Determining the structural characteristics of an object

Abstract

The present invention relates generally to a system and method for measuring structural features of an object. The object is subjected to an energy application process to provide an objective, quantitative measurement of the structural features of the object. The system may include a device, e.g., an impact instrument, that is reproducibly positionable relative to the object undergoing such measurement for reproducible positioning. The invention provides a system and method for analyzing the measured features using machine learning to create a system for predicting a pathology from the measurements.

Description

[Technical field]

[0001] [CROSS REFERENCE TO RELATED APPLICATIONS] This Patent Cooperation Treaty International Patent Application claims the benefit of and priority to the following United States Provisional Patent Applications: Serial Nos. 63 / 313,405, 63 / 313,407 and 63 / 313,409, the contents of which are hereby incorporated by reference in their entireties.

[0002] [Copyright information] A portion of the disclosure of this patent document contains material that is subject to copyright protection. The copyright owner has no objection to the facsimile reproduction by anyone of the patent document or the patent disclosure, as it appears in the U.S. Patent and Trademark Office patent file or records, but otherwise reserves any and all copyright rights whatsoever.

[0003] [Field of the Invention] The present invention relates generally to the assessment of structural properties of an object. In particular, the present invention relates to assisted assessment of structural features reflective of the integrity of an object using the application of controlled energy onto the object, and to using machine learning to facilitate the assessment of structural features reflective of the integrity of an object using the application of controlled energy onto the object. [Background technology]

[0004] Any object, anatomical or non-anatomical, including industrial or mechanical, e.g., any structure, exhibits some kind of structural features that may change over time when used in any way, including when simply left in place in the environment. For changes that are easily discernible visually or revealed through simple inspection, measurement of the changes can be easily performed. However, when such changes are not easily discernible visually or revealed through simple inspection, more complex inspection is required. Inspection to find such changes is important for the health and life of the structure, since such changes may eventually develop into the form of unrepairable defects over time if left unidentified or untreated. Several methods can be used to determine the characteristics of the structure, but when such changes are internal, most of the inspections are often destructive or invasive.

[0005] When an object is subjected to an impact force, stress waves are propagated through the object. These stress waves cause deformations in the internal structure of the object. As the object deforms, it acts, in part, as a shock absorber, dissipating some of the mechanical energy associated with the impact. The ability of an object to dissipate mechanical energy, commonly referred to as the object's "damping capacity," depends on several factors, including the type of material the object is made of and its structural integrity.

[0006] Instruments exist that can measure the damping of an object. One example of such an instrument is described in U.S. Patent 6,120,466 ("the '466 patent"). The instrument disclosed in the '466 patent provides an objective, quantitative measurement of an object's damping, referred to as the loss factor. Elastic wave energy can decay relatively slowly in a material with a relatively low loss factor, while elastic wave energy can decay relatively quickly in a material with a relatively high loss factor. Summary of the Invention

[0007] The present invention relates to a system and method for measuring and evaluating structural features of an object, whether anatomical or non-anatomical, in a non-invasive manner and / or using a non-destructive method of measurement. The structural features of an object can be identified based on measurements of the same or other objects made previously and captured by the system using the system. The system can include: applying energy to an object, which can be an anatomical or mechanical object; measuring a response, such as energy reflected from the object as a result of the application of energy, such as percussing the object, or a response, such as deceleration information of an energy application tool, such as a percussion rod, such as energy return, during a time interval; recording or compiling such measurements for analysis by a computing system; and a device capable of creating a response profile, such as an energy return curve or energy return graph (ERG), force return graph (FRG), displacement return graph (FRG) or another physical return value, as a time profile or as a frequency profile, to evaluate the characteristics of the object being measured. The response can generally be generated from measurements of force, energy, displacement or other physical return values ​​on a sensing mechanism or element, for example, over a period of time.

[0008] In general, the system may include a program logic module that is trained on a large data set of waveforms to arrive at an optimized decomposition of the waveforms. In an exemplary embodiment of the invention, the waveforms may be measured signals, such as ERG, FRG, DRG or other physical return value signals, or simulations of such signals, which may generally be grouped together based on at least one common characteristic. The program logic module may then apply an algorithm to form an initial guess (e.g., waveform) of the decomposition of the signal into its component sub-signals, perform an optimization to minimize the difference between the initial guess decomposition and the original signal, identify and address potential errors or defects in the decomposition, and perform subsequent rounds of guess decomposition and optimization to form an optimized decomposition. The optimized decomposition, its characteristics, and the method used to arrive at the optimized decomposition may be incorporated into the system by codifying algorithms, such as in machine learning or deep learning algorithms, so that the system can more efficiently and accurately decompose new signals encountered, such as those obtained from a crash measurement device that generates signals from physical objects.

[0009] In some embodiments, the object may be a real, artificial or simulated oral tissue, such as a tooth, an intraoral restoration, an appliance, an implant or splint, and / or associated tissue or an orthopedic implant. In general, the impact measurement device may apply mechanical energy to the object by percussion and measure the force / energy / displacement etc. returned to the impact measurement device, such as by measuring the force, energy, displacement or other physical return value at a sensing mechanism over a period of time. Measurements collected may also be deceleration information from the energy application tool after the energy application process.

[0010] In other embodiments, the object may be a mechanical, industrial structure, or composite that may include, but is not limited to, polymeric or metallic composite structures including honeycomb or layered honeycomb; aircraft airframes, automobiles, ships, bridges, tunnels, trains, buildings, power generation equipment, arch structures, or other similar physical structures.

[0011] Generally, some sensing mechanisms may be essentially limited to detecting a response signal in a single direction, such as only in the direction of the pressure of the piezoelectric force sensor (i.e., because the piezoelectric element typically generates a signal simply in response to the pressure). This may generally result in at least some loss of signal from the waveform or other energy profile if the amplitude includes a negative portion relative to some criteria, such as any response from the object that results in the object vibrating away from the sensing mechanism, such as application of impact-initiating energy to the object, or if mechanical contact or connection is lost, the sensing mechanism may not detect part of the signal due to the absence of pressure in the appropriate direction detected by the sensing mechanism. Thus, the ERG generated by the device may be partially incomplete.

[0012] For an object, e.g., a tooth or mechanical structure, in an initial state or without additional perturbation from an initial state, e.g., an intact canine or incisor with one root, a response profile from the object during a time interval or a signal including deceleration information of an energy application tool, e.g., ERG, FRG, DRG, etc., may generally take the shape of a Gaussian with one peak, which may resemble the upper half of a sine wave (i.e., a sinusoid or sinusoid-like shape). However, for a non-initial object, e.g., a tooth, with a crack or other defect, the response profile may be different and may vary. For example, an object with various degrees of perturbation may change, e.g., from a Gaussian with one peak to multiple peaks or other deviations from a Gaussian. In some instances, the various responses may provide good insight into the type of perturbation, which may be correlated with the type of defect, the location of the defect, etc. In other instances, some of the response signals may drown out some of the perturbations, and additional techniques of analysis may need to be used, such as decomposing the collected actual response signals into more basic subsignals, to evaluate the type of perturbation to obtain knowledge and additional information about the true structural features. For sine waves, there are many known mathematical formulas that can be used to help with the solution. However, as mentioned above, a Gaussian shape may simply resemble the top half of a sine wave, making it difficult to use these ready-made mathematical formulas to obtain information. Also, the impact system may simply produce a Gaussian curve, which is essentially a sine wave without a bottom. In some instances, additional processes may also be needed to help find the simulated missing bottom half.

