non-transitory computer-readable medium comprising instructions which, when run by a computer, causes the computer to:

a) receive the generated signal by means of the characterisation module;

b) characterise the generated signal based on the sequential structure of events in the temporal evolution thereof, by means of the characterisation module; and

c) determine whether the generated signal corresponds to the development of a specific cognitive activity based on the result of the characterisation;

wherein step (b) comprises identifying at least one temporal event in the signal, and wherein the determination of step (c) is performed based on the sequentiality of the identified temporal events; and

wherein steps (b) and (c) are performed using the trained machine learning algorithm, which uses the sequential structure of the temporal evolution of the cognitive activity;

wherein the method further comprises classifying the at least one signal into at least one subcategory of cognitive activities by means of the characterisation module; and

wherein the classification of the signal is performed by means of the machine learning algorithm previously trained with training signals which are classified into at least one predefined subcategory of cognitive activity, and wherein the result label is fed back into the machine learning algorithm to improve precision thereof.