Systems and methods for cranial fingerprint and face detection guided self registration of a head worn ultrasound neuromodulation device

Abstract

The present disclosure relates to systems, devices, and methods for determining the position of a head worn neuromodulation device relative to features of a user's face and head. The systems, devices, and methods may capture an image with a camera to detect a relative position of a feature of the head worn neuromodulation device and a feature of the user's face or head. Then, an acoustic wave propagating through or reflecting off of the user's head from a transmitting element may be captured for association with the relative positions of the head worn neuromodulation device and the feature of the user's face or head. The association between the acoustic wave and the relative positions can be used to derive the position of the feature of the head worn neuromodulation device relative to the feature of the user's face or head using only the acoustic wave.

Description

SYSTEMS AND METHODS FOR CRANIAL FINGERPRINT AND FACE DETECTION GUIDED SELF REGISTRATION OF A HEAD WORN ULTRASOUND NEUROMODULATION DEVICECROSS-REFERENCE TO RELATED APPLICATION

[0001] This application claims the benefit of U.S. Provisional Patent Application No. 63 / 662,958, filed June 21, 2024, which is incorporated by reference herein in its entirety.FIELD

[0002] The present disclosure generally relates to devices, and associated systems, methods, and uses for registering a head worn neuromodulation device relative to cranial features.BACKGROUND

[0003] Focused ultrasound (FUS) is recognized as a potent tool for non-invasive modulation of intact brain circuits. Unlike pharmacologic methods or other existing non-invasive brain stimulation approaches, it can be used to target specific, deep regions of the brain with high spatial and temporal precision. Many existing brain stimulation approaches are invasive, requiring implantation of stimulative electrodes into target regions of the brain to provide therapeutic effects, while non-invasive techniques are unable to target deeper subcortical regions.

[0004] “Ultrasound” may be defined as an acoustic wave, emission, or signal with a frequency of approximately 18 kHz or greater.

[0005] FUS can be used to modulate neural activity through the manipulation of ion channels and membrane capacitance. Specifically, mechanisms such pressure derived strain, lipid raft disruption, membrane kinetic energy change, cavitation, or radiation force can be sensed by various mechanically gated channels such as channels belonging to the Transient Receptor Potential channels, inward rectifying potassium channels, or piezo channels, among others. Modulation of these channels can lead to either increases or decreases in neural activity depending on the neural target and ultrasound parameters employed.

[0006] FUS focusing may occur with numerous wave fronts propagating across highly heterogeneous tissue. For example, a wave may pass through cortical bone, trabecular bone, bone marrow, red blood cells, neuronal sheathing, and cerebrospinal fluid, allbefore it reaches its intended target. Since all of these tissue types have different characteristic speed of sound, the time it takes for a wave to traverse from its source to its intended target may be dramatically different than another wave sourced over another part of the skull. Thus, accurate knowledge of both the position of the wave sources, as well as the acoustic characteristics of tissue between the source and the brain target facilitate proper ultrasound focusing.

[0007] Information about acoustic properties across the cranium can be effectively derived from a computed tomography (CT) scan, or variants of a magnetic resonance imaging (MRI) scan such as a zero-echo time (ZTE) or ultra-short echo time (UTE) scan. The position of ultrasound emitting sources, or ultrasound emitting elements, can be calculated by identifying the position of fiducial markings spatially registered to the ultrasound emitting elements. Together, this information allows for ultrasound focusing so long as the position of the ultrasound sources remains accurate. However, inadvertent physical displacement of the ultrasound elements could dramatically alter the ultrasound focal target within the brain or distribute the focal pressure across a wider area. Either of these outcomes could potentially reduce the clinical efficacy of a treatment or produce unintended side effects related to off target regions of the brain. As a result, FUS neuromodulation devices benefit from real time self-registration and position monitoring. There is a need for technology which can assess positional accuracy and correct for any changes in the position during actual use.

[0008] Ultrasound signals carry meaningful signals related to tissue morphology, particularly with imaging probes where high frequency linear arrays are used with very short pulses. Thus, ultrasound itself may serve as a useful tool for identifying position relative to morphology. It is possible to use high frequency ultrasound scans to localize anatomical features of the skull surface for registration to prior captured cranial imaging data. The cranial imaging data can then be used for skull-related phase aberration correction and focused ultrasound treatment targeting. Because lowering ultrasound frequency substantially decreases the axial and lateral imaging resolution, the method may utilize a dedicated high-frequency ultrasound probe for high resolution image capture required for cranial imaging registration. In particular, transducers used to image the skull surface often operate above 10 MHz, while therapeutic transducers are typically designed to operate between 250-750 kHz. Similarly, ultrasound images can be used to approximate the position of a probe or an associated medical device.

[0009] Therapeutic ultrasound transmission through the skull often utilizes very low operating frequencies between 250 and 750 KHz. Although these low frequencies transmit through the skull much better than high frequencies, they are not typically used in ultrasound imaging due to wavelength limitations on image resolution. Additionally, the high internal and external reflectivity of the skull may cause multiple layers of waves to return to the transducer receiver at time intervals which do not correlate with distance from their origin. Thus, images cannot be reasonably reconstructed from low frequency cranial ultrasound. Despite this limitation, signals related to the cranial tissue can still be transmitted and received using low frequency ultrasound. While these patterns may not be easily, or even possibly reverse-engineered to obtain spatial morphology of the cranium, it is still possible that they can be reproduced across time. Simply put, a snapshot of the signals should remain stable assuming that the position of the transducer relative to the cranial tissue has not changed. In this case, the signals can be thought of as ultrasound "fingerprints" which can be used to encode unseeable cranial features associated with the fingerprint. Much like physical fingerprints that can be uniquely associated with individuals, ultrasound fingerprints can be associated with certain positions. This association may arise from a database that maps each fingerprint to a corresponding transducer position. With this database, these unique ultrasound signals can be used to identify the face and device coordinate system even without active input from an external camera. Furthermore, dramatic changes in the pattern even with subtle movements may still have non-obvious characteristic changes indicative of placement motion. In addition to the ultrasound signal, other simultaneously captured signals detected from the head worn device, such as electroencephalogram (EEG) measurements, mounted wearable cameras, device strain sensors, infrared sensors, or other propagating waves transmitted through the head may also be included into the fingerprint. A captured fingerprint, in isolation, is simply a unique data signature. Without a corresponding spatial context, it represents a known signature of an unknown position, which limits its utility. This limitation is overcome by creating a map stored in a lookup table that pairs each unique fingerprint with a known device position, the latter being captured by an external system (e.g. a camera). This calibration process enables the device to subsequently determine its position in cranial space in real-time, using only the fingerprint as input.SUMMARY

[0010] According to aspects of the present disclosure, a method for determining relative positions of a head worn neuromodulation device that includes at least one transmitting element and at least one receiving element and features of a user's face or head is given. The method may include capturing an image with a camera to detect a relative position of a feature of the head worn neuromodulation device and a feature of the user's face or head. Then, an acoustic wave from a transmitting element may be reflected off the user’s head to a receiving element may be captured to associate the relative positions of the head worn neuromodulation device and the feature of the user's face or head. The association between the acoustic wave and the relative positions can be used to later derive a position of the head worn neuromodulation device relative to the feature of the user's face or head.

[0011] In another configuration, the acoustic wave may propagate through tissue of the user’s head.

[0012] In yet another configuration, the acoustic wave may be in an ultrasound frequency domain.

[0013] In another aspect of the above, the head worn neuromodulation device may be flexible. As such, the relative positions of at least one feature of the head worn neuromodulation device and the at least one feature of the user’s face or head may not be fixed over time.

[0014] In a further element of the present disclosure, the relative positions of at least one feature of the head worn neuromodulation device and at least one feature of the user's face or head can change during routine use.

[0015] According to another example configuration, the head worn neuromodulation device can be purposefully and iteratively moved to different positions on the head during a calibration phase. The relative positions between at least one feature of the head worn neuromodulation device and the at least one feature of the user's face or head, along with at least one signal from at least one receiving element, can be captured and associated with one another during this calibration phase.

[0016] In yet another aspect of the present disclosure, a lookup table can be created during the calibration phase. This table can associate propagating wave patterns to the relative positions between the at least one feature of the user's face or head and the at least one feature of the head worn ultrasound device.

[0017] According to a further configuration of the above, a mathematical function, computational model, or algorithm can be derived to associate acoustic wave patternsto the relative positions between the at least one feature of the user's face or head and the at least one feature of the head worn neuromodulation device.

[0018] In another aspect of the present disclosure, the derived position of the at least one feature of the head worn neuromodulation device relative to the at least one feature of the user's face or head is used to spatially define a position of the device and its contained elements in magnetic resonance imaging (MRI) or equivalent volumetric image space.

[0019] According to a further configuration of the above, the spatially defined position of the device and its contained elements in MRI or equivalent volumetric image space is used to determine acoustic wave parameters for producing focused ultrasound on at least one brain target.

[0020] In yet another configuration of the present disclosure, the image of the user’s head can be captured using an optical camera and augmented reality for facial feature position detection.

[0021] In another aspect of the present disclosure, a method for determining relative positions of a head worn neuromodulation device that includes at least one mechanical force sensor which detects at least one mechanical state of the head worn neuromodulation device and features of a user's face or head is given. The method may include capturing an image with a camera to detect a relative position of a feature of the head worn neuromodulation device and a feature of the user's face or head. Then, a signal from the mechanical force sensors can be captured to associate the relative positions of the head worn neuromodulation device and the feature of the user's face or head. The association between the mechanical force sensor signal and the relative positions can be used to derive a position of the head worn neuromodulation device relative to the feature of the user's face or head.

[0022] In another aspect of the above, the head worn neuromodulation device may be flexible. As such, the relative positions of at least one feature of the head worn neuromodulation device and the at least one feature of the user’s face or head may not be fixed over time.