[0013] The inventors have discovered using artificial intelligence that a signal can be decomposed into a series of one or more sub-signals. These sub-signals often correspond directly to vibration and / or resonant frequencies induced in the defect during impact. Furthermore, the characteristics of these sub-signals, such as frequency, amplitude, and exponential decay rate, can refine or uniquely identify certain physical and / or clinical phenomena.

[0014] In one exemplary aspect of the invention, the system may be adapted to receive or analyze signals generated from impact measurements on an object where the sensing mechanism results in a limitation or loss of at least a portion of the return signal.

[0015] In some exemplary embodiments, the system may use machine learning methods to process and / or analyze signals that essentially miss negative amplitude portions, and attempt to reconstruct or process the signal as having the missing portion of the response rather than as a complete response. In the dental field, machine learning algorithms and methods may be trained on a large set of collected impingement data (e.g., time-energy profiles) from a wide variety of teeth with different characteristics, e.g., different types (e.g., incisors, premolars, canines, molars, etc.), sizes, number of roots, different degrees of physical damage (e.g., fractures, cavities, etc.), degree or type of restoration (e.g., crowns, fillings, etc.), age, etc., to train the algorithms and methods to be able to decompose newly encountered signals into component sub-signals, e.g., a collection of sinusoidal sub-signals that form the signal or an approximation thereof.

[0016] The data set may also be grouped in other ways, for example, by the location of the impacts on the object (e.g., relative to the teeth, such as cheek or mesial, distal or proximal), by the location of the object relative to other reference points (e.g., mandible vs. maxilla in the oral cavity), by the amount / frequency / number of impacts, or by any other suitable type of grouping. In such an embodiment, the signal may be analyzed and processed as at least one sinusoidal signal with a portion of the signal missing from the signal (e.g., the lower half of the sinusoid below a given threshold is missing from the signal, forming a Gaussian-like shape within the measured time frame). This may result in a better interpretation of the data, since the entire response (or at least an approximation that accounts for the missing portion of the signal) may then be considered, rather than treating the response solely as a portion present in the signal (i.e., assuming the signal is complete without the missing portion).

[0017] In another aspect of the invention, a machine learning algorithm or set thereof, which may generally be separate or distinct from the machine learning algorithms discussed in conjunction with signal decomposition or other uses, may be trained on a large set of collected signals that may be annotated with features (which may be determined by an "expert" or other trusted characterizer or via a machine learning algorithm) to recognize the measured signals, to infer with some probability the measured signals, and / or to associate the measured signals with specific features, or to select appropriate methods of further analysis or algorithmic manipulation to produce a useful output for a user to utilize or interpret, e.g., to select an appropriate regimen for further diagnosis, monitoring, treatment, etc. Such machine learning algorithms may be used, for example, in analyzing or finding correlations between clinical signals and annotated datasets, for example, in heat maps of numerical metrics or other methods of comparison or analysis.

[0018] In an exemplary embodiment of the invention, the signal may be recognized, analyzed, and / or processed as a sum or conglomeration of multiple different sub-signals generated by the interaction of the energy applied to the tooth with the structural features of the tooth and / or surrounding tissues / structures. Without being bound to a particular theory, the signal may generally represent a number of different sub-signals generated by separate physical structures or features of or around the tooth, respectively, and may generally be an approximation of a sinusoid, such as, by way of example and without limitation, a decaying sinusoid (e.g., an exponentially decaying sinusoid in response to dissipation due to material / structure / movement of an object), resulting in a sum or conglomeration of sinusoids to form the overall shape of the signal (or the portion that is detected, i.e., without missing or "bottom" portions below a given threshold for detection or measurement). In some embodiments, without being bound to a particular theory, the signal may be decomposed into a "sparse" or limited number of sinusoids (i.e., a bounded number of sinusoids to give the original signal or an approximation thereof) to identify that each sinusoid is caused by at least one element of the tooth, its restoration if any, and / or surrounding tissue.

[0019] In some embodiments, it may be understood that the pronounced Gaussian-like shape in the signal from the tooth, particularly from a pristine or intact tooth, may generally result from the response of the PDL (periodontal ligament) absorbing the kinetic energy from the energy application tool. For an intact healthy tooth, the impact energy generated by chewing is attenuated by the PDL at the interface between the healthy bone and the natural tooth. Even if the PDL may be absent or defective, the portion of the signal referred to as the PDL portion will refer to the portion of the signal resulting from the fixation or implant of the tooth directly or indirectly within the bone, which produces a roughly pendulum-like response. In general, the PDL portion of the signal may constitute a large portion of the amplitude of the signal, and may generally be interpreted as a carrier wave that may be used to isolate and / or separate the response from other elements of the tooth or surrounding tissue, for example, by subtracting the PDL portion from the signal.

[0020] In some embodiments, the PDL signal may simply be a dominant signal that is generated and is not specifically related to a PDL.

[0021] In other embodiments, for example, the impact measurement device may perform measurements on a mechanical device or other object, such as a dental implant, industrial equipment or device, etc., and the significant signal may be approximately, for example, from the fixation of the device to its surroundings to produce a pendulum response.

[0022] In some exemplary embodiments, sinusoidal decomposition of a signal using machine learning methods may generally include: thickening the signal (e.g., to remove signal distortions close to the x-axis of the signal, e.g., distortions due to an energy application tool sticking to the teeth); finding an initial guess for the larger and / or more pronounced sinusoidal components of the signal (e.g., sinusoids typically generated by a PDL); finding an initial guess for the remaining sinusoidal components of the signal (e.g., from cracks, damage, separations between layers, or other features) including a guess for frequency (e.g., via a Fourier transform-like operation such as a Fast Fourier Transform (FFT) on the signal after subtracting the initial PDL sinusoidal guess); performing an optimization to minimize the difference between the initial signal and the resulting sum of the initial guess sinusoids to produce a candidate sinusoidal decomposition; identifying / fixing any decomposition defects or errors (e.g., unlikely or negatively indicated decomposition results) and returning to the decomposition steps described above as necessary to remove them; and cherry-picking the best or other desired candidates from the resulting decomposition. The resulting decomposition and associated data / results / visualizations may then be displayed or output in a human readable format, etc., so that a practitioner or other user may use or interpret them for clinical diagnosis, monitoring and / or treatment planning, etc. The decomposition generated by the system may also generally include the determination or calculation of uncertainty measures at various steps of the decomposition, for example, to calculate the value of the error at various steps or of a particular calculation in the decomposition.

