New PCT Application January 16, 2025 curetec GmbH C173377WO KAU/Fim Mild Coordinated Reset Neuromodulation by Entrainment 1. Field The present invention relates to an apparatus for desynchronizing neuronal activity. Also de- scribed herein is a method for desynchronizing neuronal activity. The present invention further- more relates to a computer program that, upon execution, causes a suitable apparatus to desyn- chronize neuronal activity as described herein. 2. Technical Background It is generally known that invasive as well as non-invasive stimulation may be used to reduce pathological, specifically pathologically synchronized, neuronal activity, which may result, inter alia, from Parkinson’s disease, depression, anxiety disorders and/or burn-out. An approach that has generally proven effective in desynchronizing neuronal activity is Coordi- nated Reset (CR) neuromodulation (cf. Tass, Biol Cybern 2003, 89: 81-88; Tass & Majtanik, Biol Cybern 2006, 94: 58-66, Hauptmann & Tass, BioSystems 2007, 89: 173-181; Tass et al., Physical Review E 2009, 80, 011902). CR neuromodulation comprises a consecutive, temporally equidis- tant delivery of stimuli through different stimulation units (e.g., electrodes of an implanted lead). The resulting stimulation intends to sequentially reset phases of different stimulated (sub-)populations of neurons and, hence, divide a population of neurons into phase-shifted sub- populations. This desynchronization ultimately causes an unlearning of both pathological neu- ronal synchrony and pathological synaptic connectivity by utilizing the brain’s synaptic plastic- ity. Every CR pattern is generally characterized by a cycle period T. The cycle period T may be de- fined as a (longest) time period within which each of the stimulation units is activated exactly once/provides exactly one stimulus. The cycle period may also be defined as a duration between two consecutive onset times at which a same stimulation unit is activated/a stimulus is provided by a same stimulation unit. A cycle period may hence potentially comprise one or more dura-
tions during which no stimulation unit is active/activated/provides any stimulus. The cycle pe- riod T is chosen such that its inverse, i.e., 1/T, corresponds to (is substantially the same as) a target frequency, i.e., a frequency to be targeted by the CR neuromodulation. At least in invasive approaches, the cycle period T is hence commonly chosen such that its inverse and/or the target frequency corresponds to (is substantially the same as) the frequency of the neuronal activity that is to be desynchronized, e.g., 6 Hz for Parkinson’s disease and/or tremor suppression, or 20 Hz for movement disorders and/or reduction of beta activity. Meanwhile, the individual stimuli themselves are usually brief and characterized by a high fre- quency, i.e., a frequency significantly higher than the frequency of the neuronal activity that is to be desynchronized. For example, in WO 2019/023467 A1 an individual stimulus of a CR pattern is characterized by a frequency of 250 Hz or 64 Hz. That said, in US 7,974,698 B2 (and DE 103 55652 A1 from the same family, cf. also Hauptmann & Tass, Biol Cybern 2006, 93: 463-470) it is criticized that such individual stimuli as typically employed in CR neuromodulation do not ac- count for the dynamics of the neuronal activity to be desynchronized. To address this, it is sug- gested to rather measure the neuronal activity to be desynchronized and to feedback a time-de- layed version thereof as individual stimuli. Such a delayed-feedback and/or closed-loop ap- proach is inherently complex, though. Where a user undergoing stimulation can distinctly perceive individual stimuli of a CR pattern, the relatively short cycle periods T and/or relatively high target frequencies mentioned above are unfavorable, though, as the rapid succession of stimuli originating from (provided by) differ- ent stimulation units may invoke a stressful impression with the user. This applies in particular to non-invasive, e.g., vibro-tactile or acoustic, approaches. The prior art addresses this by taking a compromise in that it chooses the cycle period T such that the frequency of the neuronal activity that is to be desynchronized corresponds to (is sub- stantially the same as) an integer multiple of the inverse of the cycle period T and/or the target frequency. For example, the cycle period T is often chosen such that the inverse of the cycle pe- riod T and/or the target frequency is about 1.5 Hz, as is disclosed, e.g., in WO 2019/023467 A1, assuming that the resulting stimulation will also affect frequencies of neuronal activity that are integer multiples of the inverse of the cycle period T and/or the target frequency, such as 6 Hz (fourfold multiple of 1.5 Hz) for Parkinson’s disease and/or tremor suppression.
However, in the exact same manner, various further, perhaps non-targeted, frequencies of neu- ronal activity will also be affected by the stimulation, e.g., also 3 Hz (twofold multiple of 1.5 Hz), 4.5 Hz (threefold multiple of 1.5 Hz), 7.5 Hz (fivefold multiple of 1.5 Hz), etc. As a result, the stimulation becomes somewhat arbitrary and thus ultimately less effective. Clearly, there exists a continuous need to further improve on CR neuromodulation, especially with respect to the issues just outlined. 3. Summary This need is addressed at least in part by an apparatus for desynchronizing neuronal activity ac- cording to independent claim 1. Such apparatus comprises a set of N≥2 stimulation units, with N being an integer, preferably N=4, N=5, N=6 or N=10. Each stimulation unit of the set is adapted to provide a respective stimulus to a user upon activation. The respective stimulus is synthesized by the apparatus based on at least one parameter, such that the respective stimulus is characterized by a repetition frequency that corresponds to a frequency of the neuronal activ- ity that is to be desynchronized. The apparatus further comprises means for individually activat- ing the simulation units of the set such that stimuli provided by different stimulation units are mutually phase-shifted by a fraction of 2π. Throughout this disclosure, any indication of a phase-shift is to be understood modulo 2π, i.e., any indication of a phase-shift presupposes that a modulo operation using 2π (as modulus) has been performed. In line therewith, throughout this disclosure, a fraction of 2π is to be under- stood as a real number between 0 and 2π. Preferably, the fraction of 2π comprises an integer multiple of 2π/N, as this may be particularly straightforward to implement and may yield an even distribution of phase-shifts across stimuli provided by different stimulation units. Furthermore, throughout this disclosure, a first frequency f1 may be understood to correspond to a second frequency f2 if 1/2⋅f2 ≤ f1 ≤ 2⋅f2 and/or 1/2⋅f1 ≤ f2 ≤ 2⋅f1, respectively, preferably if 3/4⋅f2 ≤ f1 ≤ 4/3⋅f2 and/or 3/4⋅f1 ≤ f2 ≤ 4/3⋅f1, respectively, most preferably if 7/8⋅f2 ≤ f1 ≤ 8/7⋅f2 and/or 7/8⋅f1 ≤ f2 ≤ 8/7⋅f1, respectively. Similarly, a first frequency f1 may be understood to be substantially the same as a second frequency f2 if 9/10⋅f2 ≤ f1 ≤ 10/9⋅f2 and/or 9/10⋅f1 ≤ f2 ≤
10/9⋅f1, respectively, preferably if 19/20⋅f2 ≤ f1 ≤ 20/19⋅f2 and/or 19/20⋅f1 ≤ f2 ≤ 20/19⋅f1, respec- tively, most preferably if 99/100⋅f2 ≤ f1 ≤ 100/99⋅f2 and/or 99/100⋅f1 ≤ f2 ≤ 100/99⋅f1, respec- tively. That is, according to the present invention, it is the repetition frequency characterizing an indi- vidual stimulus that is chosen such that it corresponds to (is substantially the same as) the tar- get frequency, i.e., the frequency to be targeted by the stimulation. Put differently, it is the repe- tition frequency characterizing an individual stimulus – and not necessarily the inverse of a cy- cle period T throughout which the stimulation units may be activated/stimuli may be provided thereby exactly once (the above definition of a cycle period T continues to apply in the context of the present invention) – that corresponds to (is substantially the same as) the frequency of the neuronal activity that is to be desynchronized, e.g., about 4 Hz through about 12 Hz for reduc- tion of theta activity that is linked to anxiety, depression and stress, about 6 Hz for Parkinson’s disease and/or tremor suppression, or about 20 Hz for movement disorders and/or reduction of beta activity. Thereby, the present invention allows for particularly targeted and selective and thus particu- larly effective stimulation, while circumventing the use of a relatively short cycle period T that, especially in non-invasive approaches, may result in a stressful impression with the user under- going stimulation as described above with a view to the state of the art. Accordingly, the ap- proach provided by the present invention may be dubbed mild Coordinated Reset (mCR) by En- trainment, or Coordinated Entrainment (CE) for short. The mutual phase-shifts by a fraction of 2π between stimuli provided by different stimulation units add to the effectiveness of the stimulation, i.e., they lead to a particularly strong desyn- chronization. This is because they ensure that a proper multi-cluster state is induced in the neu- ron population targeted by the stimulation, instead of an entrainment that would arise in the ab- sence of phase-shifts. Accordingly, the approach provided by the present invention may also be dubbed Coordinated Entrainment (CE, cf. above) with Phase Lags or, better yet, Desynchroniza- tion by Coordinated Entrainment (DCE). That said, as will be explained in more detail below, the use of mutual phase-shifts throughout one or more cycle periods and/or for one or more tar- get frequencies may be combined with the use of no mutual phase-shifts throughout one or more other cycle periods and/or for one or more other target frequencies. As opposed to DCE,