[0023] In a further element of the present disclosure, the relative positions of the at least one feature of the head worn neuromodulation device and the at least one feature of the user's face or head can change during routine use.

[0024] According to another example configuration, the head worn neuromodulation device can be purposefully and iteratively moved to different positions on the head during a calibration phase. The relative positions between at least one feature of the head wornneuromodulation device and the at least one feature of the user's face or head, along with at least one signal from at least one receiving element, can be captured and associated with one another during this calibration phase.

[0025] In yet another aspect of the present disclosure, a lookup table can be created during the calibration phase. This table can associate mechanical force sensor signal ensembles to the relative positions between the at least one feature of the user's face or head and the at least one feature of the head worn ultrasound device.

[0026] In yet another aspect of the present disclosure, a method for determining relative positions of a head worn neuromodulation device that includes at least one electromagnetic wave transmitting element and at least one electromagnetic wave receiving element and features of a user's face or head is given. The method may include capturing an image with a camera to detect a relative position of a feature of the head worn neuromodulation device and a feature of the user's face or head. Then, an electromagnetic wave through the user’s head from a transmitting element may be captured to associate the relative positions of the head worn neuromodulation device and the feature of the user's face or head. The association between the electromagnetic wave and the relative positions can be used to derive a position of the head worn neuromodulation device relative to the feature of the user's face or head.

[0027] In a configuration of the above, the electromagnetic can propagate through tissue of the user’s head

[0028] In another configuration of the above, the electromagnetic waves may be visible light waves from a camera configured to detect proximal features of the head.

[0029] According to aspects of the present disclosure, a system comprising a head worn neuromodulation device, a camera, and one or more processors is given. The one or more processors may be configured to capture an image with a camera to detect a relative position of a feature of the head worn neuromodulation device and a feature of the user's face or head. Then, a signal from a propagating wave receiving element may be captured to associate the relative positions of the head worn neuromodulation device and the feature of the user's face or head. The association between the propagating wave and the relative positions can be used to derive a position of the head worn neuromodulation device relative to the feature of the user's face or head.BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The present disclosure can be better understood, by way of example only, with reference to the following drawings. The elements of the drawings are not necessarily to scale relative to each other, emphasis instead being placed upon clearly illustrating the principles of the disclosure. Furthermore, like reference numerals designate corresponding parts throughout the several views.

[0031] FIG. 1 A illustrates a neuromodulation system and a computing environment, according to aspects of the present disclosure.

[0032] FIG. IB illustrates hardware of a neuromodulation device, according to aspects of the present disclosure.

[0033] FIG. 2A illustrates an example key features of a user’s face and head, according to aspects of the present disclosure.

[0034] FIG. 2B shows an example control flow for detecting key features of a user’ s face, head, and device, according to aspects of the present disclosure.

[0035] FIG. 3 A shows an example control flow for a calibration phase of the neuromodulation device, according to aspects of the present disclosure.

[0036] FIG. 3B illustrates the detection of facial features and generation of fingerprints by the neuromodulation system, according to aspects of the present disclosure.

[0037] FIG. 3C illustrates an exemplary generation of fingerprints by the neuromodulation system, according to aspects of the present disclosure.

[0038] FIG. 3D illustrates the generation of mechanical fingerprints by the neuromodulation system, according to aspects of the present disclosure.

[0039] FIG. 4 shows an example control flow for calculating phase delays based on fingerprints, according to aspects of the present disclosure.

[0040] FIG. 5 is a diagram of an example control flow for calculating phase delays based on fingerprints, according to aspects of the present disclosure.

[0041] FIG. 6 shows a control flow of a neuromodulation device, according to aspects of the present disclosure.DETAILED DESCRIPTION

[0042] The present disclosure is related to systems, devices, and methods for providing focused ultrasound to a target brain region of a user with a head worn neuromodulation device. Embodiments consistent with the present disclosure address the need for devices, systems, and methods for modulating the connectivity of the default mode network(DMN) of a user based on the application of focused ultrasound to a target brain region. The target brain region can include a portion of the thalamus, and more particularly, the anterior thalamus (ANT). Based on an application of focused ultrasound to the target brain region, the disclosed devices, systems, and methods are configured to modulate the connectivity of the default mode network. According to some aspects of the present disclosure, modulating the connectivity of the default mode network increases presenttense experiential engagement by decreasing frequency or duration of non-present focused thinking. According to some aspects of the present disclosure, modulating the connectivity of the default mode network enhances attention during tasks. According to some aspects of the present disclosure, modulating the connectivity of the default mode network results in a reduction of ruminative thinking. According to some aspects of the present disclosure, modulating the connectivity of the default mode network can result in a reduction in symptoms of major depressive disorder, obsessive-compulsive disorder, anxiety, post-traumatic stress disorder, attention deficit hyperactivity disorder, autism spectrum disorder, sleep disorder, or any combination thereof.

[0043] The neuromodulation system can comprise a wearable neuromodulation device integrated with EEG electrodes and one or more integrated ultrasound transducer arrays. The disclosed neuromodulation system can further include a stimulation control unit comprising one or more processors, and software that operates and controls the features and functionality of the ultrasound stimulation when executed by such processors. Such software includes, without limitation, EEG real-time analysis software that can continuously monitor brain functionality to identify one or more certain characteristics, phases or states of brain activity, and brain mapping software that can plot one or more specific region of the brain and accurately focus or steer ultrasound stimulation to that one or more specific brain regions. The disclosed neuromodulation system also includes a computational device that aids in neuromodulation device operation and data storage of collected information. Thus, a neuromodulation system disclosed herein noninvasively administers ultrasound stimulation in a spatially and temporally controlled manner. As such, a device disclosed herein enables a focus application of ultrasound stimulation to a specified region of the brain that largely excludes surrounding brain tissue.

[0044] The disclosed neuromodulation device can be coupled to brain substructure mapping software that identifies one or more specific regions to be targeted for ultrasound stimulation. In some embodiments, one or more specific regions of the brain areidentified by comparing a brain image scan to a brain atlas which may be publicly available or internally annotated to identify a common coordinate space. “Image” may be defined as a representation of a two- or three-dimensional structure based on a signal captured by a device. The system can use brain image scans, including scans generated by computed tomography (CT) and magnetic resonance imaging (MRI) to determine a position of the one or more transducer elements relative to the temple of a user. Nonlimiting sources of such brain image scans include scans obtained from a user of a neuromodulation device disclosed herein (personalized model customized for a particular user), scans obtained from deidentified individuals through healthcare facilities, or scans obtained from deidentified individuals through registries like the Human Connectome. Brain image scans are registered with a common brain region atlas and image segmentation performed to identify centroids in voxel space of the one or more specific regions to be targeted for ultrasound stimulation. In some embodiments, the target brain region is the thalamus. In some embodiments, the one or more specific regions to be targeted for ultrasound stimulation is a sub-region of the thalamus such as, but not limited to, the anterior thalamus and the dorsomedial thalamus (DMT). In some embodiments, the one or more specific regions to be targeted with ultrasound stimulation is the ventral capsule (VC), the anterior thalamus, the mediodorsal thalamus, the bed nucleus of the stria terminalis (BNST), or any combination thereof.

[0045] In some embodiments, the position of one or more ultrasound transducer elements relative to the temporal window (e.g., temple of a user) can be identified using biometric parameters. The position of the one or more transducers relative to the temporal window (e.g., the temple of a user) may be estimated as a point in 3D space relative to key features of the user’s face and head which have some predictive value for the positioning of the ultrasound transducer elements relative to the temporal window. This biometric may include, but is not limited to, the position of a person’s eyes, ears, eyebrow ridge, nose, mouth, jawline, or other appendage relative to a cranial landmark. This biometric may also include a relative point along a cranial feature axis, such as a fractionally defined mid-point along the forehead, between the ears and eyes, between the corner of the mouth to the base of the ear, or some other combination of cranial features or appendages.

[0046] Once the one or more specific target brain regions are identified, a brain substructure mapping software disclosed herein can identify the coordinate space of each transducerelement within the image data. According to some embodiments, the system can register the initial position of the one or more ultrasound transducer elements relative to the temporal window using one or more key features of the user’s face and head as described above. The system may then accurately calculate the temporal phase offset of ultrasound transducer elements by estimating acoustic temporal path length between the element and the target brain regions of one or more identified locations or by performing a full wave simulation. Initially the software may determine acoustic impedance by employing an algorithm that converts pixels of a brain image scan from the brain modeling database into measurements of acoustic impedance. A brain substructure mapping software may determine the appropriate phase of ultrasound emissions from the one or more ultrasound transducer elements required to effectively apply ultrasound stimulation onto a target brain region. In some embodiments, the required beam steering is determined by modeling simulations of wave equations by estimating the temporal wave path length to the target brain area, accounting for difference in sound speed across skull and tissue as well as wave refraction. The simulation then adjusts the excitation phase delay of each ultrasound transducer element until the wave fronts constructively interfere at the focus. This process may be referred to herein in shorthand as computing and applying a phase change to ultrasound emissions to create an ultrasound focus at a target brain region.

[0047] Once the initial position of the one or more ultrasound transducer elements relative to the temporal window is determined, the neuromodulation device can accurately focus ultrasound emissions from the one or more ultrasound transducer elements to the target brain region. By applying FUS to one or more target brain regions, the disclosed devices, systems, and methods are configured to modulate the connectivity of the default mode network, as described herein.

[0048] Deep brain stimulation (DBS) targeted to deep brain nuclei is an increasingly common treatment option for Parkinson's disease (PD) and has demonstrated long term success in reducing motor pathologies. Although 24% of the PD population is eligible for DBS, only 5% of these candidates are willing to undergo the procedure. Several studies found that the primary reasons for reluctance in the majority included fear of inefficacy, side effects such as deterioration of speech or personality change, and surgical complications. Even neurologists are hesitant when considering candidates for DBS, as they substantially overestimate the risk of surgical complications. In some respect, the imperfect efficacy and side effect profile may warrant these concerns.