[0023] In another exemplary aspect of the invention, the system may generate or calculate various numerical metrics from the decomposition of signals that show statistically significant differences in the data set such that these metrics may be used in probability distributions or heat maps to aid in the prediction or detection of various physical features or attributes by comparison with the numerical metrics derived from the decomposition in a clinical setting. For example, a numerical metric generated from a data set containing a known physical feature (e.g., dental damage type) may be used to compare using the same type of numerical metric derived from clinical measurements (e.g., from clinical signals or physical parameters from clinical measurements), and the probability or degree of match may be determined by the system to output the likelihood of a match with that particular physical feature.

[0024] The comparison may also be performed using machine learning algorithms to help increase efficiency and the probability of matching known data sets. For example, basic machine learning methods such as kernel density estimation and / or calibration curve fitting Bayesian networks may be used. More advanced comparison methods may also be generated using deep learning methods after the machine learning system has been fully developed and / or trained.

[0025] In some embodiments, the system may detect numerical metrics in the clinical measurement that may instruct or suggest to a user to modify some physical parameters of the clinical measurement, such as by modifying parameters of an impact measurement device, to generate better or more accurate data. For example, some numerical metrics may instruct or suggest the use of different impact forces, frequencies, or locations on the object of impacts to reveal additional information or improve the quality of the measurement.

[0026] Generally, when natural teeth are replaced with implants due to injury or disease, the ligaments are generally lost. However, the systems and methods of the present invention can also be used to evaluate the structural characteristics of implant structures using abutments. Some materials used for abutments, such as composites, gold, and zirconia, can produce side signals that are somewhat similar to the PDL response.

[0027] Additionally, the systems and methods as described above and below may be useful for measuring dynamic response when forces are applied to the abutment material, and may also be useful for predicting fit or compatibility prior to implantation, or for selecting appropriate materials to protect natural teeth adjacent to the implant, and for making better material selections to minimize differences between how implants and natural teeth react to impacts.

[0028] In some aspects, the invention relates to a system for compiling inspection results from multiple objects, which may or may not include inspection results of objects inspected over a period of time. In some embodiments, each inspection result may be generated using an instrument having a housing with an open or closed end along with an energy application tool that may apply energy to the object to generate a response, e.g., an impact response that may reveal structural features of the object without substantially affecting existing structural features of the object.

[0029] For example, the device (i.e., the impact measurement device) may include a housing having an open end and a longitudinal axis including an energy application tool mounted inside the housing for movement from a stationary configuration to an activated configuration using a drive mechanism supported inside the housing. The housing may include an object contacting portion at its open end, or a sleeve may protrude a distance from the open end of the housing and include an object contacting portion at its open end adapted to rest the device on at least a portion of an object and to activate the drive mechanism and thus the energy application tool to impact the object when the object contacting portion of the housing or sleeve is at rest on at least a portion of the object and to measure the response after the impact to generate a response versus time curve or graph. The response may be captured by a computer coupled to the device, and the response versus time curve.

[0030] The system and method may include a device having an energy application tool, e.g., a percussion tool, that may apply energy to an object to generate a response, e.g., an impact response that may reveal structural features of the object without substantially affecting the object's existing structural features. The energy application tool may be programmed to impact the object a predetermined number of times per minute at substantially the same speed for a certain time interval during the inspection. The system may measure the impact response, e.g., energy reflected from the object as a result of the application of energy, e.g., by percussion or applying energy, or deceleration information of the energy application tool, i.e., energy return, for a certain time interval. The responses may be fed to a computer and the information recorded or compiled for analysis by the system, which may include creating an impact response profile, e.g., ERG, FRG, DRG, etc., or a frequency-based profile of such, e.g., based on energy / force reflected from the object during a time interval, and / or evaluating the impact response profile, e.g., a time response profile, etc., to determine the structural characteristics of the object, e.g., vibration damping capacity; acoustic damping capacity; defects including, e.g., defects inherent in the bone structure or material that makes up the object; cracks; microcracks; fractures; microfractures; loss of cement seal; cement failure; adhesion failure; microleaks; lesions; cavities; the structural characteristics, general structural integrity, or general structural stability of the substrate or environment to which the object may be fixed or within which it may reside.

[0031] The system of the invention may, for example, include a device for performing an impact action on an object. Impact measurement devices useful in the invention may come in different configurations, and inspection results produced from some configurations may produce better models than others. In general, the device includes an impact instrument that can be reproducibly placed directly on the object to be measured for reproducible measurements.

[0032] In some exemplary embodiments of the invention, the device used may include a housing having an open end and a hollow interior through which energy may be applied by an energy application tool, including any tool capable of applying any type of energy to an object, including mechanical, sonic, or electromagnetic energy. According to some embodiments, a tool capable of applying mechanical energy to an object, such as a percussion rod or impact rod, positioned or mounted inside the housing, passes through to reach the object to be measured. According to some other embodiments, a source of electromagnetic energy of any frequency, such as light energy, may be positioned inside the housing, for example. According to further examples, a source of sonic energy, such as an ultrasonic transducer, or any acoustic energy source may be positioned inside the housing.

[0033] The device of the present invention may be, for example, an impact instrument, which may include a handpiece having a housing with a longitudinal axis with an open end and an energy application tool, such as a percussion or impact bar, mounted inside the housing for axial movement along the longitudinal axis of the housing or for vibrational movement relative to the longitudinal axis of the housing. In some embodiments, the housing may include an object contacting portion that may be reproducibly placed in contact with an object to be measured. In some other embodiments, the housing may include at least a portion, such as a sleeve portion extending a distance from the housing, that may be reproducibly placed in contact with an object to be measured. The energy application tool, such as a percussion bar, may have a length and may be located inside the housing and may be programmed to impact the object at substantially the same speed a predetermined number of times per minute, and the deceleration information of the tool or the object's response from the impact may be recorded or compiled for analysis by the system. In some embodiments, the device and hardware may communicate via a wired connection. In some other embodiments, the device and hardware may communicate via a wireless connection.

[0034] Systems and methods useful for performing and acquiring measurements on an object may include devices having energy application tools that may apply energy to the object to generate measurements. For example, impact measurement devices may be useful in the present invention and may come in different configurations and forms, e.g., desktop or portable devices such as handheld devices, and inspection results produced from some configurations may generate better models than other configurations.

[0035] In some embodiments, the energy application tool, e.g., a percussion bar, has a length in a retracted or resting form or configuration and an extended or activated form or configuration, the retracted form being retracted from or substantially coextensive with the open end of the housing when the energy application tool is a percussion bar. Movement of the energy application tool, e.g., a percussion bar, may be accomplished by a drive mechanism mounted inside the housing for axially driving the percussion bar within the housing between the retracted and extended positions described above during operation. In the extended position, the free end of the percussion bar is capable of extending or protruding from the open end of the housing. In some other embodiments, the energy application tool, such as a percussion rod, may be in a resting form or configuration with the tip of the tool substantially perpendicular to the longitudinal axis of the housing, and may be moved to an activated form or configuration where the energy application tool forms an acute angle with the longitudinal axis of the housing, while the tip of the energy application tool remains substantially perpendicular to the longitudinal axis of the housing by rocking back and forth about a pivot point on the longitudinal axis. In other words, the energy application tool may oscillate from a position substantially parallel to the longitudinal axis of the housing to a position at which the pivot point forms an acute angle with the longitudinal axis of the housing. The energy application tool may be held either horizontally or in other positions during measurement, with the tip position substantially perpendicular to the main portion of the tool, and maintain a constant length whether at rest or during impact. Movement of the energy application tool, such as a percussion rod, may be accomplished by a drive mechanism mounted inside the housing for driving the percussion rod from a position substantially parallel to the longitudinal axis of the housing to a position at a pivot point at an acute angle with said axis, with the tip oscillating up and down in sequence. Movement of the energy application tool, such as a percussion rod, may be accomplished by a drive mechanism mounted inside the housing for driving the energy application tool.