such stimulation not using any mutual phase-shifts (but otherwise in accordance with the pre- sent invention) may be dubbed Synchronization by Coordinated Entrainment (SCE). SCE may be used to amplify certain (parts of) neuronal activity that may be pathologically reduced, as is the case for theta activity in individuals suffering from Alzheimer’s disease. Moreover, since the respective stimuli that are provided by the stimulation units upon activation are synthesized (e.g., artificially generated, e.g., from scratch) by the apparatus based on at least one parameter, the present invention may circumvent the complexities of delayed-feedback and/or closed-loop approaches known in the state of the art. Rather, it may be implemented as an open-loop approach, i.e., it may not use any feedback, e.g., obtained by measuring the neu- ronal activity to be desynchronized. Alternatively, the present invention may likewise be imple- mented as a semi-open-loop and/or non-delayed-feedback approach, i.e., it may use some feed- back, e.g., obtained by measuring the neuronal activity to be desynchronized, but it may not feedback a time-delayed (and perhaps further processed) version of the measured neuronal ac- tivity as individual stimuli. The present invention may be implemented as an invasive or as a non-invasive approach. For example, the stimulation units may each be adapted to provide a respective electrical stimu- lus to the user, preferably to at least a portion of a population of neurons of the user. To this end, the stimulation units may comprise electrodes implanted in the user, preferably in a brain of the user, and/or electrodes (e.g., electroencephalography, EEG, electrodes) placed on the user, preferably on a scalp of the user (e.g., using a headband with frontal electrodes). In such cases, charge-balanced components may be used, and stimulation may be kept at sub-threshold intensities to stabilize brain activity at a controlled level. Alternatively, the stimulation units may each be adapted to provide a respective vibro-tactile stimulus to the user, preferably to a hand, in particular one or more fingers and/or a thumb, of the user. To this end, the stimulation units may be arranged on an apparatus as disclosed in US 2023/405264 A1 (and DE 102022205898 A1 from the same family). Said apparatus may have a generally spherical shape with a diameter between 5 cm and 100 cm and/or it may be in a shape of a pillow. Additionally or alternatively, said apparatus may comprise one or more re- cesses adapted to receive at least a tip of a finger and/or thumb of the user, wherein each of the
one or more recesses comprises at least one stimulation unit. Further additionally or alterna- tively, said apparatus may be at least partially hollow such that it is positionable over another object. Even further additionally or alternatively, said apparatus may be in a shape of a glove. Further alternatively, the stimulation units may each be adapted to provide a respective acoustic stimulus to the user. In some embodiments, the at least one parameter, based on which the respective stimulus is characterized by a repetition frequency and synthesized by the apparatus, may comprise an indi- cation of the repetition frequency, or of a range of frequencies for the repetition frequency, for the respective stimulus. That is, an operator may directly input the desired repetition frequency and/or the target frequency, i.e., the frequency to be targeted by the stimulation, and/or the fre- quency of the neuronal activity that is to be desynchronized. Likewise, an operator may directly input a desired range of frequencies for the repetition frequency and/or a target range of fre- quencies, i.e., a range of frequencies to be targeted by the stimulation, and/or a range of fre- quencies of neuronal activity that is to be desynchronized. The frequency, or the range of fre- quencies, of the neuronal activity that is to be desynchronized may for example be known from previous experiments, studies and/or literature. Herein, a range of frequencies may refer to a contiguous, but also to a non-contiguous set of frequencies. Additionally or alternatively, the at least one parameter may comprise an indication of a shape for the respective stimulus. For example, it may be indicated that the respective stimulus is to comprise a sinusoidal shape or a rectangular pulse shape. An indication of a shape for the re- spective stimulus may comprise a starting amplitude (e.g., given in arbitrary units) and/or a starting phase-shift (e.g., given as a fraction of 2π). As such, it may for example be indicated that the respective stimulus is to comprise a sinusoidal shape, however with a starting amplitude of 1 and/or a starting phase-shift of π/2, effectively indicating a cosinusoidal shape without any starting phase-shift. Similarly, it may for example be indicated that the respective stimulus is to comprise a rectangular pulse shape, either with a starting amplitude of 1 (implying that the first change in amplitude of the respective stimulus will be a decrease to 0) or with a starting ampli- tude of 0 (implying that the first change in amplitude of the respective stimulus will be an in- crease to 1).
Additionally or alternatively, the at least one parameter may comprise an indication of an inten- sity for the respective stimulus. For example, it may be indicated that the respective stimulus comprises an intensity, e.g., a maximum intensity (occurring, e.g., when the amplitude of the stimulus is at a maximum and/or a minimum), in a range of 0.1 mA to 10 mA, preferably in a range of 0.5 mA to 5 mA, most preferably in a range from 0.5 mA to 2 mA (e.g., where the stim- ulation units are adapted to provide a respective electrical stimulus to the user, e.g., via elec- trodes) or in a range of 0.1 mm to 10 mm, preferably 0.2 mm to 4 mm, most preferably 0.5 mm to 2 mm (e.g., where the stimulation are adapted to provide a respective vibro-tactile stimulus to the user). Additionally or alternatively, the at least one parameter may comprise an indication of a dura- tion for the respective stimulus. For example, it may be indicated that the respective stimulus comprises a duration in a range of 0.01 s to 0.5 s, preferably in a range of 0.05 s to 0.2 s. Prefer- ably, such indication may be in terms of a positive integer multiple of an inverse of the repeti- tion frequency characterizing the respective stimulus. For example, it may be indicated that the respective stimulus comprises 2, 3, 5, 10, 20, 50 or 60 repetitions of a certain shape, e.g., a sinus curve or a rectangular pulse. Synthesizing the respective stimulus based on an indication of any one or more of the above pa- rameters allows to tailor the individual stimuli, providing flexibility in designing the stimulation (which is lacking, e.g., in delayed-feedback approaches known in the state of the art as outlined above) according to the needs of the desired application and/or the targeted neuronal activity, in turn increasing effectiveness of the stimulation. Especially the possibility to indicate the repeti- tion frequency for the respective stimulus may enable a particularly easy setup of the present in- vention and a particularly precise targeting of (pathologically synchronized) neuronal activity. As mentioned above, in some embodiments, the present invention may be implemented as a semi-open-loop approach, i.e., it may use (some) feedback, e.g., obtained by measuring the neu- ronal activity to be desynchronized. To this end, the apparatus may comprise means for deter- mining the at least one parameter based on a measurement of the neuronal activity. In some embodiments, the apparatus may additionally comprise means for performing the measurement of the neuronal activity. This may allow to even better adjust (finetune) the stimulation, and in particular to tailor the individual stimuli, such as to further enhance stimulation effectiveness.