[0049] Although PD has been studied extensively, the brain origins of the various pathologies are not always fully understood. Primary candidate targets include subthalamic nuclei of the thalamus, Globus pallidus interim (GPI), and Ventral Intermediate Nucleus of the Thalamus, all of which have deep connectivity to the substantia nigra. Even within these larger brain regions, certain spatial subsets are thought to have varying levels of contribution, evidenced by the need to titrate voltage and spatial range to achieve effects. Despite numerous clinical studies comparing the benefits of each target, it is unclear whether any target offers superior clinical outcomes. As a result, the brain target is often selected based on the probability of side effects in a given patient. For instance, the GPi is typically favored over the subthalamic nucleus for patients suffering from depression or dementia. Thus, it is possible that non-optimal target selection is responsible for the -25% of patients who do not experience symptom relief. Furthermore, it is unknown whether stimulating multiple targets might have additive benefits. The lack of certainty in these treatments invites the use of reversible, non- invasive treatment modalities such as focused ultrasound for several purposes.

[0050] Accordingly, embodiments of the present disclosure could perform pre-surgical assessment of multiple target's potential efficacy through iterative focused ultrasound exposure. Such tests could include the Motor Unified Parkinson's Disease Rating Scale Examination. As a secondary correlated measure, automated monitoring of head and hand tremor detected through accelerometry built into embodiments of the present disclosure could also be used. Such an application of this technology could lower the portion of non-responders, increasing both patient and physician confidence in the surgery and the number of candidates opting in. Alternatively, the neuromodulation device could be used temporarily in situations which require rapid relief of motor symptoms. While the application of FUS for Parkinson's patients is clear from a therapeutic standpoint, the practical aspects are far more complicated. These patients are often subject to both whole body tremor and limb rigidity, limiting their ability to perform simple motor tasks such as tying shoelaces or basic use of utensils. Thus, it is unlikely these patients would be able to reliably place a cranially worn device, even with the presence of special fit tools or an aid. Furthermore, repetitive shaking of the head may slowly shift the device over time, irrespective of initial placement error. Elements of the present disclosure may remedy these issues associated with both poor placement and device spatial drift over time and allow for consistently targeted FUS neuromodulation in Parkinson's patients, among other cohorts.

[0051] In another embodiment, focused ultrasound is used to modulate the efficacy of sleep. Numerous brain regions and cell types within those regions have been identified as sleep regulators and can modulate sleep through optogenetic or designer drug targeted manipulation. In another embodiment, the device may be used to temporarily stimulate regions responsible for slow wave rhythms to improve synaptic downscaling and / or metabolite clearance in the brain. Furthermore, metabolic clearance from the brain during sleep may be aided by acoustic streaming effects through targeting of cerebrospinal fluid channels. The use of FUS allows for translation of these findings without genetic manipulation. Polysomnography is a method used to monitor sleep and employs numerous electrodes and sensors attached to a subject's head. While these sensors can typically monitor sleep during a quiescent state, many signals are interrupted frequently during the night through mechanical perturbation from pillows, blankets, limb contact, and the subject's partner. Similar effects may be expected with the neuromodulation device described herein. If left uncorrected, a device would likely experience accumulated focal targeting error throughout the night, lowering or completely abolishing any therapeutic benefits of the device. Thus, the methods described herein are particularly useful for focused ultrasound modulation therapies during sleep.

[0052] In yet another embodiment, physical or task training exercises may benefit from the use of focused ultrasound targeted to arousal or cognitive systems in the brain including the locus coeruleus, the ventral tegmental area, or the prefrontal cortex. The supplemented tasks may involve high levels of mental and physical exertion which could place both mechanical force on the neuromodulation device coupled with increased perspiration. The resulting emitting and receiving element offsets would benefit from the corrective measures described herein.System Overview

[0053] Some implementations of the disclosed technology will be described more fully with reference to the accompanying drawings. This disclosed technology may, however, be embodied in many different forms and should not be construed as limited to the implementations set forth herein. The components described hereinafter as making up various elements of the disclosed technology are intended to be illustrative and not restrictive. Many suitable components that would perform the same or similar functions as components described herein are intended to be embraced within the scope of the disclosed electronic devices and methods.

[0054] In some embodiments, and as shown in FIGS. 1A-B, an exemplary neuromodulation device 110 can include a wearable device housing 120, which can support two array elements 130 each containing a transducer array 140. In some example embodiments, the transducer array 140 may be capable of producing and receiving a propagating wave sent through or reflected from a user’s head. The propagating wave may be an ultrasound wave, another type of acoustic wave, an electromagnetic wave, or any other wave capable of propagating through the user’s head.

[0055] When worn, wearable device housing 120 is configured to encircle the head in a transverse plane that positions the main band along the forehead, temples and back of the head. Wearable device housing 120 provides rigid stereotactic placement of the transducer arrays 140 over the temporal window of the user's head.

[0056] Wearable device housing 120 can include a main band 122, a secondary band 124, and an optional securing strap 126. Main band 122, secondary band 124 and securing strap 126 can be adjustable to facilitate accurate positioning and securing of neuromodulation device 110 to a user's head. Secondary band 124 can be attached to main band 122 via first and secondary band attachment points and configured to extend over the top of the head. First and second band attachment points can be static or configured to allow movement between secondary band 124 and main band 122. Optional securing strap 126 is attached to main band 122 via first and second securing strap attachment points and configured to extend under the chin. First and second securing strap attachment points can be static or configured to allow movement between securing strap 126 and main band 122. In aspects of these embodiments, main band 122 has front and back portions composed of a semi-rigid material and side or temple portions composed of a flexible material, secondary band 124 and a first and second attachment hubs each being composed of a semi-rigid material, and a securing strap being composed of an elastic material. Although the embodiment shown in FIGS.1A-B includes optional securing strap 126, it should be understood that in some embodiments, the optional securing strap 126 is omitted, and in yet other embodiments, optional securing strap 126 is replaced with a band similar to the construction of main band 122 and / or secondary band 124. Similarly, in some embodiments, main band 122 and secondary band 124 may be of unitary construction. In yet other embodiments, wearable device housing 120 can include a single band that stretches from approximately the forehead.

[0057] Neuromodulation device 110 can include one or more transducer arrays 140 contained in a housing attached to main band 122 of wearable device housing 120. The one ormore transducer arrays 140 can be located on the inner surface of main band 122 and configured to interface with a user's head. In some embodiments, a neuromodulation device disclosed herein contains a single transducer array 140 located on the main band. In some embodiments, neuromodulation device 110 contains a single transducer array 140 located on one side of main band 122 positioned at either the left or right temple region of a user above the ears. In some embodiments, neuromodulation device 110 contains a single transducer array located on each side of main band 122 positioned at the left and right temple region of a user above the ears. In some embodiments, neuromodulation device 110 contains multiple transducer arrays 140 located on each side of main band 122 positioned at the left and right temple region of a user above the ears. In aspects of these embodiments, and as shown in FIG. 1A-B, neuromodulation device 110 comprises two transducer arrays 140 one located on the left side of main band 122 and one located on the right side of main band 122. In aspects of these embodiments, neuromodulation device 110 comprises two transducer arrays 140 located on the left side of main band 122 and two transducer arrays 140 located on the right side of main band 122. It should be understood that while each transducer array 140 is shown having a plurality of emitting and receiving elements 142, in some embodiments, each transducer array 140 can comprise a single emitting and receiving element 142. In some embodiments, the neuromodulation device 110 can optionally include photoplethysmography (PPG) sensors (not shown). In some embodiments, the neuromodulation device 110 can include one or more accelerometers (not shown) that may be used to capture head movement of a user wearing the neuromodulation device 110.

[0058] According to some embodiments, the emitting and receiving elements 142 can be designed to interface with the “temporal window”, a thin portion of skull bone posterior to the eyes that allows access to centralized deep brain structures. In some embodiments, the neuromodulation system 100 determines beam steering parameters that are unique to each person’s brain and skull morphology that are used to accurately target certain brain regions with FUS. In some embodiments, the neuromodulation system 100 utilizes a combination of custom automated MRI scan segmentation, transducer spatial mapping, and acoustic simulation tools to optimize off-line targeting of brain regions.

[0059] In some example embodiments, the neuromodulation system 100 may include a mobile device 260 such as a smartphone, tablet, or handheld computer. Other computingdevices may also be used in place of mobile device 260. The neuromodulation system 100 may also include one or more cameras, such as camera 262. The cameras may be located on or part of the mobile device 260. The camera 262 may be a device that captures information about a scene by detecting electromagnetic radiation, typically visible light, infrared, or near-infrared, and converting it into a data representation. This includes traditional cameras, which form two-dimensional images based on intensity and color, as well as active or passive depth-sensing systems, such as LiDAR, structured light, or time-of-flight sensors, which measure spatial information to generate depth maps or three-dimensional reconstructions of the environment In FIG. 1 A, the camera 262 is the camera of the mobile device 260. The mobile device 260 may be configured to obtain imaging, video, or other data associated with the neuromodulation device 110. This may include video and imaging data of the neuromodulation device 110, video and imaging data of the neuromodulation device 110 placed on a user’s head, video and imaging data of the user’s face and head, and video and imaging data of the neuromodulation device 110 moving in three- dimensional (3D) space.

[0060] According to some embodiments, the neuromodulation device 110 may be communicatively coupled to the mobile device 260. This may be done through a Wi-Fi connection, a BLUETOOTH® connection, a Near-Field Communications (NFC) connection, a physical connection over a protocol such as Universal Serial Bus (USB), or another coupling protocol.

[0061] According to some embodiments, the head worn neuromodulation device 110 may include one or more fiducial elements 150. The fiducial elements 150 may be markers or other elements mounted on flat-plane mounting areas of the neuromodulation device 110. Signals emitted from, reflected by, or otherwise transmitted by or from the fiducial elements 150 may be captured by a camera to determine a position of at least a portion of the fiducial elements 150.