[0036] The driving mechanism may be, for example, an electromagnetic mechanism and may include an electromagnetic coil. In some embodiments, the driving mechanism may include a permanent magnet fixed to the rear end of the energy application tool, for example a percussion rod, and the magnetic coil may lie axially behind the permanent magnet. The magnetic coil together with the rear of the handpiece housing and any power supply lines form a structural unit that may be integrally operable and may be connected to the remaining device, for example, by a suitable removable connection, for example a screw-type connection or a plug-type connection. This removable connection may facilitate cleaning, repair, etc. In some other embodiments, the driving mechanism may be an electromagnetic mechanism and may include an electromagnetic coil and a permanent magnet fixed to the rear end of the energy application tool, for example a percussion rod, by an interface, for example a coil mount. The coil, for example an electromagnetic coil, may lie axially behind the permanent magnet. The electromagnetic coil may also act directly on a metallic or conductive component, such as a ferromagnetic component. Other types of linear motors may also be used.

[0037] An energy application tool, such as a percussion bar, is mounted to the front of the housing, and the mounting mechanism for the percussion bar may include frictionless bearings. These bearings may include one or more axial openings so that adjacent chambers formed by the housing and the percussion bar communicate with each other for air exchange, depending on how much information is expected from the test.

[0038] For a given energy application tool, e.g., a physical tool such as a percussion rod, changing the impact force may be achieved, for example, by changing the voltage, current, or both, which may change the coil actuation time (changing the length of time the coil is energized or activated), changing the speed of the percussion rod moving towards the object upon impact, changing the coil delay time (changing the time between actuation activities), the number of coil energizations (i.e., changing the number of actuation pulses applied), the coil polarity, and / or a combination / multiple of them.

[0039] According to some embodiments, the drive mechanism may include a measurement device, e.g., a piezoelectric force sensor, mounted within the handpiece housing for coupling with an energy application tool, e.g., a percussion rod. The measurement device may be adapted to measure the deceleration of the percussion rod upon impact with the object in motion, or any vibrations caused by the percussion rod on the specimen. The piezoelectric force sensor may detect changes in the properties of the object and objectively quantify its internal features. Data transmitted by the piezoelectric force sensor may be processed by a system program, discussed further below.

[0040] According to some other embodiments, the drive mechanism may include a linear variable differential transformer adapted to sense and / or measure the displacement of an energy application tool, such as a percussion rod, before, during, and after the application of energy. The linear variable differential transformer may be a non-contact linear displacement sensor. The sensor may utilize inductive technology and therefore may be capable of sensing any metal object. The non-contact displacement measurement may also allow the computer to determine the velocity and acceleration immediately prior to impact, such that the effects of gravity may be omitted from the results. The communication between the drive mechanism and the energy application may be wired or wireless.

[0041] The open end of the housing may be provided with an object contacting portion that may or may not include a sleeve. In some embodiments, the open end of the housing may be placed in direct contact with the object during measurement, thereby stabilizing the device on the object. In some other embodiments, the sleeve may attach and / or surround at least the length of the free end of the housing and may protrude a distance from the housing, substantially coextensive with the end of the percussion bar in its extended form when the percussion bar moves axially. Thus, the length of the sleeve may depend on the length of extension of the extended percussion bar desired. The free end of the sleeve may be placed towards the object to be measured. The sleeve may be placed in direct contact with the object during measurement, thereby stabilizing the device on the object. In other embodiments, additional features may be included to further stabilize the device, and may also allow for some repeatability of placement of the device on the object, as discussed below.

[0042] In other exemplary embodiments of the invention, the device may be as described in the above exemplary embodiment, except that the sleeve may include a tab protruding from at least a portion of its end such that when the open end of the sleeve is in contact with at least a portion of the surface of the object to be measured, the tab may rest on a portion of the top of the object. The tab and sleeve together may aid in repeatable positioning of the handpiece relative to the object, and therefore the results provided are more reproducible than without the tab. In rare circumstances, the tab may not protrude at all to allow for testing at a lower position of the object. The tab may be substantially parallel to the longitudinal axis of the sleeve. In one aspect, the surface of the tab in contact with the object may be contoured with a concave or convex surface to better position on the top of the object, e.g., a tooth. In another aspect, the surface of the tab in contact with the object may be flat to match the topography of the object, e.g., a flat surface. In a further aspect, the surface of the tab in contact with the object includes a groove or a groove to match an object having an uneven surface. Additionally, the tabs can be adapted to be repeatedly positioned at substantially the same location on the top of the object each time, hi some embodiments, the tabs can be substantially parallel to the longitudinal axis of the sleeve.

[0043] In rare cases where the tabs may mate with a stable position on, for example, a dental implant transfer abutment, a sleeve portion without the tabs may be used for a lower, more stable placement on the abutment.

[0044] In further exemplary embodiments of the invention, the sleeve portion may include not only tabs, but also feature components, such as ridges, protrusions, or other features, that are substantially orthogonal to the surface of the tab on the side adapted to face the surface of the object. For example, with respect to teeth, the ridges or protrusions may fit between adjacent teeth or other orthogonal surfaces and thus help to prevent any substantial lateral or vertical movement of the tab across the surface of the object and / or further aid in repeatability. The tabs may be of sufficient length or width, depending on the length or width of the top portion of the object, so that the ridges or protrusions can be properly positioned during operation. Additionally, the tabs and features also aid in more repeatable results than without the tabs.

[0045] In the exemplary embodiments described above, the device may be of any form factor, as described above, including a handpiece having a longitudinal housing for housing the device as described above, or a desktop, or any form of portable part. The device, e.g., any portable form or handpiece, may be held at any angle relative to the horizontal during testing.

[0046] The stability of the instrument achieved by the tabs, or tabs and / or components, can minimize jerky actions by the operator that can confuse the test results, e.g., any defects inherent in the bone structure or physical or industrial structure can be drowned out by the jerky actions of the examiner. This type of defect detection is important because the location and extent of the defect can dramatically affect the stability of the implant or physical or industrial structure. In general, when a lesion such as a crestal or apical defect is detected in an implant, for example, the stability of the implant is affected if both crestal and apical defects are present. Previously, there was no other way to collect this type of information other than a process requiring expensive radiation. With the present device, this type of information can be collected and done without radiation in an unobtrusive and non-invasive manner.