At the same time, a degree of flexibility is maintained as, different from delayed-feedback ap- proaches known in the state of the art, it is not the (entire) measured neuronal activity that is fed back, as such, as individual stimuli. In a sense, the semi-open-loop approach envisaged here exploits the advantages of delayed-feedback approaches known in the state of the art (measur- ing neuronal activity and the effect of the stimulation thereon to adjust), but dispenses with, or at least ameliorates, their disadvantages (lack of flexibility and complexity). In this manner, for example an intensity and/or a target frequency of a respective stimulus may be varied/adapted according to demand. For example, if the neuronal activity to be desynchro- nized lessens (in intensity), an intensity of a respective stimulus may also be chosen to decrease, and vice versa. Likewise, if the neuronal activity to be desynchronized shifts frequency, a repeti- tion frequency of a respective stimulus may also be chosen to shift. Such variations/adaptations of an intensity and/or a target frequency of a respective stimulus may be performed stepwise. That is, such a variation/adaptation may only be performed if the intensity and/or frequency of the neuronal activity to be desynchronized changes by more than 5%, preferably 10%, more preferably 15%, most preferably 20%. It is to be understood that such (e.g., stepwise) variations/adaptations of an intensity and/or a target frequency of a respective stimulus may also be performed independently of any measure- ment of the neuronal activity. For example, varying/adapting an intensity and/or a target fre- quency of a respective stimulus, e.g., within less than 20%, preferably less than 15%, more pref- erably less than 10%, most preferably less than 5% of an initial value and/or an initial range of values (a respective stimulus may, e.g., be synthesized based on an indication of a range of fre- quencies for the repetition frequency, as discussed above) may further enhance stimulation ef- fects. To this end, using the means for performing the measurement of the neuronal activity, the neu- ronal activity (that is to be desynchronized) may be measured and/or tracked, e.g., throughout a duration during which no stimulation unit is active/activated/providing a respective stimulus (as will become clear below, the present invention also envisages such durations), or according to an algorithm that considers (e.g., counters, balances and/or subtracts) any artefacts caused by ongoing stimulation/stimuli.
In some embodiments, the respective stimulus may comprise a plurality of bursts, the inverse of a distance in time between onset times of consecutive bursts of the plurality of bursts corre- sponding to (being substantially the same as) the repetition frequency characterizing the respec- tive stimulus. Therein, at least one burst of the plurality of bursts may optionally be character- ized by a burst frequency higher than the repetition frequency characterizing the respective stimulus. A plurality of bursts may for example comprise 2, 4, 5, 6, 8 or 10 bursts. For example, the distance in time between onset times of consecutive bursts of the plurality of bursts may be in a range of 10 ms to 500 ms, preferably in a range of 20 ms to 200 ms, e.g., 66 ms, such that the inverse of this distance in time, corresponding to (being substantially the same as) the repe- tition frequency characterizing the respective stimulus, may be, e.g., 15 Hz (if the distance in time between onset times of consecutive bursts of the plurality of bursts is 66 ms). An individual burst may comprise a plurality of pulses, e.g., 2, 4, 5, 6, 8 or 10. For example, the distance in time between onset times of consecutive pulses of an individual burst may be in a range of 0.5 ms to 20 ms, preferably in a range of 1 ms to 10 ms, e.g., 4 ms, such that the inverse of this dis- tance in time, corresponding to the burst frequency characterizing the respective burst, may be, e.g. 250 Hz (if the distance in time between onset times of consecutive pulses of an individual burst is 4 ms). This shape of an individual stimulus has proven to be particularly effective in bringing about the desired desynchronization of neuronal activity. Generally, stimuli provided by the same stimulation unit may be identical or different, e.g., with respect to repetition frequency and/or shape. The same applies to stimuli provided by different stimulation units, i.e., stimuli provided by different stimulation units may be identical or differ- ent, e.g., with respect to repetition frequency and/or shape. This allows for flexibility in design- ing the stimulation according to the needs of the desired application and/or the targeted neu- ronal activity, eventually enhancing effectiveness of the stimulation. For example, at least two stimuli provided by different stimulation units may be identical in at least repetition frequency and/or shape; it may also be that all stimuli provided by the stimulation units are identical in at least repetition frequency and/or shape. But it may also be that at least two stimuli provided by a same stimulation unit differ in at least repetition frequency and/or shape. For example, repeti- tion frequency and/or shape of stimuli provided by a same stimulation unit may vary over time, e.g., repetition frequency may first increase and then decrease again, preferably in steps in a range of 0.5 Hz to 7.5 Hz, preferably in a range of 2 Hz to 5 Hz, e.g., 2.5 Hz, eventually returning to an initial repetition frequency (e.g., a repetition frequency of a first stimulus provided by the same stimulation unit throughout a given duration). Notably, it may be possible that at least two
stimuli provided by a same stimulation unit differ in at least repetition frequency and/or shape, while at least two stimuli provided by different stimulation units are identical in at least repeti- tion frequency and/or shape. For example, it may be that repetition frequency and/or shape of stimuli provided by different stimulation units vary over time in the exact same manner. Similarly, a mutual phase-shift between stimuli provided by different stimulation units may vary over time. For example, a first pair of stimuli provided by different stimulation units may be mutually phase-shifted by a first fraction of 2π and a second pair of stimuli provided by the dif- ferent stimulation units may be mutually phase-shifted by a second fraction of 2π different from the first fraction of 2π. Not only does this allow for additional flexibility in designing the stimu- lation according to the needs of the desired application and/or the targeted neuronal activity. A variation in mutual phase-shifts has also proven, as such, to enhance effectiveness of the stimu- lation. In some embodiments, the apparatus’ means for individually activating may be configured to sequentially activate the stimulation units of the set such that that stimuli provided by the stim- ulation units do not overlap in time. This may be advantageous to prevent an over-stimulation of the user, e.g., it may limit a total charge applied to a user where the stimulation units adapted to provide a respective electrical stimulus to the user. Generally, whether or not to have stimuli provided by the stimulation units overlap in time is another dimension of flexibility in designing the stimulation according to the needs of the desired application and/or the targeted neuronal activity, which may eventually enhance effectiveness of the stimulation. It is to be noted that stimulation units may be activated sequentially (i.e., not simultaneously, but one after the other), but the respective stimuli may nevertheless overlap (e.g., because their respective dura- tion is longer than the separation in time between the activation of the respective stimulation units). To achieve the desired mutual phase-shift between stimuli provided by different stimulation units it may be necessary to consider not only the shape (including any starting amplitude and/or starting phase-shift as discussed above) of the respective stimuli, but also the separation in time between the activation of the respective stimulation units, i.e., the separation in time be- tween onset times T1…TN at which the apparatus’ means for individually activating may be con- figured to sequentially activate the N stimulation units of the set. If the stimuli provided by two