[0062] The fiducial elements 150 may be located on or within the neuromodulation device 110 at known, replicable coordinates on each device to ensure consistent spatial relationships between the fiducial elements 150 and their associated device. This may facilitate real-time tracking of the position and orientation of the neuromodulation device 110. By determining the position of the fiducial elements 150 based on input information, such as visual or acoustic position information, and based on the known position of the fiducial elements 150 relative to the other components of theneuromodulation device 110, it is possible to determine six degrees of freedom (6-DOF) pose estimation for each detected fiducial element 150, enabling the tracking of both position and orientation of the marker, and thus the headset components, in real-time.

[0063] One exemplary embodiment of the fiducial elements 150 employs a set of infrared reflective markers placed either directly on the secondary band 124, the main band 122, the securing strap 126, or over a portion of the array elements 130. These infrared reflective markers may include one or more infrared elements that may reflect, emit, transmit, or receive infrared light.

[0064] In another exemplary embodiment of the fiducial elements 150, the fiducial elements 150 may be fiducial markers or other optical markers. These fiducial markers may be stickers, decals, embossed designs, images, or other designs disposed at various points on the secondary band 124, the main band 122, the securing strap 126, or a portion of the array elements 130. For example, the fiducial elements 150 can be a triangular shape located on the array elements 130 of the neuromodulation device 110, as shown in FIG. 1A. Other visual designs, such as other shapes, QR codes, ArUco markers, or other patterns can be used as fiducial elements 150.

[0065] In some embodiments, the fiducial elements 150 can be features or landmarks of the neuromodulation device 110 that a sensor and a computational model or algorithm may be configured to recognize. For example, the array elements 130 themselves may serve as the fiducial elements 150. In this embodiment, an optical camera and a computing system configured to execute an algorithmic process may be used to record images or video footage of the neuromodulation device 110. The algorithmic process may be used to recognize the array elements 130 of the neuromodulation device 110 and determine their positions relative to the neuromodulation device 110 and the user’s face and head. As such, the fiducial functions of the fiducial elements 150 could be accomplished by the array elements 130.

[0066] The camera 262 or other electro-optical device with visual access to the markers may be used to determine fiducial points of the neuromodulation device 110 based on the fiducial elements 150 and the user’s head and face. The camera 262 may be a smartphone camera communicatively coupled with a device such as mobile device 260. The camera 262 may provide image data of the neuromodulation device 110. The image data of the neuromodulation device 110 may include 2D video images of the neuromodulation device 110.

[0067] The image data of camera 262 may also include three-dimensional (3D) volumetric information collected by camera 262 or an associated sensor. For example, the camera 262 may include a structured light projector, a time-of-flight sensor, a stereoscopic image sensor array, or any other suitable depth-sensing modality configured to obtain spatial depth data. In some embodiments, the camera 262 generates a dense or semi- dense point cloud dataset corresponding to the physical contours, features, and surface topology of the neuromodulation device 110. The camera 262 may further acquire multispectral or photometric data in conjunction with spatial depth measurements to augment the dataset with surface material properties, reflectance characteristics, or tissue-device contrast enhancements.

[0068] Other example embodiments may use sensors other than camera 262 to help generate 2D data or 3D volumetric information to obtain spatial location data associated with the neuromodulation device 110.

[0069] To further enhance positional awareness and tracking fidelity, for example, the neuromodulation system 100 may incorporate a magnetic tracking system. In such embodiments, one or more magnetic field emitters may be positioned in the environment of the neuromodulation system 100 to establish a spatial reference field. The neuromodulation device 110, or a mounting structure coupled thereto, may be equipped with a magnetic sensor unit configured to detect changes in the emitted magnetic field and derive position and orientation information based on the field vector data.

[0070] In additional or alternative embodiments, the neuromodulation device 110 may include an inertial measurement unit (IMU) comprising one or more gyroscopes, accelerometers, and / or magnetometers. The IMU may be embedded within or affixed to the neuromodulation device 110 and configured to generate motion data indicative of the device’s orientation, acceleration, and translational displacement over time. The IMU data may be used independently or fused with optical point cloud data and / or magnetic tracking data. This hybrid tracking approach enables real-time position tracking of the neuromodulation device, even in environments where optical occlusion or magnetic interference may occur.Facial and Head Feature Key Point Identification and Mesh Alignment

[0071] As shown in FIG. 2 A and FIG. 2B, certain embodiments may provide methods and systems for identifying key points of a user’s face and head and aligning meshes.

[0072] FIG. 2A shows the identification of key points of a user's face and head and of the neuromodulation device 110, according to example embodiments. The method may include receiving an initial scan of at least a portion of the user’s face and head. This initial scan may be a volumetric scan containing 3D volumetric data associated with the user’s face and head.

[0073] For example, a user may first undergo cranial imaging which can approximate acoustic information. This image may be a CT, MRI, or any imaging modality that can serve as an acoustic information source. The CT / MRI image may be performed and received by the neuromodulation system 100 at any time prior to the use of the neuromodulation system 100.

[0074] According to other example embodiments, structured light scanning may be used to capture high-resolution external facial geometry. This technique employs one or more digital projectors that cast light patterns (such as sinusoidal fringes, binary patterns, or Gray code sequences) onto the user's facial surface. Cameras such as camera 262 positioned at known geometric relationships to the projectors capture the deformed light patterns as they conform to the facial topography.

[0075] In another example embodiment, photogrammetric techniques may be used to generate the initial scan. This may involve capturing one or more samples of 2D image data from cameras such as camera 262. The photogrammetric process may utilize feature detection and matching algorithms that identify corresponding points across multiple camera views. For example, structure-from-motion algorithms determine camera poses and sparse 3D point clouds, while multi -view stereo (MVS) techniques densify the point cloud to create detailed surface representations. Other machine learning algorithms may be used to generate the volumetric initial scan data.

[0076] In yet another example embodiment, laser emitting and receiving devices associated with the camera 262 may be used to generate the initial scan data. This embodiment may use one or more laser emitting and receiving devices that sweep across the surface of the user’s face and head while cameras such as camera 262 observe the laser reflections from predetermined angles. In some configurations, laser detection and ranging (LiDAR) sensors can be used to generate the initial scan data.

[0077] Another example embodiment may involve time-of-flight sensing technology. One or more infrared light emitters and sensors may be used. The infrared light devices may emit and receive infrared light across the user’s face and head. The topographical data received may then be used to generate the initial scan data.

[0078] According to further example embodiments, additional processing may be performed to identify internal key features of the user’s face or head. For example, one or more post-processing algorithms may convert the initial scan data into volumetric representations, enabling visualization of soft tissue structures including facial muscles, fat pads, salivary glands, neural pathways, blood vessels, and intracranial anatomy. In some embodiments, this can be combined with propagating waves generated by the neuromodulation device 110.

[0079] In some scenarios, the user may be wearing the neuromodulation device 110 during the generation of the initial scan. In this scenario, volumetric data associated with the neuromodulation device 110 may also be collected by any of the exemplary systems described above, or by other methods capable of generating volumetric data. In other scenarios, a user may put on the neuromodulation device 110 after the initial scan has been completed. In this situation, the initial scan may still identify key facial features of the user’s face and head. An additional update scan may also be performed to generate updated initial scan data associated with the neuromodulation device 110 and the user’s face and head.

[0080] Based on the initial scan, the location of key facial features within those scans would be manually or automatically determined. These positions could include but are not limited to the tip of the nose, the inner and outer edges of each eye, and the nasion, or deepest part of the nose bridge. Other key facial features may be identified. Identification of key features may begin with an initial segmentation step. The initial segmentation step may first isolate facial structures of the user’s face and head. This may be done by a computational model or algorithm configured to recognize and segment sections of the user’s face and head. For example, the segmentation step may segment the initial scan data into an eye segment, a cranial segment, a mouth segment, a nasal segment, and other segments. Based on these initial scan segments, dense 3D representations of the user’s face and head can be generated. In some embodiments, these are dense 3D point clouds of the user’s face and head.

[0081] In some embodiments, key features of the neuromodulation device 110 may also be identified by the methods described herein. For example, key feature 203 and key feature 205 of the neuromodulation device 110 have been identified in FIG. 2 A. These key features may be determined by the locations of fiducial elements 150.

[0082] In some example embodiments, the initial segmentation may also include segmenting the initial scan data into segments associated with the user’ s face and head and segmentsassociated with the neuromodulation device 110. In this way, the initial scan can separate the shape of the neuromodulation device 110 from the shape of the user’s face and head.

[0083] The initial scan data can also be used to generate one or more surface meshes of the segments. For example, a surface mesh generation algorithm may process volumetric data from the initial scan to create continuous surface representations. Poisson reconstruction, Delaunay triangulation, or other techniques may be used to generate high-quality meshes of the user’s face and head and the neuromodulation device 110. Mesh optimization procedures including smoothing, remeshing, and topology correction may also be applied.

[0084] The systems and methods may then process the initial scan data and determine key features of the user’s face and head. The processing may include facial contour mapping. Based on the initial scan data, the surface meshes, and other data, facial contours and key facial landmarks such as the tip of the nose, the inner and outer edges of each eye, and the nasion, or deepest part of the nose bridge can be identified. Further soft tissue profile analysis may incorporate forehead curvature, cheek prominence, nasolabial fold depth, and chin projection measurements. A computational model or algorithm may then be used to process the volumetric data and identify landmarks of the user’s face and head. These landmarks can then be identified and tracked against temporal changes such as changes in pose, position, and morphology, or other changes.