[0047] In further exemplary embodiments of the invention, an inclinometer may be present as part of the electronic control system of any of the exemplary embodiments described above, which may, for example, trigger an audible warning if the device is outside of its angular range of operation, for example for a percussion rod, the system may trigger a warning to return the device to a more horizontal orientation when it is + / - about 45 degrees from horizontal, and even further, for example, + / - about 30 degrees.

[0048] In yet another exemplary embodiment of the invention, any or all of the exemplary embodiments described above may also include a force sensor not for detecting or measuring the force exerted by the energy application tool on the object during inspection or the response after impact of the energy application tool, but for detecting and / or monitoring that an appropriate contact force is exerted by the sleeve portion on the object undergoing measurement. As described above, during measurement, for example, the device may contact the object at the end of the housing or the sleeve portion. The contact force may vary depending on the operator. It is desirable that the force is applied consistently in a certain range and that the range is not excessive, regardless of the operator. A force sensor may be included in the device for detecting this force and a visual signal, audio or digital readout may be fitted. This sensor may also be used to ensure that a proper alignment is obtained with respect to the object being measured. The sensor, for example, the force sensor, may be in physical proximity and / or in contact and / or physically coupled to at least a portion of the device other than the energy application tool, for example, it may be in physical proximity and / or in contact and / or physically coupled to the housing and / or the sleeve portion in the case where the open end of the sleeve includes the object contact portion.

[0049] In general, the sensor may surround the energy application tool and may not be in physical contact with the tool. For example, the sensor may be positioned such that the energy application tool, even a physical tool, may pass through it to impact the object to be measured. The sensor may include a strain gauge, a piezoelectric element, a sensing pad, or any other sensor that may be clamped. The sensor, e.g., a force sensor, may be disposed anywhere inside the housing and may be in physical proximity and / or contact and / or physically coupled with at least a portion of a device other than the energy application tool, e.g., as described above, when the open end of the sleeve portion includes an object contact portion, it may be in physical proximity and / or contact and / or physically coupled with the housing and / or the sleeve portion. In one embodiment of the invention, the sensor may include at least one strain gauge for sensing. A strain gauge may be attached or mounted to the cantilever between the device housing and the sleeve portion such that when the object contacting portion of the sleeve portion is pressed onto the object, it also deforms the cantilever, which is measured by the strain gauge, thus providing a force measurement. In some embodiments, multiple strain gauges mounted on a single or separate cantilever may be utilized. The cantilever may also be on a separate component from the remainder of the housing or sleeve portion, for example, on a mounting device. In another embodiment of the invention, the sensor may include a sensing pad that may be positioned between the rigid surface and the slider, such that a force is measured when the pad is pressed or squeezed as the slider moves towards the rigid surface. According to some embodiments, the rigid surface may be, for example, a coil interface that holds an electromagnetic coil in a drive mechanism within the device housing of any of the above and below exemplary embodiments. The slider may be a force-transmitting sleeve-like component or member disposed within the housing, coupled to the object contacting portion of the sleeve portion, and adapted to slide within the housing when a force is exerted by the object contacting portion of the sleeve portion on an object. In some embodiments, it may be disposed within the sleeve portion.The sliding distance may be very small, for example, on the order of about 0.3 mm to about 1 mm (in millimeters or mm), and further for example, on the order of about 0.5 mm. The sensing pad may include a layered structure, which may be generally referred to as a "shunt mode" FSR (force sensing resistor), that may change resistance depending on the force applied to the pad to provide a force measurement. According to some alternative embodiments, the force transfer sleeve-like component or member may be biased forward by a spring such that when a force is applied on an object by the object contacting portion of the sleeve portion, the force transfer sleeve-like component or member may transfer the force towards the spring. According to one aspect, the force sensing may be performed by a linear position sensor, which knows, for example, that if the force transfer sleeve-like portion is at position X, a force of Y must be applied to it (against the reaction force of the spring) to move it to that position. According to another ... force sensor pressed against the spring, which may be biased forward by a force sensor pressed against the spring. In some further embodiments of the invention, the relative position of the object contacting portion of the sleeve portion on the object may be determined by having one or more strain gauges that may be attached at one end to the moving portion, e.g., a force sensing sleeve-like component, and at the other end to a static element, e.g., a housing. In further embodiments of the invention, the device may include a piezoelectric element for directly measuring the force. In still further embodiments of the invention, a Hall effect sensor may be used to detect a change in the magnetic field when a magnet (attached to the moving element) is moving relative to the position of the sensor. In still further embodiments of the invention, a capacitive linear encoder system, such as found in digital calipers, may be used to measure the force.

[0050] In addition to monitoring and detecting the contact force exerted by the operator on the object when the object contacting portion or sleeve portion of the housing contacts the object, the sensor can also be configured to activate the device when the correct amount of force is exerted on the object by the sleeve portion.

[0051] Although the sensor is not physically or mechanically coupled to the energy application tool in any way, the sensor may be in electronic communication with the energy application tool and may act as an on / off switch for the device or meter, as described above. For example, when an appropriate force is exerted on the object by the object-contacting portion of the housing or sleeve, it may trigger an activation mechanism of the device or meter to activate movement of the energy application tool to initiate a measurement. Therefore, as described above, no external switch or push button is required to activate the system on and off. An indication of the appropriate force may be indicated by a visible or audible signal.

[0052] The sleeve portion may be mounted on a force-transmitting sleeve-like component or member that forms or protrudes from the front of the housing and protects the energy application tool, e.g., a percussion bar, from damage in the absence of the sleeve portion, e.g., the sleeve portion may form part of a disposable assembly, as discussed below. The force-transmitting sleeve-like component or member may be around and surround the energy application tool, e.g., a percussion bar, held at the front by the housing and mounted at the rear on the front of the electromagnetic coil. The force-transmitting sleeve-like component or member may be adapted to slide a small amount, thereby acting on a force sensor, e.g., a force-sensing resistor, located between the rear of the force-transmitting sleeve-like component or member and the coil mount. The energy application tool, e.g., a percussion bar, may be actuated and a force may be detected when the object-contacting portion of the sleeve portion is pressed against the object to be measured, e.g., a tooth. If a correct force within a certain range is detected, the meter is turned on to start the measurement.

[0053] As described above and in all embodiments of the sensor, the sensor may be positioned to form a channel through which an energy application tool, such as a percussion rod, may pass to impact the object being measured, i.e., surround the percussion rod.

[0054] If the device is oriented such that the axis of motion is greater than about 45 degrees, even greater than about 30 degrees, from the horizontal position when a pushing force is detected on the object contacting portion of the sleeve portion, this may result in an alarm sound being emitted by a speaker mounted on the device, such as a printed circuit board (PCB) within the device. In such a situation, the impact action will not be initiated until the device is returned to an acceptable angle. In some instances, if an impact action is initiated when the above-mentioned deviation from the range is detected, the device may not actually stop operation, but may simply sound an alarm so that a correction can be made.