stimulation units comprise a same shape, the respective two onset times may need to be sepa- rated in time by a non-integer multiple of an inverse of the repetition frequency. Otherwise, the stimuli provided by the two stimulation units may not be mutually phase-shifted by a fraction of 2π. Conversely, if the stimuli provided by two stimulation units comprise a mutually different shape, in particular different starting phase-shifts, the respective two onset times may be sepa- rated in time by an integer multiple of an inverse of the repetition frequency. In such cases, choosing the separation in time to be an integer multiple of an inverse of the repetition fre- quency may be particularly straightforward to implement and may yield an even distribution of the separation in time between the activation of the respective stimulation units, while ensuring that the stimuli provided by the two stimulation units are nevertheless mutually phase-shifted by a fraction of 2π. In some embodiments, the apparatus’ means for individually activating may be configured to sequentially activate the stimulation units of the set at respective onset times T1…TN throughout a cycle period T, wherein the onset times T1…TN are not arranged substantially uniformly throughout the cycle period T. In particular, the onset times T1…TN may not be arranged in uni- form intervals substantially equal to T/N throughout the cycle period T. That is, the present in- vention may implement Dynamic Coordinated Reset (DCR) neuromodulation as disclosed in WO 2021/028119 A1 (and US 2022/288396 A1 from the same family) for invasive approaches and in DE 102020208431 A1 for non-invasive approaches, which may result in significantly stronger and longer-lasting desynchronization within a population of neurons, which may in turn have a stronger and longer-lasting effect on symptoms resulting from pathologically syn- chronized neuronal activity. Additionally or alternatively, the apparatus’ means for individually activating may be configured to sequentially activate the stimulation units of the set at respective onset times T1…TN through- out each cycle period T of a plurality of cycle periods. That is, it is envisaged that stimuli may be provided in recurring patterns, as in (classical) CR neuromodulation and in DCR neuromodula- tion as disclosed in the above-referenced applications. Therein, the apparatus’ means for individually activating may be configured to sequentially acti- vate the stimulation units of the set at the same respective onset times T1…TN throughout the plurality of cycle periods. Alternatively, the means for individually activating may be configured
to sequentially activate at least two stimulation units of the set such that the respective two on- set times are separated in time by a first value in a first cycle period of the plurality of cycle peri- ods and by a second value different from the first value in a second cycle period of the plurality of cycle periods; that is, onset times T1…TN may change from one cycle period to the next. At the same time, stimuli provided by the two stimulation units may or may not be mutually phase- shifted by a same fraction of 2π in the first cycle and in the second cycle; that is, the mutual phase-shift may or may not remain the same from one cycle period to the next. Additionally or alternatively, the means for individually activating may be configured to sequen- tially activate a further set of M≥2 stimulation units different from the set of N≥2 stimulation units throughout at least one of the plurality of cycle periods, wherein M is an integer. Additionally or alternatively, the means for individually activating may be configured not to acti- vate any stimulation units for a duration corresponding to at least one cycle period in alteration with activating stimulation units throughout the plurality of cycle periods. It is also envisaged that there may be one or more cycle periods during which stimuli provided by different stimulation units are mutually phase-shifted by a fraction of 2π, in alteration with one or more cycle periods during which stimuli provided by different stimulation units are not mutually phase-shifted by a fraction of 2π, potentially in further alteration with durations corre- sponding to at least one cycle period during which no stimulation units are activated. To this end, the apparatus’ means for individually activating may be configured to sequentially activate the stimulation units of the set such that stimuli provided by different stimulation units are mu- tually phase-shifted by a fraction of 2π throughout at least a first cycle period and such that stimuli provided by different stimulation units are not mutually phase-shifted throughout at least a second cycle period (different from the first); and optionally further, the apparatus’ means for individually activating may be configured not to activate any stimulation units throughout a duration corresponding to at least one cycle period. Throughout different cycle pe- riods, different frequencies and/or ranges of frequencies may be targeted by the stimulation. That is, while the means for individually activating is configured to sequentially activate the stimulation units of the set such that stimuli provided by different stimulation units are mutu- ally phase-shifted by a fraction of 2π, the respective stimuli may be synthesized by the appa- ratus, such that they are characterized by a first repetition frequency, or a first range of frequen-
cies for the repetition frequency, that corresponds to a first frequency, or a first range of fre- quencies, of the neuronal activity. Then, while the means for individually activating is config- ured to sequentially activate the stimulation units of the set such that stimuli provided by differ- ent stimulation units are not mutually phase-shifted, the respective stimuli may be synthesized by the apparatus, such that they are characterized by a second repetition frequency, or a second range of frequencies for the repetition frequency, that corresponds to a second frequency, or a second range of frequencies, of the neuronal activity. Combining the use of mutual-phase shifts (DCE) and the use of no mutual phase-shifts (SCE) in this manner allows to deliberately tailor and/or shape neuronal activity. Certain (parts of) neu- ronal activity may be targeted for reduction (using mutual phase-shifts/DCE), while other (parts of) neuronal activity may be targeted for amplification (using no mutual-phase shifts/SCE). For example, considering an individual suffering from Alzheimer’s disease with symptoms of move- ment disorder, combining DCE and SCE it becomes possible to selectively reduce movement disorder symptoms (linked to neuronal activity at about 15 Hz through 30 Hz), while selectively amplifying neuronal activity at about 4 Hz through 12 Hz that is reported to be pathologically reduced in individuals suffering from Alzheimer’s disease. A suitable stimulation could include DCE for 10 minutes at about 15 Hz through about 30 Hz, followed by SCE for 5 minutes at about 8 Hz through about 13 Hz. This succession of DCE and SCE could then be repeated one or more times for even better results. The present invention may be used for the treatment of any diseases associated with, or caused by, (pathologically) synchronized neuronal activity. This includes neurological disorders like Parkinson’s disease, movement disorders, psychological diseases like depression, anxiety disor- ders, obsessive-compulsive disorder (OCD) or burn-out, or tinnitus. Accordingly, it is further described herein a method for desynchronizing neuronal activity. Therein, a plurality of stimuli is provided to at least two parts of a user, wherein each stimulus of the plurality of stimuli is synthesized based on at least one parameter, such that the respective stimulus is characterized by a repetition frequency that corresponds to a frequency of the neu- ronal activity, wherein stimuli provided to different parts of the user are mutually phase-shifted by a fraction of 2π. The above explanations concerning the apparatus according to the present invention apply mutatis mutandis to such method.
Such method may be performed by a suitable apparatus, e.g., by an apparatus as discussed above. Accordingly, in a further aspect, the present invention also relates to a computer program com- prising instructions which, when the program is executed by a computer of an apparatus, cause the apparatus to carry out a method as just described. The above-discussed need is also ad- dressed at least in part by such a computer program. 4. Brief Description of the Figures Fig. 1A: Four stimuli comprising a sinusoidal shape each, mutually phase shifted by a fraction of 2π; Fig. 1B: Four stimuli comprising a rectangular pulse shape each, mutually phase shifted by a fraction of 2π; Fig. 2A: mCR pattern comprising three cycle periods T, each cycle period T comprising four stimuli comprising a sinusoidal shape, mutually phase shifted by a fraction of 2π; Fig. 2B: mCR pattern comprising three cycle periods T, each cycle period T comprising four stimuli comprising a rectangular pulse shape, mutually phase shifted by a fraction of 2π; Fig. 3: Four stimuli comprising four bursts each, each burst comprising four pulses, the four stimuli mutually phase shifted by a fraction of 2π; Fig. 4: mCR pattern comprising three cycle periods comprising four stimuli each, each stimulus comprising four bursts, and a duration corresponding to two cycle peri- ods not comprising any stimulus; Fig. 5: Stepwise variation of a repetition frequency characterizing consecutive stimuli over time.