[0085] For example, FIG. 2A shows a scan 200 of the user’s face. In FIG. 2A, the scan 200 may be a point cloud of the shape and contours of the user’s face. As shown, the scan 200 may also include scans taken at multiple angles of the user’s face. Based on this scan 200, one or more segments may be identified. These segments may then be further processed by the neuromodulation system 100 to identify key features of the user’s face and head. As shown by key features 202, 204, and 206, the user’s nose bridge, eye comer, and nose tip have been identified as key features of the user’s face based on the scan 200. Other features may also be identified as key features. “Features of the user's face and head” may also include or be inferred from a proxy marker that is physically or virtually associated with anatomical landmarks. A proxy marker may comprise any marker, device, reflector, tag, or other detectable element that can be coupled to the user's face and head. A proxy marker may be identified using optical, infrared, electromagnetic, or other sensing modalities. For example, key features of the user'sface and head may be determined by the positions of one or more infrared reflectors, fiducial markers, or wearable sensors positioned on or near the face or head.

[0086] Based on these key features, the surface meshes and volumetric data of the user’s face and head and the neuromodulation device 110 may be aligned to indicate a position of the neuromodulation device with respect to the key features identified of the user’s face and head.

[0087] In some example embodiments, this may be done by computing distances between the key features of the user’s face and head and the fiducial points indicated by fiducial elements 150 of the neuromodulation device 110. For example, the distance between key features 202, 204, 206, and fiducial elements 150 could be computed, and based on these distances and the known positions of the fiducial elements 150, positions of the neuromodulation device 110 with respect to the key features 202, 204, and 206 can be determined.

[0088] According to some example embodiments, the surface mesh of the user’ s face and head may be received from a device such as mobile device 260. Mobile device 260 may perform any of the imaging and sensing techniques described above to establish a surface mesh of the user’s face and head. In some configurations, additional cameras, sensors, laser emitters, MRI / CT scans, or other volumetric information may be used by the mobile device 260 to establish the face mesh.

[0089] According to some example embodiments, a calibration phase may be performed. The calibration phase may include the user placing the neuromodulation device 110 on their head with one or more cameras, such as camera 262, actively capturing images of their face and the neuromodulation device 110. The user may then move their head or the neuromodulation device 110 around in 3D space, with the camera 262 or other sensors actively recording 2D or volumetric data associated with the user’s face, head, and the neuromodulation device 110. This can produce additional volumetric data used to determine the position of the neuromodulation device 110 with respect to the key facial features of the user.

[0090] Based on the distances between the key features of the user’ s face and head, and fiducial elements 150 of the neuromodulation device 110, surface meshes, and other volumetric information associated with the neuromodulation device 110, can provide real-time position and face tracking for the neuromodulation device 110 placed on the user’s face and head.

[0091] This alignment process may use the key features of the user’s face and head described previously as an initial starting point. This is followed by an iterative alignment algorithm that translates vertices on the mesh associated with the user’s face and head to minimize the distance between the closest point on the target reference mesh optionally derived from the initial scans or other scans. Once this alignment converges, the transformed face and head mesh is intersected with the wearable device mesh that was automatically extracted from the subject’s initial scans or other scans. This may use algorithmic processes configured to align meshes associated with the user’s face and head and the neuromodulation device 110. In some embodiments, this process may involve identifying intersection points of the face mesh and the device mesh, and further identifying at least some of those points as anchor points. Furthermore, a rigid transformation matrix can be defined to fit these anchor points. Anchor points can be points where the neuromodulation device 110 interacts with, intersects, or is anchored against the user’s face and head. For example, the anchor points may be defined by the array elements 130 or the securing strap 126.

[0092] FIG. 2B shows an example control flow for the mesh alignment techniques described above. In FIG. 2B, step 208 describes using one or more models to extract key features of the user’s face and head and the neuromodulation device 110 from the initial scan data or additional scan data. Furthermore, surface meshes of the user’s face and head and the neuromodulation device 110 may be established. Step 210 describes iteratively aligning the mesh associated with the user’s face and head and the neuromodulation device 110. This can be based on a surface mesh received from a mobile device such as mobile device 260, as shown in step 212 of FIG. 2B. Step 212 also shows how the face mesh can be uploaded to a server running a mobile application or cloud application.

[0093] Once the surface mesh of the user’s face and head and the surface mesh of the neuromodulation device 110 have been aligned in step 210, step 214 may perform an intersection of the aligned face and device associated meshes to determine anchor points. Step 215 shows how these coordinates may be stored pairwise with device coordinates and fingerprints in a lookup table or database. The device positions may be generated by exemplary steps 218, 220, and 220 using fiducial elements 150 and cameras such as camera 262. Based on these intersections, the anchor points, and the known distances between the user’s key facial features, step 216 demonstrates how the neuromodulation device 110 with respect to the user’s face and head may be aligned in real-time. In some example embodiments, a visual depiction or indication of thisalignment may be transmitted to and displayed on mobile device 260, as shown in element 264 of FIG 1 A.Fingerprint Generation

[0094] The neuromodulation device 110 can be used to establish “fingerprints” based on propagating wave patterns through the user’s head and the positional information determined through the techniques described in the present disclosure. Fingerprints may comprise data associated with propagating waves transmitted and received through and reflected off a user’s head while emitting and receiving elements 142 are located at various locations with respect to the user’s face and head.

[0095] One example type of propagating wave that can be transmitted through or reflected off of a user’s face, head, or skull is ultrasound. Ultrasound signals can carry meaningful information related to tissue and skeletal morphology, particularly with imaging probes where high frequency linear arrays are used with very short pulses. Thus, ultrasound itself may serve as a useful tool for identifying position relative to morphology. It is possible to use high frequency ultrasound scans to localize anatomical features of the skull surface for registration to prior initial scan data. The initial scan data can then be used for skull-related phase aberration correction and high intensity focused ultrasound treatment targeting. Because lowering ultrasound frequency substantially decreases the axial and lateral resolution, the method described herein may utilize a dedicated high- frequency ultrasound probe for high resolution image capture required for head image registration. In particular, transducers used to image the skull surface often operate above 10 MHz, while therapeutic transducers are typically designed to operate between 250-750 kHz. Similarly, ultrasound images can be used to approximate the position of a probe or an associated medical device. Obtaining an image of tissue may be difficult at frequencies below 1 MHz, however.

[0096] Given the direct relationship between higher ultrasound frequencies and increased absorption, reflection, and refraction, therapeutic ultrasound transmission through the skull often utilizes very low operating frequencies between 250 and 750 KHz. Signals related to the cranial tissue can still be transmitted and received using low frequency ultrasound. It is possible that such signals can be reproduced across time. There, a snapshot of the ultrasound signals should remain stable assuming the position of the transducer relative to the cranial tissue has not changed. This snapshot of signals associated with a certain positioning of the transducer elements in relation to the user’s face and head can establish a fingerprint. Fingerprints can be used to identify unseeablestructural elements and use them to infer positioning. By way of analogy, an actual fingerprint can be used to identify the human to which they belong, even if the human's face, body, or voice cannot be captured. This is simply because a database has associated that fingerprint with the human. Similarly, these fingerprints can be used to identify the device within an individual’s face and head coordinate system. Furthermore, dramatic changes in the fingerprint pattern, even with subtle movements, may still have non-obvious characteristic changes indicative of placement or motion. In addition to the ultrasound signal, other simultaneously captured signals detected from the head worn device, such as EEG, mounted wearable cameras, or device strain sensors, may also be included into the fingerprint. When combined with indications of the position of the neuromodulation device 110, which can be determined using techniques as described in the present disclosure, real-time tracking of the position and orientation of the neuromodulation device 110 with respect to the cranial space of the user can be established.

[0097] In an example embodiment, low frequency ultrasound pulses can be used to generate fingerprints. Any of the emitting and receiving elements 142 can be ultrasound transducer elements and may emit and / or receive ultrasound. Each emitting and receiving element 142, or any combination thereof, can be pulsed with a single waveform, or multiple waveforms, while all receivers, or a subset of receivers, of the neuromodulation device 110 can listen to the temporal response. The data collected by the receivers represents complex interactions with waves reflecting and refracting within the cranium.

[0098] In another example embodiment, electromagnetic waves may be used to generate fingerprints. Electromagnetic waves — ranging from visible light to near-infrared (NIR) and beyond — readily penetrate and scatter through the soft tissues and superficial bone of the human head. By transmitting a known beam of light or infrared to the skull and measuring the returning intensity and phase (or time-of-flight) at one or more detectors, a unique electromagnetic wave fingerprint of that individual’s cranial anatomy can be captured. Similar principles underlie pulse-oximetry, where NIR LEDs illuminate a fingertip (or earlobe), and photodiodes detect light modulated by arterial blood. Small shifts in sensor position or exact alignment relative to local vasculature beneath the probe can result in measurable changes in the AC and DC components of the received waveform, making the signal highly sensitive to a given position on the head.

[0099] In one example embodiment, a visible-light camera and / or depth-sensing camera such as LiDAR can be used to detect salient superficial facial features such as curves of the nose, eyes and / or mouth, or subtle features such as texture and color gradients that vary with anatomical structure and blood perfusion. In another embodiment, multiple illumination wavelengths spanning the visible to NIR spectrum can be used in sequence or combination to form a composite multispectral image, where the pattern of attenuation across wavelengths is sensitive to underlying bone thickness, soft tissue layers, and vascular structure. Multiple wavelengths in the NIR spectra may be used to detect different tissue properties with varying absorption. In another embodiment, the emitted light can be polarized, and reflectance is measured at one or more angles to extract polarization-resolved subsurface scattering features, which shift as a function of device placement. These signals can be collected using onboard cameras, photodiodes, or photonic sensors integrated into the head worn device and may be processed locally or remotely to generate a placement-specific electromagnetic signature that correlates with the position of the device relative to the cranial anatomy.

[0100] According to some example embodiments, a calibration phase may be established. In the calibration phase, a database consisting of pairwise fingerprints and the device coordinates is established. FIG. 3A depicts a portion of such a calibration phase.