[0055] In further exemplary embodiments of the invention, any of the exemplary embodiments, systems and methods described above may also include a device operable by holding the device at variable angles from horizontal and adjusting the energy application process to simulate a substantially horizontal position during measurement, providing a system that may apply an optimal amount of energy to the object in all circumstances. In some embodiments, the device may produce substantially the same impact force on the object at various angles from the normal to the object surface as if the device were operating such that the direction of propagation was normal to the surface of the object. Thus, even if the device is operating at about + / - 45 degrees, or even, for example, about + / - 30 degrees from normal to the object surface, the device may still generate approximately the same amount of comparable impact force, for example, about 20-30 Newtons, for optimal results. The system may include a visual indicator, for example, an LED in this example, if the handpiece is held at an angle that does not allow for reliable measurements. The LED in this example may be red or any other pre-set color to indicate such a situation, to alert the user to readjust the angle at which the handpiece is being held.

[0056] The systems and methods of the present invention may, for example, accommodate reaching hard to reach objects, both anatomically and non-anatomically, to generate more reproducible measurements and also to be better able to detect any abnormalities that may be present in the object, increase flexibility of operation, etc. The device may include a housing having a hollow interior and an open end through which any tool capable of applying any type of energy to the object passes, such as a percussion rod positioned inside the housing through which the tool can apply mechanical energy, electromagnetic energy of any frequency, sound waves, such as light, acoustic energy, etc., to the object.

[0057] For example, the system may include a device for performing an impact action on an object. The device may be positioned with a housing having a hollow interior and an open end through which energy may be applied by an energy application tool, including any tool capable of applying any type of energy, including mechanical, sonic, or electromagnetic energy, to an object. In some embodiments, a tool capable of applying mechanical energy to an object, such as a percussion rod, may be positioned inside the housing through which it passes to reach the object to be measured. In some other embodiments, a source of electromagnetic energy of any frequency, such as light energy, may be positioned inside the housing, for example. In further examples, a source of sonic energy, such as an ultrasonic transducer, or any acoustic energy source may be positioned inside the housing.

[0058] The energy application tool may be held horizontally or in other positions during measurement and may have a tip substantially perpendicular to the main portion of the tool, maintaining a constant length whether at rest or under impact. In this subsequent embodiment, if the energy application tool is a mechanical tool such as a percussion rod, it may or may not include a removable tool tip that is substantially perpendicular to the longitudinal axis of the tool and housing.

[0059] An energy application tool, such as a percussion rod, can be programmed to strike the object a predetermined number of times per minute at substantially the same speed, and the deceleration information can be recorded or compiled for analysis by the system, as described above. In addition to helping to position the device, if it is of a material that has some dampening properties, it can also help to dampen any vibrations caused by the impact so as not to disturb sensitive measurements.

[0060] With regard to electromagnetic energy, the application of energy may be in the form of pulses or energy bursts that may be programmed to impact the object a predetermined number of times per minute, each time with substantially the same amount of energy, and the effects on the object may be recorded or compiled for analysis by the system. In some instances, repeated impacts may provide an average measurement that may be more representative of the actual underlying characteristics. In addition to helping to position the device, if it is of a material that has some damping properties, it may also help to dampen any vibrations caused by the impacts so as not to disturb the sensitive measurements.

[0061] Upon activation of a mechanical energy application tool, such as, for example, pressing a finger switch on the device, or when a certain amount of force is applied by the object contacting portion of the housing or sleeve as described above, a magnetic coil in the device extends at a speed toward the object being measured, propelling the energy application tool, such as a percussion rod, to strike or impact the object or specimen multiple times per measurement cycle, for example, with an impact force. In some instances, where the handpiece may be positioned in a mount, such as when it is desired to generate multiple signals in a controlled laboratory setting, the handpiece may be set up to be activated without waiting for the appropriate amount of contact force applied by the object contacting portion of the housing or sleeve. The impact force on the object may create a stress wave that travels through the energy application tool, such as a percussion rod, and by measurement of a sensing device or mechanism installed in the device, the deceleration of the tool, such as a percussion rod, upon impact with the object may be measured and transmitted to the rest of the system for analysis. The system may measure an impact response, such as energy reflected from the object as a result of the application of energy, over a time interval, for example by percussing or applying energy, which may include creating an impact response profile, e.g., a signal or a frequency response profile, based on the response from the object during the time interval, and / or evaluating the impact response profile, such as a signal, to determine the damping capacity or other characteristics of the object. The measuring device or sensing mechanism may detect characteristics of the impact from the impact of the energy application tool with the object. In general, the measuring device or sensing mechanism may be physically coupled, operably coupled, or otherwise in contact with the energy application tool such that it may detect characteristics of the impact. The coupling may be wired or wireless.

[0062] After impact with the object, the energy application tool, e.g., the percussion rod, decelerates as described above. The deceleration of the energy application tool, e.g., the percussion rod, can be measured by a measuring device or sensing mechanism, e.g., an accelerometer internal to the device. For example, an accelerometer in a device coupled to the energy application tool can be adapted to measure the deceleration of the energy application tool upon impact with the object in motion, the impact response from the object, measure the vibrations caused by the impact, or measure signals corresponding to the resulting stress waves. The measuring device or sensing mechanism can detect changes in the properties of the object and objectively quantify its internal features. The data transmitted by the measuring device or sensing mechanism can be processed by a system program as described above or below.

[0063] The measurement mechanisms described above may also be applicable to other than the mechanical energy application tools described above, using a similar sensor setup, for example when such energy application tools perform an impact action.

[0064] In some embodiments, the inclinometer may include an accelerometer, such as a three-axis device that measures gravity on all three axes, X, Y, and Z. In some embodiments of the invention, a device, such as a handpiece, may include software to measure the value of gravity (G-force) on the Y-axis (i.e., vertical). For example, if the G-force on the Y-axis is greater than a threshold of about + / - say 15 degrees, the handpiece may make an audible noise, such as a beep, a light signal, such as a flashing light, or a light of a certain color. If the G-force on the Y-axis is greater than a threshold of 30 degrees, the handpiece may emit a faster beep, or in the case of a light signal such as a flashing light, it may be a faster flashing light. The accelerometer may be sampled at a period of say 100 ms. Five valid measurements in a row may be required (500 ms) to trigger the threshold and therefore the beep or flash, etc. The thresholds for both the 15 and 30 degree thresholds may be determined empirically.

[0065] For example, for a device without the features of the present invention, in operation, if the equivalent impact force is about 26 Newtons at +15 degrees from horizontal, the equivalent impact force may be about 32 Newtons at the horizontal position, and at -15 degrees from horizontal, the impact force may be about 35 Newtons. With the present invention, all impact forces at all angles mentioned above may be about 25 Newtons, or whatever the optimum impact force is programmed.

[0066] As mentioned above, the system can be turned on and off with or without an external switch or remote control. In some embodiments, the handpiece energy application process can be actuated via a mechanical mechanism, such as by a switch mechanism. In one aspect, a finger switch can be located at a convenient location on the handpiece for easy activation by the operator. In another aspect, the switch mechanism can be actuated by applying pressure to an object through a sleeve. In some other embodiments, the handpiece energy application process can be actuated via voice or foot control or a button within a computer software user interface.

[0067] Generally, any external switching device, such as a flip switch, locking switch or push button switch, may tend to limit how an operator holds the instrument, for example when it is hand-held during measurement, in order to allow easy access by the operator to the switching device to turn it on and / or off, and therefore may limit the positioning of the instrument on the object.