5. Detailed Description of Some Embodiments For the sake of brevity only a few embodiments will be described in the following. The skilled person will recognize that the specific features described with reference to these embodiments may be modified and combined differently and that individual features may also be omitted if they are not essential. The general explanations in the sections above will also be valid for the following more detailed explanations. Fig. 1A through 4 show four signals, Signal 1, Signal 2, Signal 3 and Signal 4, over time. These four signals may correspond to stimuli provided to four different locations on and/or in a user by a set of respective N=4 stimulation units. That is, the four signals may for example be (have been) output by the set of respective N=4 stimulation units and/or output to the four different locations on and/or in the user. Similarly, the four signals may for example be (have been) measured at the set of respective N=4 stimulation units and/or at the four different locations on and/or in the user. Likewise, the four signals may correspond to control signals that may be (have been) provided to a set of N=4 stimulation units of an apparatus for desynchronizing neu- ronal activity, such as to control the set of N=4 stimulation units to in turn provide correspond- ing stimuli to four respective locations on and/or in the user. For the sake of brevity and reada- bility, the expression stimulus/stimuli shall be understood herein to include such control sig- nals, too. Accordingly, the four signals, Signal 1, Signal 2, Signal 3 and Signal 4, are shown in Fig.1A through 4 in arbitrary units. Generally, the stimuli illustrated in Fig.1A through 4 may be (have been) synthesized (e.g., arti- ficially generated, e.g., from scratch), based on at least one parameter, by an apparatus for desynchronizing neuronal activity. The stimuli may be (have been) provided to and/or by a set of N=4 stimulation units of the apparatus for desynchronizing neuronal activity, which set of N=4 stimulation units may provide (have provided), upon activation, the stimuli to a user. To this end, the apparatus may comprise means for individually activating the set of N=4 stimula- tion units. The means for individually activating the simulation units of the set may be config- ured to individually activate the simulation units of the set such that stimuli provided by differ- ent stimulation units are mutually phase-shifted by a fraction of 2π. While the following exclusively assumes a set of N = 4 stimulation units (and/or four locations on and/or in a user to be stimulated), the present invention is not limited thereto. The present
invention may use any set of two or more stimulation units, i.e., any (integer) number of stimu- lation units larger than two (N≥2). Preferred embodiments use a set of N≤20, e.g., N = 4, N = 5, N = 6 or N = 10 stimulation units. It is however to be understood that an apparatus for desyn- chronizing neuronal activity may comprise a larger set, i.e., a larger number, of stimulation units, but may only use a subset of these according to the present invention. That is, it is still ac- cording to the present invention if an apparatus for desynchronizing neuronal activity comprises a set of M stimulation units, M being an integer, but only uses a set of N stimulation units as de- fined by the claims, with M>N≥2. Similarly, while the following exclusively assumes that the set of N = 4 stimulation units is al- ways activated in the same order (such that Signal 1 is the first to be non-zero, followed by Sig- nal 2, followed by Signal 3, followed by Signal 4), the present invention is not limited thereto. The present invention may activate stimulation units and/or provide stimuli thereby in any or- der, e.g., an order in which stimulation units are activated and/or stimuli are provided thereby may be random or pseudo-random. Especially in such cases, it may be advantageous if a shape, including a starting amplitude and/or a starting phase-shift, of stimuli provided by a same stim- ulation unit remains constant over time, as is assumed in the following, to ensure that stimuli provided by different stimulation units are mutually phase-shifted by a fraction of 2π. However, it is to be understood that, generally, a shape, including a starting amplitude and/or a starting phase-shift, of stimuli provided by a same stimulation unit may vary over time. Fig. 1A shows four stimuli 1100a, 1200a, 1300a and 1400a. All four stimuli 1100a, 1200a, 1300a and 1400a comprise a sinusoidal shape, i.e., they comprise a plurality of (e.g., 59) repetitions of a sinus curve 1101a, 1201a, 1301a and 1401a, respectively. The four stimuli 1100a, 1200a, 1300a and 1400a, and likewise the sinus curves 1101a, 1201a, 1301a and 1401a, all comprise a starting amplitude and/or a starting phase-shift of 0. The sinus curves 1101a, 1201a, 1301a and 1401a are identical in shape. As a result, all four stimuli 1100a, 1200a, 1300a and 1400a are characterized by a same repetition frequency, specifically a repetition frequency of 20 Hz. This repetition fre- quency corresponds to, and is in fact substantially the same as, the frequency of neuronal activ- ity that is pathologically synchronized due to movement disorders and/or reduction of beta ac- tivity. In other embodiments, the repetition frequency characterizing stimuli such as stimuli 1100a, 1200a, 1300a and 1400a may be different, e.g., 6 Hz, such that it corresponds to, and is in fact substantially the same as, the frequency of neuronal activity that is pathologically syn- chronized due to Parkinson’s disease and/or tremor.