[0101] In FIG. 3 A, while actively tracking the user's facial and device features with an external camera such as camera 262, the user is instructed to move the device on their head in a specified manner, as shown in position diagrams 301, 303, 305, and 307. Other tracking devices may also be used. In one embodiment, arrows on an application overlaying the neuromodulation device 110 features on a camera feed displayed on mobile device 260 provide these instructions. In another embodiment, a mesh depicting positions that the user needs to move the neuromodulation device 110 to may be overlaid on a camera feed displayed on a mobile device 260. In yet another embodiment, the user may be instructed to move the neuromodulation device 110 around randomly or according to their own direction on their head for a period of time.

[0102] In some example embodiments, a search space 311 may be defined by the neuromodulation system 100. The search space 311 may determine one or more suggestion vectors that indicate areas of the user’s face and head that the neuromodulation device 110 needs to be moved to. Based on these suggestion vectors, the arrows displayed on the application displayed on mobile device 260 may be determined.

[0103] At specified intervals, both the ultrasound fingerprint and facial and device features are captured and saved to a database. The database may be stored on a storage of the neuromodulation device 110, on a storage of the mobile device 260, or on a storage of another computing system communicatively coupled to the neuromodulation system 100. In FIG. 3 A, two of these example positions at exemplary intervals are shown. As seen in orientation one, fingerprint 302 may be taken at a certain position on the user’s head, associated with a position 304 of certain device key features with respect to the user’s head. The distance between the position 304 of the device key features and example facial feature 202 may then be stored in the database.

[0104] In orientation 2, a similar process can be repeated. The user, potentially guided by the neuromodulation system 100 or the mobile device 260, can move the neuromodulation device 110 on their head so that the key features of the device are located at position 308 with respect to the user’s head. At position 308, the neuromodulation device 110 may use a propagating wave pulse to then generate fingerprint 306. Fingerprint 306, the position 308 of the key features of the neuromodulation device 110, and the position of facial feature 206 can then be stored in the database. Each fingerprinting interval may also record one or more facial features or device key feature positions. Likewise, at each interval, multiple fingerprints can be measured.

[0105] FIG. 3B shows an example diagram of the pairwise database based on the embodiment shown in FIG. 3 A. In orientation 1, the first coordinates 314 associated with the positions of the facial features and key features of the neuromodulation device 110 may be recorded. For example, the position of the device features at orientation 1 and the key facial feature 202 would be recorded in coordinates 314. Likewise, at orientation 2, the second coordinates 316 may be collected and associated with the positions of the facial features and key features of the neuromodulation device 110. For example, the position 308 of the device features at orientation 2 and the key facial feature 204 and 206 would be recorded in coordinates 314. In each orientation, the associated fingerprint, such as the fingerprint 302 and fingerprint 306, may be recorded as a pair with their associated coordinates. This database may then be used to map the propagating wave receive signals to the face and device positions.

[0106] In some example embodiments, the calibration phase may involve collecting propagating wave data across many discrete locations on the user’s head to create a unified coordinate system for the user’s head and the neuromodulation device 110. For example, one exemplary embodiment can collect propagating wave data at 2,601discrete locations across a 51 * 51 spatial coordinate grid with 0.1 mm resolution in the X-Z plane. At each coordinate position, the neuromodulation device 110 may execute a sequence of transmit-receive events employing a predetermined set of transmit element combinations, where each combination pairs one element from the left array element 130 with one element from the right array element 130. For each transmit event, all 128 elements can simultaneously record echo signals over a 200-microsecond acquisition window, digitized at 2 MHz sampling frequency.

[0107] A time-to-peak value for each emitting and receiving element 142, defined as the temporal point within a 20-80 microsecond post-transmit analysis window where the signal amplitude reached its maximum value, can be computed. This temporal window can isolate near-field acoustic reflections originating from the skull's temporal acoustic windows. Each spatial coordinate can be characterized by a 128*n matrix of time-to- peak values. By vectorizing these matrices into single-dimensional feature arrays, a unique spatiotemporal fingerprint corresponding to each grid coordinate may be generated.

[0108] To compute the relationships between collected fingerprints, a correlation matrix illustrating pairwise similarity across a 5x5 mm grid region along the X-Z plane can be used. This matrix demonstrates the correlation of each fingerprint within the measurement grid relative to a central reference fingerprint in the center of the grid. Fingerprints collected proximal to this central reference exhibited elevated Pearson correlation coefficients, indicating high similarity, with correlation values decreasing as spatial distance from the central reference coordinate increased.

[0109] The Pearson correlation coefficient can be used to quantify similarity between fingerprints. The K-nearest neighbors, defined as the K-fingerprints exhibiting the highest Pearson correlation coefficients, can be used to generate interpolation weights (w,) using a Gaussian radial basis function.

[0110] where d, (equal to 1-p,) represents a dissimilarity measure derived from the Pearson correlation coefficient for the z-th nearest neighbor fingerprint, and d represents the mean of the top-K corresponding dissimilarity measures. A predicted coordinate x for a fingerprint can be calculated as a weighted average of the corresponding coordinates of the top-K nearest neighbors:Si WiXt X = -Si wt

[0111] where the summation index z iterates over the K-nearest neighbors, w, represents the interpolation weight for the z-th nearest neighbor, and x, denotes the known spatial coordinate of the z-th nearest neighbor fingerprint. This weighted averaging computation produces a predicted coordinate x that represents the estimated position of the current fingerprint based on the spatial locations of its most similar training fingerprints. This computational approach enables continuous position estimation between coarsely sampled grid coordinates by prioritizing neighbors with similar propagating wave fingerprint profiles.

[0112] FIG. 3C shows depictions of signals gathered from propagating waves during an experiment of example embodiments of the present disclosure. In subject depiction 319, a depiction of a subject wearing the neuromodulation device 110 is shown. While wearing the neuromodulation device 110, a fingerprint at a certain coordinate value 321 may be taken. In the experiment, at each designated spatial coordinate such as coordinate value 321, the neuromodulation system 100 performed a predetermined sequence of transmit-receive operations, shown here as three sequential events using predefined transmit element combinations. Each combination consisted of one element from the left transducer array paired with one from the right transducer array. Echo signals were simultaneously recorded across all 128 receiving elements. For each receiving element, a time-to-peak metric representing the temporal point within a 20- 80 microsecond post-transmit window where the received signal achieved maximum amplitude was calculated. The collection of time-to-peak values generated a locationspecific spatiotemporal fingerprint.

[0113] FIG. 3C demonstrates the temporal relationship of the fingerprint at this coordinate value 321 between time TX1 320 (marking the initiation of an emitting and receiving element 142 pair) and time RX1 322 (indicating the corresponding receive period for this emitting and receiving element 142 pair). The Y-axis shows the full array 318 of various emitting and receiving elements 142 of the neuromodulation device 110. By vectorizing the matrices generated by each element over the temporal window into single-dimensional feature arrays, as described above, a unique spatio-temporal fingerprint responding to each grid coordinate can be created.

[0114] Plot 324 shows a correlation matrix that demonstrates pairwise similarity relationships between fingerprints across a 5x5 mm spatial region. The correlation of each signature within the measurement grid relative to a designated central reference signature (marked with a white circle) was quantified. Fingerprints demonstrated elevated similarity coefficients at proximal positions, with correlation values diminishing as spatial separation from the central reference position increased. This shows the interpolatable nature of fingerprints.

[0115] In graphs 325, position prediction performance is presented as a function of system acquisition parameters. The left panel shows mean localization error relative to grid sampling density variations. The right panel presents the percentage of test coordinates achieving sub-0.5 mm accuracy as a function of the number of transmit element pairs incorporated into each spatiotemporal signature

[0116] Histogram 327 presents the distribution of localization errors across the complete test dataset (0.1mm coordinate spacing) when utilizing a weighted K-nearest neighbors algorithm trained on signatures collected at sparse 1 mm coordinate intervals. All position predictions achieved sub-millimeter precision, with mean error values substantially below 0.5 mm, demonstrating effective performance even with reduced training data density.

[0117] In one example embodiment, the fingerprinting methods described above enable realtime inference of device-deformation states — such as inward and outward flexion — by sampling distributed joint-angle measurements. Each joint angle may be captured by a strain sensor that bridges adjacent segments, with wiring or wireless telemetry embedded in the neuromodulation device 110 to convey the strain values to the controller.

[0118] In another embodiment, the same device-position fingerprint can be derived from any deformation cue sensed in the elastomeric body, including, without limitation: piezoresistive strain networks (carbon-black, CNT, graphene, MXene or liquid-metal microchannels whose resistance varies with elongation); capacitive stretch sensors (parallel-plate or interdigitated elastomer capacitors whose capacitance rises with area change); piezoelectric pressure, force or bend sensors (PVDF, PZT or nanofiber films that generate charge under dynamic stress); triboelectric nanogenerator sensors that harvest contact-separation charge; magnetic or magneto-elastic sensors (rubber loaded with hard or soft magnetic particles read by Hall-effect, GMR or AMR chips); inductive coils whose inductance or quality factor shifts with geometry or proximity; opticalstrain sensors such as polymer or silica fibre-Bragg gratings and elastomeric waveguides reporting wavelength- or intensity-shifts; fluidic or pneumatic microbladders or channels whose internal pressure or impedance tracks deformation; ionicconductor hydrogel sensors whose impedance changes with strain; ultrasonic or acoustic time-of-flight transducers that map local thickness or curvature; thermal or conductivity-gap sensors that infer deformation from heat-flow or contact-resistance changes; embedded joint goniometers; or any combination thereof. Triboelectric and piezoelectric elements can provide self-powered, high-bandwidth updates for impacts and vibrations, while piezoresi stive, capacitive and optical channels furnish stable readings of slower postural changes, allowing hybrid architectures that monitor the full dynamic range of wearable deformation.