[0068] In some embodiments, voice or remote control may be commonly used to provide greater flexibility in instrument positioning, but such voice or remote control may add additional complexity to the system. With the present invention, the same advantages of flexibility may be obtained without such remote control or additional complexity.

[0069] In some other embodiments, to obtain more flexibility in the positioning of the meter, activation of the device may be controlled by an appropriate contact force between the object and a sleeve portion located at the open end of the housing, as described above and below. This appropriate contact force may also add other desired features to the system, as discussed below. The sleeve portion may be open at its free end with an object resting, pressing, or contacting portion for resting, pressing, or contacting on at least a portion of the object during measurement. The contact by the sleeve portion helps to stabilize the device on the object. During measurement, the force exerted by the sleeve portion on the object is controlled by the operator, unlike the impact force of the energy application tool, which may be controlled by various factors of the system described above, and the appropriate force on the object may be important and may need to be monitored, since, for example, insufficient or excessive force exerted by the operator may complicate the measurement and even produce less accurate results. For better reproducibility, there may be a sensor that is deployed inside the housing and is not physically or mechanically coupled to the energy application tool to ensure that the appropriate contact force by the contact portion of the sleeve portion can be applied even by different operators.

[0070] In some embodiments, the meter may be turned on immediately once the proper contact force is exerted on the object by the object contacting portion of the sleeve, as indicated by a visible or audible signal. In some embodiments, there may be a delay before turning on the meter once the proper contact force is exerted on the object by the object contacting portion of the sleeve, as indicated by a visible or audible signal. In further embodiments, the meter may be turned on to begin measuring once a certain pressing force between the object contacting portion of the sleeve portion and the object is detected and maintained for a period of time, for example, about 1 second, or even for example, about 0.5 seconds. In this embodiment, a green light will illuminate the tip and impact will begin about 1 second, or even for example, 0.5 seconds after the force within the correct range has been maintained.

[0071] For example, an appropriate force exerted by the operator on the object through the sleeve portion acts as a switch for the system. If the system does not switch on, it may be desirable to know if it is faulty or if insufficient or excessive force is exerted. In some embodiments, the force measurement may be connected to a visible output such as a light emitter. The light emitter may be mounted in any convenient location on the device or instrument, for example, one or more LEDs may be mounted on the front of the device or instrument. In one aspect, multiple light systems may be included. For example, two LEDs may be used. If the force is within the correct range, a blue light may be illuminated. If too much force is detected, the LED may turn red and the instrument will not activate unless the pressure is reduced. In some embodiments, if the user presses too hard on the object, the light may first turn amber and then red. If the pressure is enough to turn the light red, the impact may not be initiated or may be aborted if already initiated. There may also be an amber LED state that warns if the user is approaching too much pressure. At that stage, the gauge may still operate if the LED is illuminated amber. In another embodiment, the light may not indicate any too little force, and a blue light may indicate the right amount of force, while a red light may indicate too much force. In yet another embodiment, one light system may be included. For example, the light may not give any signal of too little force, and a red light may give a signal of too much force. In a further embodiment, a flashing red light may indicate too much force, and the light may not indicate any too little force.

[0072] In some alternative embodiments, the force measurement may be connected to an audible output. In one aspect, the audible output may include a single beep to indicate too little force and multiple beeps to indicate too much force. In another aspect, the audible output may include a beep to indicate too little force and a beep with a flashing red light to indicate too much force. In a further aspect, the force measurement may be connected to an audio alarm system to warn of too much or too little force. In a further aspect, the force measurement may be connected to an audio alarm system to warn of too little force and an audio alarm and a flashing red light to warn of too much force.

[0073] During the measurement, as described above, the system may measure the impact response, such as energy reflected from the object as a result of the application of energy, for example by percussion or applying energy, or deceleration information of the energy application tool, i.e., energy return, for a certain time interval. The response may be fed to a computer and the information recorded or compiled for analysis by the system, which may include creating an impact response profile and / or evaluating the impact response profile to help determine the structural characteristics of the object, such as vibration damping capacity; sound damping capacity; defects including inherent defects in the framework or materials that make up the object; cracks, microcracks, fractures, microfractures; loss of cement seal; cement failure; adhesion failure; microleakage; lesions; caries; general structural integrity or general structural stability.

[0074] Generally, the loss factor is an indication of the overall capacity of the damping capacity within the object or structure being inspected. In the impact process, it is based on the maximum energy return or the square of the impact force measured with a measuring or sensing mechanism coupled to an energy application tool, e.g., an impact bar, as discussed above and below. The normal fit error (NFE) or damage or instability is the total error (difference) between the ideal curve (produced by an object without defects) and the actual inspection data. These results can be calculated from ERG, FRG, or other physical return values ​​or metrics. All response curves are normalized to a maximum of 1 before determining the NFE, therefore, it is not directly related to the loss factor.

[0075] As mentioned above, the systems and methods of the present invention may include devices that are non-destructive and non-invasive and can operate by holding the device at variable angles from horizontal and adjusting the energy application process to simulate a substantially horizontal position during measurement. The system may or may not include disposable parts and / or features that aid in repositionability. The present systems and methods for measuring structural features may minimize or even micro-impact the impact on the object being measured without compromising the sensing of the measurement and the operation of the system. If the energy application tool is a percussion rod, the amount of impact energy may also vary depending on, for example, the length of the rod, the diameter of the rod, the weight of the rod, or the speed of the rod before impacting, etc. In some embodiments, the system includes an energy application tool that is lighter and / or can move at a slower speed to minimize the force of impact on the object being measured while presenting, maintaining, or providing an equal or better sensitivity of the measurement. In one aspect, the energy application tool, e.g., a percussion rod, may be made from a lighter material to minimize the weight of the handpiece and therefore the impact on the object being measured. In some other embodiments, the energy application tool, e.g., a percussion rod, may be made shorter and / or with a smaller diameter, which may also minimize the size of the hand piece and therefore minimize the impact on the object being measured. In further embodiments, the system may include a drive mechanism that may lessen the acceleration of the energy application tool and therefore minimize the impact on the object being measured. For example, the drive mechanism, whether it is lightweight or not and / or smaller in length or diameter or not, may include a separate drive coil to lessen the acceleration of the energy application tool, minimizing the impact force on the object being moved while maintaining the sensitivity of the measurement. These embodiments may be combined with one or more of the embodiments described above and below, including a lighter hand piece housing. It may also be desirable for the speed at which the measurement is made not to increase the initial speed of the impact and minimize the impact on the object being measured. The system may or may not have disposable parts and / or features that aid in repositionability, as described above and below.

[0076] The system may include a drive mechanism that can change the travel distance of the energy application tool while maintaining the initial velocity of impact of the object with the energy application tool. For example, if the energy application tool includes a percussion tool, the distance can vary from about 2 mm to about 4 mm. Reducing the travel distance of the energy application tool, for example, from about 4 mm to about 2 mm, while maintaining the same initial velocity on impact or contact, can allow faster measurements without compromising the operation of the system. The system may or may not include various exemplary embodiments described above or below. For example, the system may or may not have disposable parts and / or features to aid repositionability and / or reduce impact with features described previously or below.