The four stimuli 1100a, 1200a, 1300a and 1400a are mutually phase-shifted by a fraction of 2π. More specifically, the four stimuli 1100a, 1200a, 1300a and 1400a are mutually phase-shifted by (an integer multiple of) 2π/N = 2π/4 = π/2. That is, stimulus 1200a is phase-shifted with re- spect to stimulus 1100a by π/2, stimulus 1300a is phase-shifted with respect to stimulus 1200a by π/2, and stimulus 1400a is phase-shifted with respect to stimulus 1300a by π/2. In other em- bodiments, mutual-phase shifts between stimuli provided by different stimulation units may not be an integer multiple of the same fraction of 2π (in Fig. 1A: 2π/N = 2π/4 = π/2); e.g., mutual- phase shifts between stimuli provided by different stimulation units may be random or pseudo- random fractions of 2π. In the embodiment of Fig. 1A, said mutual phase-shifts are realized by delaying (offsetting in time) a subsequent stimulus from another stimulation unit and/or activa- tion of that stimulation unit. With respect to an onset time of stimulus 1100a, an onset time of stimulus 1200a is delayed by a quarter of the inverse of the repetition frequency, i.e., by 1/4⋅1/20 s = 1/80 s, an onset time of stimulus 1300a is delayed by half (two quarters) of the in- verse of the repetition frequency, i.e., by 1/2⋅1/20 s = 1/40 s, and an onset time of stimulus 1400a is delayed by three quarters of the inverse of the repetition frequency, i.e., by 3/4⋅1/20 s = 3/80 s. Such a pattern (more specifically: mCR pattern) may be achieved using a means for individually activating the simulation units of an apparatus for desynchronizing neuronal activity that is con- figured to individually activate the simulation units such that stimuli provided by different stim- ulation units are mutually phase-shifted by a fraction of 2π, specifically by (an integer multiple of) 2π/N = 2π/4 = π/2. Assuming, on the one hand, that the means for individually activating is configured to sequentially activate the stimulation units of the set at respective onset times T1…TN, and, on the other hand, that stimuli provided by two stimulation units comprise a same shape (as is the case in Fig.1A), the respective two onset times may need to be separated in time by a non-integer multiple of an inverse of the repetition frequency, e.g., by a quarter, a half (two quarters) and/or three quarters, as just discussed. Fig. 1B shows four stimuli 1100b, 1200b, 1300b and 1400b similar to stimuli 1100a, 1200a, 1300a and 1400a of Fig.1A. However, stimuli 1100b, 1200b, 1300b and 1400b comprise a rec- tangular pulse shape, i.e., they comprise a plurality of (e.g., 59) repetitions of a rectangular pulse 1101b, 1201b, 1301b and 1401b, respectively. The four stimuli 1100b, 1200b, 1300b and 1400b,
and likewise the rectangular pulses 1101b, 1201b, 1301b and 1401b, all comprise a starting am- plitude of 1 and an ending amplitude of 0. That is, the rectangular pulses compromise an ampli- tude of 1 in their first half and an amplitude of 0 in their second half. The rectangular pulses 1101b, 1201b, 1301b and 1401b are identical in shape. As a result, all four stimuli 1100b, 1200b, 1300b and 1400b are characterized by a same repetition frequency, specifically a repetition fre- quency of 20 Hz, just as stimuli 1100a, 1200a, 1300a and 1400a of Fig.1A. The above explana- tions concerning the embodiment of Fig.1A apply mutatis mutandis to the embodiment of Fig. 1B. The mCR pattern of Fig. 1B may be particularly effective in non-invasive, e.g., vibro-tactile and/or acoustic, approaches. Fig.2A shows an evolved mCR pattern that may be derived from the mCR pattern of Fig.1A as will be described further below. The mCR pattern of Fig.2A comprises three cycle periods, and – in line with the definition of a cycle period as given above – each cycle period comprises four stimuli, wherein each of the four stimuli may be (have been) provided to and/or by a respective stimulation unit of a set of N=4 stimulation units. That is, a first cycle period comprises stimuli 2110a, 2210a, 2310a and 2410a, a second cycle period comprises stimuli 2120a, 2220a, 2320a and 2420a, and a third cycle period comprises stimuli 2130a, 2230a, 2330a and 2430a. Put dif- ferently, each stimulation unit of a set of N=4 stimulation units may provide (have provided) three stimuli in (a single pass through) the mCR pattern of Fig.2A. That is, a first stimulation unit may provide (have provided) stimuli 2110a, 2120a and 2130a, a second stimulation unit may provide (have provided) stimuli 2210a, 2220a and 2230a, a third stimulation unit may pro- vide (have provided) stimuli 2310a, 2320a and 2330a, and a fourth stimulation unit may pro- vide (have provided) stimuli 2410a, 2420a and 2430a. The stimuli 2110a, 2120a and 2130a, 2210a, 2220a and 2230a, 2310a, 2320a and 2330a, 2410a, 2420a and 2430a comprise a sinusoidal shape, i.e., they comprise a plurality of repetitions, namely three (not 59 as stimuli 1100a, 1200a, 1300a and 1400a of Fig.1A) of a sinus curve 2111a, 2121a and 2131a, 2211a, 2221a and 2231a, 2311a, 2321a and 2331a, 2411a, 2421a and 2431a, respectively. The stimuli 2110a, 2120a and 2130a, 2210a, 2220a and 2230a, 2310a, 2320a and 2330a, 2410a, 2420a and 2430a, and likewise the sinus curves 2111a, 2121a and 2131a, 2211a, 2221a and 2231a, 2311a, 2321a and 2331a, 2411a, 2421a and 2431a, all comprise a starting amplitude of 0 and/or a starting phase-shift of 0. The sinus curves 2111a, 2121a and 2131a, 2211a, 2221a and 2231a, 2311a, 2321a and 2331a, 2411a, 2421a and 2431a are identical in shape. As a result, all stimuli 2110a, 2120a and 2130a, 2210a, 2220a and 2230a, 2310a, 2320a
and 2330a, 2410a, 2420a and 2430a are characterized by a same repetition frequency, specifi- cally a repetition frequency of 20 Hz. With respect to the repetition frequency, the explanations on the embodiment of Fig. 1A apply mutatis mutandis to the embodiment of Fig. 2A. Notably, it is the repetition frequency of a single stimulus, and not the inverse of the cycle period, that cor- responds to, or is in fact substantially the same as, 20 Hz, i.e., the frequency of neuronal activity that is pathologically synchronized due to movement disorders and/or reduction of beta activity. In the embodiment of Fig. 2A, stimuli within a same cycle period, and more generally stimuli provided by different stimulation units, are mutually phase-shifted by a fraction of 2π., specifi- cally by (an integer multiple of) 2π/N = 2π/4 = π/2. This may be the case because the mCR pat- tern of Fig.2A may be derived from the mCR pattern of Fig.1A in that only portions of the latter are used: Baseline signals (dotted lines) 2100a, 2200a, 2300a and 2400a shown in Fig. 2A cor- respond to stimuli 1100a, 1200a, 1300a and 1400a, respectively, as shown in Fig.1A. In a sense, the mCR pattern of Fig.2A may be arrived at by intermittently muting (and/or unmuting) stim- uli 1100a, 1200a, 1300a and 1400a of Fig.1A. Alternatively, these mutual phase-shifts may be considered realized by delaying (offsetting in time) a subsequent stimulus from another stimulation unit and/or activation of that stimulation unit. With respect to an onset time of stimulus 2110a, an onset time of stimulus 2210a is delayed by four and a quarter of the inverse of the repetition frequency, i.e., by 41/4⋅1/20 s = 17/80 s, an onset time of stimulus 2310a is delayed by eight and a half of the inverse of the repetition fre- quency, i.e., by 81/2⋅1/20 s = 17/40 s, and an onset time of stimulus 2410a is delayed by twelve and three quarters of the inverse of the repetition frequency, i.e., by 123/4⋅1/20 s = 51/80 s. The same applies mutatis mutandis to stimuli 2120a, 2220a, 2320a and 2420a as well as 2130a, 2230a, 2330a and 2430a. In the exact same manner, the evolved mCR pattern of Fig.2B may be derived from the mCR pattern of Fig. 1B. The above explanations apply mutatis mutandis. Like numerals denote like elements. Generally, this manner of deriving an evolved mCR pattern from another (e.g., less evolved) one, may be exploited to implement the present invention in a particularly efficient way. That is, in some embodiments, an apparatus for desynchronizing neuronal activity may first synthesize, for each of its stimulation units, a baseline signal, such as baseline signals 2100a, 2200a, 2300a and