[0119] In FIG. 3D, an articulated mechanical structure 326 consisting of a series of rotational joints 328 embedded within the neuromodulation device 110 is shown. Each joint 328 is equipped with a miniature strain or flexion gauge that passively reports the angular displacement between adjacent segments of the neuromodulation device 110. These sensors can provide analog or digital readings of relative joint angles, measured in degrees, radians, or another measurement, at each articulation point. The device structure allows controlled flexion along multiple degrees of freedom, enabling the device to conform to individual head shapes or respond to external forces, such as inward pressure at the temporal window.

[0120] As shown in element 350 of FIG. 3D, mechanical deformation patterns under different loading conditions may be characterized by constructing a custom armature of the headset composed of a network of bones connected by rotational joints such as the rotational joints 328. Each joint 328 incorporates a strain gauge that measures angular displacement. By configuring various joint angles, three discrete device poses may be defined: a neutral Rest Pose 329, an Inward Flex Pose 330 (simulating compression at the temporal windows), and an Outward Flex Pose 340 (simulating lateral expansion). The Rest Pose 329 serves as the baseline configuration, exhibiting minimal or zero strain across all joints. The Inward Flex Pose 330 (Figure 1C, Top) and Outward Flex Pose 340 (Figure ID, Top) can be induced by applying symmetric forces to the lateral wings of the headset. Each pose produces a pattern of joint angle changes across the articulated structure.

[0121] For each pose, a mechanical fingerprint may be defined based on the angles measured at each joint 328. For example, the mechanical fingerprint may be defined as the vectorof joint angle differentials (A Joint Angle) relative to the Rest Pose baseline. These fingerprints are extracted from the embedded strain gauges and represented as fixed- length feature vectors indexed by joint identifiers (e.g., L1-L41, R1-R41). Graph 360 and graph 370 display bar plots of A Joint Angle for the Inward and Outward Flex Poses, illustrating pose-specific deformation signatures.

[0122] Each mechanical fingerprint may be associated with a corresponding spatial position and orientation of the device, measured relative to key features of the user’s face and head. These fingerprint-position pairs serve as calibration entries that can be stored in a lookup table. After collecting one or more calibration samples, inference is performed by comparing a new fingerprint to all entries in the table. A V-Nearest Neighbors (KNN) algorithm may be used to identify the most similar fingerprints, and the predicted device position can be computed as the weighted average of the positions corresponding to the top K matches.Phase Delay Determination

[0123] Based on the database of pairwise coordinates and fingerprints, the position of all of the emitting and receiving elements 142 of the neuromodulation device 110 can be determined. This may be done by computing the known positions of the emitting and receiving elements 142 with respect to the neuromodulation device 110. For example, it may be known where each of the emitting and receiving elements 142 are placed with respect to the fiducial elements 150 of the neuromodulation device 110. As the positions of the fiducial elements 150 can be determined by the techniques illustrated in the present disclosure, the position of the emitting and receiving elements may then be approximated.

[0124] In some example embodiments, the flexing or other deformation of the neuromodulation device 110 when it is placed on the user’s head may change the position of at least a portion of the emitting and receiving elements 142. Various methods may be used to account for this flex.

[0125] In one example embodiment, a computational model or algorithm may be used to estimate a flex or a mechanical fingerprint of the neuromodulation device based on volumetric and positional data collected by the camera 262, other cameras, sensors, and the fingerprinting techniques described above. Based on the angles, reflectance, and relative position of the fiducial elements 150 of the neuromodulation device 110, an estimated mechanical fingerprint may be determined. The computational model or algorithm may be configured to estimate the mechanical fingerprint of theneuromodulation device 110 based on the known properties of the materials that make up the neuromodulation device 110 and the positions of the fiducial elements 150. In other embodiments, the relative positions of the key features of the user’s face and head and the fiducial elements 150 may be compared to previous measurements of these elements’ positions, allowing for the computational model or algorithm to determine the mechanical fingerprint of the neuromodulation device 110.

[0126] As described in the present disclosure and shown in FIG. 3D, a mechanical fingerprint may be determined based on the position of angle sensors included in one or more joints 328 of the neuromodulation device 110. The mechanical fingerprint may further be used to determine a position of every emitting and receiving element 142 of the device.

[0127] Based on the positions of the emitting and receiving elements 142 of the neuromodulation device 110, phase delays may be computed for a therapeutic propagating wave application by the emitting and receiving elements 142. Various types of acoustic or other propagating wave simulations may be performed. Some examples include but are not limited to: the finite difference time domain, hybrid angular spectrum, pseudospectral, finite differences, boundary element, and ray tracing methods of simulation.

[0128] Based on the simulation of propagating wave propagation through the user’s head, the propagating wave phase delay may be computed to focus at least a portion of the propagating waves on a target region of the user’s head. The computation of the phase delay may be based on the acoustic simulations performed by the neuromodulation device 110. The phase delays may also be calculated for the position of the neuromodulation device 110 and stored in a lookup table or database. When at the position, the neuromodulation device 110 may then use the stored phase delay.

[0129] In some example embodiments, the phase delay may also be calculated by a time reversal of the received propagating waves of the simulation. By placing a wave generating source at the target region during the simulation and treating the original elements as sensors, time reversal may be performed to generate the appropriate phase delay for the emitting and receiving elements 142.

[0130] FIG. 4 shows an exemplary control flow for the phase delay calculation process. In step 402, one or more fingerprints may be generated by the methods described in the present disclosure. In step 404, the one or more fingerprints can be used to determine positions within an associated coordinate system, including the key features of the device and the user’s face and head. The associated coordinate system may also include positionalinformation of the neuromodulation device 110, the array elements 130, and the emitting and receiving elements 142. By mapping the key features of the user’s face and head to the initial scan data, the coordinates of the emitting and receiving elements 142 can be obtained within the initial scan’s associated coordinate system, as shown in step 406. Step 408 then shows how an acoustic simulation or calculation is used to determine the phase delays needed to target a specific area in the brain. The phase delays are transmitted into the device. The phase delays may then be used by the neuromodulation device 110 to generate propagating waves that target the intended area of the user’s brain using the emitting and receiving elements 142, as shown in step 410. In some embodiments, the entire process may last on the order of milliseconds.

[0131] Given the computational complexity of the fingerprint acquisition and subsequent simulations, it may be beneficial to avoid repeating such calculations. For instance, if a patient is lying still on their back, the device is unlikely to move relative to the cranium. Thus, a repeated calculation of updated phases may be returning the same solution while blocking the computational resources from performing alternative functions. To limit computation to periods when it is beneficial, the device may implement a closed loop detection algorithm which only performs the signature when the device detects motion of the emitting and receiving elements 142 elements relative to the user’s face and head. This motion detection could include but is not limited to changes in thresholded accelerometry data or EEG impedance values. In another embodiment, the motion of the emitting and receiving elements 142 relative to the cranium may be detected by monitoring electrical impedance of a circuit containing both the coupling medium and the patient skin. This coupling medium could be a coupling pad, ultrasound gel, a faceplate of at least one of the emitting and receiving elements 142, or the emitting and receiving elements 142 themselves.

[0132] In one embodiment, an electrical contact is placed in contact with the neuromodulation device 110 and an additional electrical lead placed somewhere on the patient's skin. During wear, the neuromodulation system 100 may monitor changes in capacitance which exceed a given threshold at which point the system can trigger the processes described herein. The system can also be used to alert the user when the contact between the neuromodulation device 110 and the user’ s skin is insufficient for treatment through use of a tone, light, or other sensory stimuli.Exemplary Control Flow

[0133] FIG. 5 shows a flowchart of an example method 500 for determining relative positions of a head worn neuromodulation device, according to certain embodiments of the present disclosure.

[0134] Step 502 of method 500 includes capturing data to detect relative positions of at least one feature of the head worn neuromodulation device and at least one feature of the user's face or head. The data may be image data captured by a camera such as camera 262. The data may also include volumetric data collected by the camera 262, an initial scan, or by other sensors associated with the neuromodulation system 100. Based on the data, relative positions of features of the neuromodulation device and features of the user’s face can be computed by any of the techniques described above.

[0135] Step 504 of method 500 includes capturing a signal from a propagating wave receiving element and associating the captured signal with the detected relative position of the at least one feature of the head worn neuromodulation device 110 and at least one feature of the user’s face or head. The propagating wave may be an ultrasound wave, an electromagnetic wave, another type of acoustic wave, or any other sort of propagating wave. In some configurations, the propagating wave may be emitted and received from one or more transducer elements of the neuromodulation device. For example, the emitting and receiving elements 142 of the neuromodulation device 110 may be the transducers producing and capturing the propagating wave. Based on these propagating waves, the relative positions of the neuromodulation device and the features of the user’s face or head can be determined. For example, fingerprints associated with the propagating waves may be collected and stored in a database or table alongside coordinates of the relative positions of key features of the neuromodulation device and the user’s face or head.

[0136] Step 506 of method 500 includes using the association between the captured signal and the relative positions to derive a position of the head worn neuromodulation device 110 relative to the at least one feature of the user's face or head. For example, the fingerprintcoordinate pairs of the database may be used to establish a coordinate system of the device on the user’s head. Further positional information may be used to establish the relative positions of features of the neuromodulation device to features of the user’s face or head. In other example embodiments, initial scan data and volumetric data associated with the camera 262 may be used to establish surface meshes of the neuromodulation device and the user’s face and head. Based at least in part on the signals and data associated with the propagating wave, the meshes may be aligned andused to establish the relative positions of the neuromodulation device and features of the user’s face and head.

[0137] In FIG. 6, an example control flow 600 is shown. The example control flow 600 may begin with cranial mapping 602. In cranial mapping 602, an initial scan of the user’s head may be taken. This may be an MRI / CT, or another volumetric imaging method. Based on the cranial mapping, known properties of tissue, and other information gathered from the neuromodulation device 110, an acoustic / anatomical spatial map of the user’s head may be generated.