2400a of Fig.2A and baseline signals 2100b, 2200b, 2300b and 2400b of Fig.2B. The appa- ratus may then synthesize the respective stimulus, to be provided to a user by the respective stimulation unit, based on the respective baseline signal. In such embodiments, the at least one parameter, based on which the apparatus is to synthesize the respective stimulus, may define how to derive the respective stimulus from the respective baseline signal. For example, the at least one parameter may indicate one or more start points, durations and/or end points that de- fine one or more portions of the respective baseband signal that are to be synthesized as respec- tive stimuli. Based on Figs. 2A and 2B, it may also be readily understood why it may be advantageous if a shape, including a starting amplitude and/or a starting phase-shift, of stimuli provided by a same stimulation unit remains constant over time if stimulation units are activated and/or pro- vide stimuli in different orders throughout different cycle periods. As can be seen from Figs.2A and 2B, whatever the order in which the set of N = 4 stimulation units is activated, stimuli pro- vided by different stimulation units will always be mutually phase-shifted by a fraction of 2π. The temporal structure of three exemplary cycle periods throughout which stimulation units are activated and/or provide stimuli in different orders may be as follows, wherein the inverse of the repetition frequency is denoted PRep. Stimulus such as 2110a (3⋅PRep) – 11/4⋅PRep – Stimulus such as 2210a (3⋅PRep) – 11/4⋅PRep – Stimulus such as 2310a (3⋅PRep) – 11/4⋅PRep – Stimulus such as 2410a (3⋅PRep) (as shown in Fig. 2A and 2B) 3/4⋅PRep – Stimulus such as 2410a (3⋅PRep) – 11/2⋅PRep – Stimulus such as 2210a (3⋅PRep) – 13/4⋅PRep – Stimulus such as 2110a (3⋅PRep) – 11/2⋅PRep – Stimulus such as 2310a (3⋅PRep) 1/4⋅PRep – Stimulus such as 2210a (3⋅PRep) – 13/4⋅PRep – Stimulus such as 2110a (3⋅PRep) – 11/2⋅PRep – Stimulus such as 2310a (3⋅PRep) – 11/4⋅PRep – Stimulus such as 2410a (3⋅PRep) Fig. 3 shows another embodiment of stimuli, i.e., four stimuli 3100, 3200, 3300 and 3400. All four stimuli 3100, 3200, 3300 and 3400 comprise a plurality of bursts 3110, 3120, 3130 and 3140, 3210, 3220, 3230 and 3240, 3310, 3320, 3330 and 3340, 3410, 3420, 3430 and 3440, re- spectively. In the embodiment of Fig. 3, the stimuli 3100, 3200, 3300 and 3400 comprise N1 = 4
bursts, but in other embodiments there may be a different number of stimuli comprising a dif- ferent number N1 of bursts, e.g., N1 = 2, N1 = 5, N1 = 6, N1 = 8 or N1 = 10. The number N1 of bursts may also vary across stimuli. The respective bursts within the stimuli 3100, 3200, 3300 and 3400, e.g., their onset times, are separated in time by P1. In the embodiment of Fig.3, this distance in time between onset times of consecutive bursts of the plurality of bursts P1 is the same across the stimuli 3100, 3200, 3300 and 3400, but in other embodiments, the distance in time between onset times of consec- utive bursts of the plurality of bursts P1 may be different across stimuli. The inverse of this dis- tance in time between onset times of consecutive bursts of the plurality of bursts P1, i.e., 1/P1, corresponds to, or is in fact substantially the same as, the repetition frequency characterizing the respective stimulus, and in turn also corresponds to, or is in fact substantially the same as, the frequency of the neuronal activity, e.g., the frequency of the neuronal activity to be desyn- chronized by the stimulation. In the embodiment of Fig. 3, the distance in time between onset times of consecutive bursts of the plurality of bursts is P1 = 66 ms, such that 1/P1 = 15 Hz. Each burst 3110, 3120, 3130 and 3140, 3210, 3220, 3230 and 3240, 3310, 3320, 3330 and 3340, 3410, 3420, 3430 and 3440 comprises a plurality of pulses 3111, 3121, 3131 and 3141, 3211, 3221, 3231 and 3241, 3311, 3321, 3331 and 3341, 3411, 3421, 3431 and 3441, respectively. In the em- bodiment of Fig.3, the bursts 3110, 3120, 3130 and 3140, 3210, 3220, 3230 and 3240, 3310, 3320, 3330 and 3340, 3410, 3420, 3430 and 3440 comprise N2 = 4 pulses, but in other embodi- ments there may be a different number N1 of bursts comprising a different number N2 of pulses, e.g., N1 = 2, N1 = 5, N1 = 6, N1 = 8 or N1 = 10 and/or N2 = 2, N2 = 5, N2 = 6, N2 = 8 or N2 = 10. The number N1 of bursts and the number N2 of pulses may also vary across stimuli and bursts, respectively. The respective pulses within the bursts 3110, 3120, 3130 and 3140, 3210, 3220, 3230 and 3240, 3310, 3320, 3330 and 3340, 3410, 3420, 3430 and 3440, e.g., their onset times, are separated in time by P2. In the embodiment of Fig.3, this distance in time between onset times of consecu- tive pulses of the plurality of pulses P2 is the same across the bursts 3110, 3120, 3130 and 3140, 3210, 3220, 3230 and 3240, 3310, 3320, 3330 and 3340, 3410, 3420, 3430 and 3440, but in other embodiments, the distance in time between onset times of consecutive pulses of the plu- rality of pulses P2 may be different across bursts and/or stimuli. The inverse of this distance in
time between onset times of consecutive pulses of the plurality of pulses P2, i.e., 1/P2, corre- sponds to a burst frequency characterizing the respective burst. Such a burst frequency may be higher than the repetition frequency characterizing the respective stimulus, and in turn also higher than the frequency of the neuronal activity, e.g., the frequency of the neuronal activity to be desynchronized by the stimulation. In preferred embodiments, a burst frequency characteriz- ing a burst may be significantly higher than a repetition frequency characterizing a stimulus comprising said burst, e.g., such that a user undergoing stimulation cannot distinctly perceive individual pulses, but rather perceives a burst as a whole. In the embodiment of Fig. 3, the dis- tance in time between onset times of consecutive pulses of the plurality of pulses is P2 = 4 ms, such that 1/P2 = 250 Hz. In other embodiments, the inverse of a distance in time between onset times of consecutive pulses of the plurality of pulses P2, i.e., 1/P2, may be in in a range of 50 Hz to 1000 Hz, preferably in a range of 100 Hz to 500 Hz, more preferably in a range of 200 Hz to 300 Hz. The four stimuli 3100, 3200, 3300 and 3400 are mutually phase-shifted by a fraction of 2π. More specifically, the four stimuli 3100, 3200, 3300 and 3400 are mutually phase-shifted by (an integer multiple of) 2π/N = 2π/4 = π/2. That is, stimulus 3200 is phase-shifted with respect to stimulus 3100 by π/2, stimulus 3300 is phase-shifted with respect to stimulus 3200 by π/2, and stimulus 3400 is phase-shifted with respect to stimulus 3300 by π/2. In other embodi- ments, mutual-phase shifts between stimuli provided by different stimulation units may not be an integer multiple of the same fraction of 2π (in Fig. 3: 2π/N = 2π/4 = π/2); e.g., mutual-phase shifts between stimuli provided by different stimulation units may be random or pseudo-ran- dom fractions of 2π. In the embodiment of Fig.3, where the distance in time between onset times of consecutive bursts of the plurality of bursts is P1 = 66 ms, a mutual phase shift by π/2 corresponds to an offset in time O of O = 1/4⋅P1 = 16.5 ms. More generally, such an offset in time O may, e.g., be between 0 and 1/2⋅P1, e.g., it may be 1/N⋅P1. In the embodiment of Fig. 3, these mutual phase-shifts are realized by delaying (offsetting in time) a subsequent stimulus from an- other stimulation unit and/or activation of that stimulation unit. With respect to an onset time of stimulus 3100, an onset time of stimulus 3200 is delayed by 21/4⋅P1 = 9/4⋅P1 = 148.5 ms, an onset time of stimulus 3300 is delayed by 41/2⋅P1 = 9/2⋅P1 = 297 ms, and an onset time of stim- ulus 3400 is delayed by 63/4⋅P1 = 9/2⋅P1 = 445.5 ms. As indicated in Fig.3 (cf. greyed out bursts not provided with reference signs), the shown, evolved mCR pattern may be derived from another (e.g., less evolved) one, e.g., based on four