[0138] The device position detection calibration phase 604 shows an example workflow for determining the position, orientation, and location of the neuromodulation device 110 relative to the user’s face and head. The calibration phase may include the user placing the device on their head and moving it according to a predetermined pattern of movements. The user may also randomly move the device, or move it in a pattern that they determine. During the calibration phase, the neuromodulation device may determine one or more fingerprints using mechanical, acoustic, or electromagnetic signals.

[0139] The position of the neuromodulation device 110 may also be determined using fiducial elements 150, cameras such as camera 262, and other techniques described in the present disclosure. The calibration phase may further include associating the fingerprint data with the positional data of the neuromodulation device in a lookup table or database.

[0140] During usage of the neuromodulation device in the device usage step 606, the lookup table may be used in conjunction with a “black-box” function, such as a trained machine learning model, that is configured to estimate the position of the device with respect to the user’s face and head. The acoustic simulations are then used to determine phase delays to be used to target a region of the user’s brain with target propagating waves such as focused ultrasound.

[0141] It should be noted that processing of data and algorithms described herein may be performed by system components implemented in hardware or a combination of hardware and software (see exemplary description of components in FIGS. 1 A-1C). As an example, such a system component may include at least one processor, such as a digital signal processor (DSP) or central processing unit (CPU), configured to execute instructions stored in memory for performing the functions described herein. In some embodiments, application-specific integrated circuits (ASICs) or gate arrays, such asfield-programmable gate arrays (FPGAs), may be used to implement any of the functions described herein. Various configurations of circuitry for the processing of data and algorithms described here are possible.

[0142] The foregoing is merely illustrative of the principles of this disclosure and various modifications may be made by those skilled in the art without departing from the scope of this disclosure. The above described embodiments are presented for purposes of illustration and not of limitation. The present disclosure also can take many forms other than those explicitly described herein. Accordingly, it is emphasized that this disclosure is not limited to the explicitly disclosed methods, systems, and apparatuses, but is intended to include variations to and modifications thereof, which are within the spirit of the following claims.

[0143] As a further example, variations of apparatus or process parameters (e.g., dimensions, configurations, components, process step order, etc.) may be made to further optimize the provided structures, devices, and methods, as shown and described herein. In any event, the structures and devices, as well as the associated methods, described herein have many applications. Therefore, the disclosed subject matter should not be limited to any single embodiment described herein, but rather should be construed in breadth and scope in accordance with the appended claims.

[0144] The terms “about” and “approximately” shall generally mean an acceptable degree of error or variation for the quantity measured given the nature or precision of the measurements. Typical, exemplary degrees of error or variation are within 20 percent (%), preferably within 10%, more preferably within 5%, and still more preferably within 1% of a given value or range of values. Numerical quantities given in this description are approximate unless stated otherwise, meaning that the term “about” or “approximately” can be inferred when not expressly stated.

[0145] With reference to the use of the word(s) “comprise,” “comprises,” and “comprising” in the foregoing description and / or in the following claims, unless the context requires otherwise, those words are used on the basis and clear understanding that they are to be interpreted inclusively, rather than exclusively, and that each of those words is to be so interpreted in construing the foregoing description and / or the following claims.

[0146] The term “including” should be interpreted to mean “including but not limited to...” unless the context clearly indicates otherwise.

[0147] The term “consisting essentially of’ means that, in addition to the recited elements, what is claimed may also contain other elements (steps, structures, ingredients,components, etc.) that do not adversely affect the operability of what is claimed for its intended purpose. Such addition of other elements that do not adversely affect the operability of what is claimed for its intended purpose would not constitute a material change in the basic and novel characteristics of what is claimed.

[0148] The term “adapted to” means designed or configured to accomplish the specified objective, not simply able to be made to accomplish the specified objective.

[0149] The term “capable of’ means able to be made to accomplish the specified objective.

[0150] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting. As used herein, the singular forms “a”, “an” and “the” are intended to include the plural forms as well (i.e. “at least one”), unless the context clearly indicates otherwise.

[0151] The terms “first”, “second”, and the like are used herein to describe various features or elements, but these features or elements should not be limited by these terms. These terms are only used to distinguish one feature or element from another feature or element. Thus, a first feature or element discussed below could be termed a second feature or element, and similarly, a second feature or element discussed below could be termed a first feature or element without departing from the teachings of the present disclosure.

[0152] Terms such as “at least one of A and B” should be understood to mean “only A, only B, or both A and B.” The same construction should be applied to longer list (e.g., “at least one of A, B, and C”).

Claims

CLAIMSWhat is claimed is:

1. A method for determining relative positions of a head worn neuromodulation device to at least one feature of a user's face or head, wherein the head worn neuromodulation device includes at least one acoustic wave emitting element and at least one acoustic wave receiving element, the method comprising: capturing at least one image with a camera to detect the relative position of at least one feature of the head worn neuromodulation device and at least one feature of the user's face or head; capturing a signal from the at least one acoustic wave receiving element and associating the captured signal with the detected relative position of the at least one feature of the head worn neuromodulation device and the at least one feature of the user's face or head; and using the captured signal and the association between the signal and the relative position of the head worn neuromodulation device and the at least one feature of the user’s face or head to derive a position of the head worn neuromodulation device relative to the at least one feature of the user's face or head.

2. The method of claim 1, wherein the acoustic wave propagates through tissue of the user’s head.

3. The method of claim 1, wherein the acoustic wave is in an ultrasound frequency domain.

4. The method of claim 1, wherein the head worn neuromodulation device is flexible, and the relative positions of the at least one feature of the head worn neuromodulation device and the at least one feature of the user’s face or head are not fixed over time.

5. The method of claim 1, wherein the relative positions of the at least one feature of the head worn neuromodulation device and the at least one feature of the user's face or head change during routine use.

6. The method of claim 1, wherein the head worn neuromodulation device is purposefully and iteratively moved to different positions on the head during a calibration phase, and the relative positions between the at least one feature of the head worn neuromodulation device and the at least one feature of the user's face or head, along with the at least one signal from the at least one receiving element, are captured and associated with one another.

7. The method of claim 6, wherein a lookup table is created during the calibration phase to associate propagating wave patterns to the relative positions between the at least one feature of the user's face or head and the at least one feature of the head worn neuromodulation device.

8. The method of claim 6, wherein a mathematical function or model is derived to associate acoustic wave patterns to the relative positions between the at least one feature of the user's face or head and the at least one feature of the head worn neuromodulation device.

9. The method of claim 1, wherein the derived position of the head worn neuromodulation device relative to the at least one feature of the user's face or head is used to spatially define a position of the device and its contained elements in magnetic resonance imaging (MRI) or equivalent volumetric image space.

10. The method of claim 9, wherein the spatially defined position of the device and its contained elements in MRI or equivalent volumetric image space is used to determine acoustic wave parameters for producing focused ultrasound on at least one brain target.

11. The method of claim 1, wherein the image of the user’s head is captured using an optical camera and augmented reality for facial feature position detection.

12. A method for determining relative positions of a head worn neuromodulation device to at least one feature of a user's face or head, wherein the head worn neuromodulation device includes at least one mechanical force sensor, the method comprising: capturing at least one image with a camera to detect the relative position of at least one feature of the head worn neuromodulation device and at least one feature of the user's face or head;capturing a signal from the at least one mechanical force sensor and associating the captured signal with the detected relative position of the at least one feature of the head worn neuromodulation device and the at least one feature of the user's face or head; and using the captured signal and the association between the captured signal and the relative position of the head worn neuromodulation device and the at least one feature of the user’s face or head to derive a position of the head worn neuromodulation device relative to the at least one feature of the user's face or head.

13. The method of claim 12, wherein the at least one mechanical force sensor signal is associated with at least one mechanical configuration state of the head worn neuromodulation device.

14. The method of claim 12, wherein the relative positions of the at least one feature of the head worn neuromodulation device and the at least one feature of the user's face or head change during routine use.

15. The method of claim 12, wherein the head worn neuromodulation device is purposefully and iteratively moved to different positions on the head during a calibration phase, and the relative positions between the at least one feature of the head worn neuromodulation device and the at least one feature of the user's face or head, along with at least one signal from the at least one mechanical force sensor, are captured and associated with one another.

16. The method of claim 15, wherein a lookup table is created during the calibration phase to associate mechanical force sensor signal patterns to the relative positions between the at least one feature of the user's face or head and the at least one feature of the head worn neuromodulation device.

17. A method for determining relative positions of a head worn neuromodulation device to at least one feature of a user's face or head, wherein the head worn neuromodulation device includes at least one electromagnetic wave emitting element and at least one electromagnetic wave receiving element, the method comprising:capturing at least one image with a camera to detect the relative position of at least one feature of the head worn neuromodulation device and at least one feature of the user's face or head; capturing a signal from the at least one electromagnetic wave receiving element and associating the captured signal with the detected relative position of the at least one feature of the head worn neuromodulation device and the at least one feature of the user's face or head; and using the signal and the association between the signal and the relative position of the head worn neuromodulation device and to the at least one feature of face or head to derive a position of the head worn neuromodulation device relative to the at least one feature of the user's face or head.

18. The method of claim 17, wherein the electromagnetic wave propagates through tissue of the user’s head.

19. The method of claim 17, wherein the electromagnetic waves are visible light waves from a camera configured to detect proximal features of the face.

20. A system comprising: a head worn neuromodulation device including at least one propagating wave emitting element and at least one propagating wave receiving element; a camera; one or more processors; and a memory, the memory having stored thereon instructions to cause the one or more processors to: capture at least one image with the camera to detect a relative position of at least one feature of the head worn neuromodulation device and at least one feature of a user's face or head; capture a signal from the at least one propagating wave receiving element and associate the captured signal with the detected relative positions of the at least one feature of the head worn neuromodulation device and the at least one feature of the user's face or head; and use the captured signal and the association between the captured signal and the relative positions to derive a position of the head wornneuromodulation device relative to the at least one feature of the user's face or head.

Images