baseline signals (mutually phase-shifted by, if applicable integer multiples of, 2π/N = 2π/4 = π/2, or offset by, if applicable integer multiples of, O = 1/N⋅P1 = 1/4⋅P1, respectively) as de- scribed above, e.g., with reference to Figs.2A and 2B. Fig. 4 shows an evolved mCR pattern that may in turn be derived from the evolved mCR pattern of Fig. 3. Specifically, the evolved mCR pattern of Fig. 4 may be derived by repeating the mCR pattern of Fig. 3 C = 3 times, followed by a duration, corresponding to P = 2 times a duration of the evolved mCR pattern of Fig.3, without any stimuli (e.g., because no stimulation unit is acti- vated during this duration). Accordingly, bursts 4110a-4140a, 4110b-4140b and 4110c-4140c may be the same as bursts 3110, 3120, 3130 and 3140, bursts 4210a-4240a, 4210b-4240b and 4210c-4240c may be the same as bursts 3210, 3220, 3230 and 3240, bursts 4310a-4340a, 4310b-4340b and 4310c-4340c may be the same as bursts 3310, 3320, 3330 and 3340, and bursts 4410a-4440a, 4410b-4440b and 4410c-4440c may be the same as bursts 3410, 3420, 3430 and 3440. The duration of the evolved mCR pattern of Fig. 3 corresponds to its cycle period T, in line with the above definition of a cycle period, as can be seen in Fig.4. The evolved mCR pattern of Fig.4 may in turn be repeated a number of times to derive a (more evolved) mCR pattern. Generally, by repeating a (less evolved) mCR pattern C times, followed by a duration, corre- sponding to P times a duration (e.g., a cycle period) of said (less evolved) mCR pattern, without any stimuli, one may derive other evolved mCR patterns. By repeating (e.g., such other evolved) mCR patterns, yet other (even more evolved) mCR patterns may be derived that promise ever better stimulation results, i.e., ever stronger desynchronization of neuronal activity pathologi- cally synchronized at the repetition frequency. Fig. 5 shows a stepwise variation of a repetition frequency characterizing consecutive stimuli, provided by a same stimulation unit or by different stimulation units, over time. As illustrated, a first stimulus, or a first set of consecutive stimuli, may be characterized by a first repetition fre- quency 1/P1 (cf. embodiment of Fig. 3) 5001. A second, consecutive stimulus, or a second, con- secutive set of consecutive stimuli, may be characterized by a second repetition frequency 1/P1 5002 higher than the first repetition frequency 1/P15001. A third, consecutive stimulus, or a
third, consecutive set of consecutive stimuli, may be characterized by a third repetition fre- quency 1/P15003 higher than the second repetition frequency 1/P15002. A fourth, consecutive stimulus, or a fourth, consecutive set of consecutive stimuli, may be characterized by a fourth repetition frequency 1/P15004 higher than the third repetition frequency 1/P15003. A fifth, consecutive stimulus, or a fifth, consecutive set of consecutive stimuli, may be characterized by a fifth repetition frequency 1/P15005 higher than the fourth repetition frequency 1/P15004. A sixth, consecutive stimulus, or a sixth, consecutive set of consecutive stimuli, may be character- ized by a sixth repetition frequency 1/P15006 lower than the fifth repetition frequency 1/P1 5005; the sixth repetition frequency 1/P15006 may be substantially the same as the fourth repe- tition frequency 1/P15004. A seventh, consecutive stimulus, or a seventh, consecutive set of consecutive stimuli, may be characterized by a seventh repetition frequency 1/P15007 lower than the sixth repetition frequency 1/P15006; the seventh repetition frequency 1/P15007 may be substantially the same as the third repetition frequency 1/P15003. An eighth, consecutive stimulus, or an eighth, consecutive set of consecutive stimuli, may be characterized by an eighth repetition frequency 1/P15008 lower than the seventh repetition frequency 1/P15007; the eighth repetition frequency 1/P15008 may be substantially the same as the second repetition frequency 1/P15002. A ninth, consecutive stimulus, or a ninth, consecutive set of consecutive stimuli, may be characterized by a ninth repetition frequency 1/P15009 lower than the eighth repetition frequency 1/P15008; the ninth repetition frequency 1/P15009 may be substantially the same as the first repetition frequency 1/P15001. Generally, a repetition frequency characterizing consecutive stimuli, provided by a same stimu- lation unit or by different stimulation units, may vary over time, e.g., in a range from 2 Hz to 100 Hz, preferably in a range from 10 Hz to 50 Hz, most preferably in a range from 15 to 30 Hz. A variation may occur in (preferably equally sized) steps, e.g., 3, 4, 5, 6, 8 or 10 steps, and/or af- ter a predefined time, e.g., specified as a number of, e.g., 3, 4, 5, 6, 8, 10, 15 or 20, cycle periods of an underlying mCR pattern. At least during such a predefined time, there may occur no changes to the mCR pattern used, e.g., the stimuli provided by a same stimulation unit may be identical in at least repetition frequency and/or shape. Additionally or alternatively, during such a predefined time, an order in which the stimulation units are activated may stay the same, but may be different after such a predefined time. As discussed with reference to Fig.5, a variation may be such that the repetition frequency eventually returns to an initial value, e.g., the repeti- tion frequency may, starting from a lower limit of a respective range as just given, increase a
number of times/steps, e.g., up to an upper limit of the respective range, and may then decrease a same number of times/(respectively equally sized) steps. The present invention may be implemented or performed with hardware, e.g., an apparatus as per the claims, and/or software, e.g., a computer program as per the claims. It may be imple- mented or performed with a general-purpose processor, a digital signal processor (DSP), an ap- plication specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device (PLD), discrete gate or transistor logic, discrete hardware compo- nents, or any combination thereof designed to perform the functions described herein. A gen- eral-purpose processor may be a microprocessor, but in the alternative, the processor may be any commercially available processor, controller, microcontroller, or state machine. A processor may also be implemented as a combination of computing devices, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunc- tion with a DSP core, or any other such configuration. If implemented in software, the functions may be stored or transmitted over as one or more instructions or code on a computer-readable medium. Software shall be construed broadly to mean instructions, data, or any combination thereof, whether referred to as software, firmware, middleware, microcode, hardware descrip- tion language, or otherwise. Computer-readable media include both computer storage media and communication media including any medium that facilitates transfer of a computer pro- gram from one place to another. A computer-readable medium may be coupled to a processor such that the processor can read information from, and write information to, the medium. In the alternative, the medium may be integral to the processor. Examples of computer-readable media may include, by way of example, RAM (Random Access Memory), flash memory, ROM (Read Only Memory), PROM (Programmable Read-Only Memory), EPROM (Erasable Program- mable Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), registers, magnetic disks, optical disks, hard drives, or any other suitable storage me- dium, or any combination thereof. Also, any connection is properly termed a computer-readable medium. For example, if the software is transmitted from a website, server, or other remote source using a coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wire- less technologies such as Bluetooth, infrared (IR), radio, and microwave, then the coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as Bluetooth, infrared, radio, and microwave are included in the definition of medium.
In particular, means for individually activating and means for determining at least one parame- ter based on a measurement of neuronal activity as described herein may be implemented or performed with hardware and/or software, as just described. Means for performing a measurement of neuronal activity as described herein may comprise an apparatus and/or device for functional magnetic resonance imaging (fMRI), magnetoenceph- alography (MEG), local field potentials (LFP) and electroencephalography (EEG).