[0077] As with any of the above exemplary embodiments, the systems and methods for measuring structural characteristics using an energy application tool may also include disposable features to help eliminate or minimize contamination through transmission from the system to the object being measured or cross-contamination from objects previously measured without impeding the measurement or the system's capabilities. The instrument includes a housing having a hollow interior with an open end and an energy application tool, e.g., a percussion rod or impact rod, mounted within the housing for movement within the housing. The system provides a method of non-destructive measurement with some contact with the object being measured, without the need for wiping or autoclaving the energy application tool, and at the same time, without disposal of the energy application tool and / or housing, whatever may be contained within the housing of the instrument.

[0078] Disposable features may include a cover to cover or encase a portion of the system that may be in close proximity and / or in contact with the object being measured without interfering with the sensitivity, reproducibility, or general operation of the instrument to any substantial degree if desired.

[0079] The disposable features may include any of the features described below or as disclosed in U.S. Patent No. 9,869,606, entitled “System and Method For Determining Structural Characteristics Of An Object,” or WO2011 / 160102A9, the contents of which are hereby incorporated by reference in their entireties.

[0080] The disposable feature may include a sleeve portion extending from and / or enveloping the open end of the housing. In one example, with respect to a mechanical energy application tool, the sleeve portion includes a hollow interior and an open free end having an object resting or contacting portion for resting on, pressing against, or contacting an object at the open end during measurement. Features such as contact features may or may not be movable and may include a closed end that has a length and is deployed toward the open end of the sleeve portion to substantially close the free end of the sleeve portion by, for example, frictionally nestling against the interior of the sleeve portion to substantially close it. The contact feature may be, for example, a short tubular segment or loop and may include a closed end to substantially close the free end of the sleeve portion. The contact feature may be positioned between the tip of the energy application tool and the surface of the object to be measured. The contact feature described above may include a thin film that may be integrally attached or formed as described above and below to form a closed end. The film may be thick or thin, as long as it is selected to have a minimal effect on the operation of the energy application tool. For example, the closed end, whether it is closed by a membrane or other structure, may have some elasticity or be deformable and may adapt to various surface configurations of the object being measured so that intimate contact with the object may be achieved during impact.

[0081] The closed end may include a polymeric thin film that may or may not be of the same material as the remainder of the contact feature, or it may be a material that has substantially the same properties as the remainder of the contact feature. The polymer may include any polymeric material that can be molded, cast, or stretched into a thin film so as not to substantially adversely affect the measurement. In some other embodiments, the closed end may include an insert molded metal foil film. The metal may be any metallic material that can be stretched, cast, or formed into a thin film so as not to substantially adversely affect the measurement. The film may also be formed to conform to the shape of the energy application tool, or vice versa, for optimal transfer of force / energy. In some embodiments, the film may be constructed from stainless steel foil or sheet, and may be, for example, pressed and / or molded. In other embodiments, the closed end may be integral to the contact feature. For example, the contact feature may be formed from a material that can be shaped, such as by pressing a metal (e.g., stainless steel, aluminum, copper, or other suitable metal), into a tubular or hoop structure with a closed end of a desired thickness.

[0082] Additionally, the sleeve portion, contact features, and contact tabs, and / or the sleeve, tabs, and components may be made from recyclable, compostable, or biodegradable materials, which is particularly useful in embodiments that are to be disposed of after a single use.

[0083] The device itself may be tethered to an external power supply, such as a battery, a capacitor, a transducer, a solar cell, an external power source, and / or any other suitable power source, or may be powered by a power source contained within the housing.

[0084] The system and method may be applicable to inspect various objects that are mechanical, as previously mentioned. For mechanical objects, which may include, but are not limited to, polymer composite structures, including honeycomb or layered honeycomb, or metal composite structures; aircraft airframe structures, automobiles, ships, bridges, tunnels, trains, buildings, industrial structures, including but not limited to power generation facilities, arch structures, or other similar physical structures, the inspection may also be performed on stationary or moving movable objects. Thus, mechanical objects may be inspected even when they are stationary or moving, which may give specific insight into the object under real operating conditions. For moving objects, such as trains, the inspection may be performed over many different points. This may be performed using one energy application tool over multiple points on the object to obtain a general average state of the object, or may be performed over the same point using many separate tools or devices to obtain average results over the same point. With regard to performing measurements on the same point using many energy application tools, the devices or tools, e.g., an array of percussion rods striking the object, may be positioned consecutively along the path of the moving object over a distance, and by controlling the spacing between the tools or devices, it may be possible to match the speed of the moving object, e.g., a train, to the spacing of the energy applications on the same point of the object to obtain an average value for the point. In this example, measurements may be performed under real operating conditions. In some embodiments, the array of devices may be a linear array, either a vertical or horizontal array, or a curved array. In another aspect, the array may be arranged in a two-dimensional array, flat, or curved.

[0085] In embodiments where the object is large, measurements at different locations on the object, for example impacts on multiple parts of the object, may allow a better assessment of structural characteristics that are more representative of the object.

[0086] In general, structural characteristics as defined herein may include vibration damping capacity; acoustic damping capacity; defects, including, for example, defects inherent in the structure or materials that compose the object; cracks, microcracks, fractures, microfractures; loss of cement seal; cement failure; adhesion failure; microleakage; lesions; caries; general structural integrity or general structural stability. With respect to anatomical objects such as tooth structures, natural teeth, natural teeth having fractures due to wear or trauma, natural teeth that have become at least partially abscessed, or natural teeth undergoing bone augmentation procedures, prosthetic dental implant structures, dental structures, orthopedic structures, or orthopedic implants, such characteristics may be indicative of the health of the object or the health of an underlying foundation to which the object may be fixed or attached. The health of the object and / or underlying foundation may also relate to density or bone density, or level of osseointegration; any inherent or other defects; or cracks, fractures, microfractures, microcracks; loss of cement seal; cement failure; adhesion failure; microleakage; lesions; or caries. For objects in general, such as polymer composite structures including honeycomb or layered honeycomb, or metal composite structures; industrial structures including, but not limited to, aircraft airframes, automobiles, ships, bridges, buildings, power generation equipment, arch structures, or other similar physical structures, such measurements may also relate to defects or cracks, as well as any structural integrity or structural stability such as hairline cracks or microcracks.

[0087] Additionally, changes in the structure of the tooth, or any underlying structure to which a mechanical structure is attached or secured, that reduce the ability to dissipate mechanical energy associated with impact forces and therefore, for example, reduce the overall structural stability of the tooth, can be detected by evaluation of the energy return data compared to an ideal, undamaged sample.As also mentioned above, the present invention also generally contributes to the accuracy of detection location of defects, cracks, microcracks, fractures, microfractures, leaks, lesions, loss of cement seal; microleaks; caries; structural integrity of cement failure; adhesion failure; general or structural stability.

[0088] In some exemplary embodiments, the invention includes: A method for providing a machine learning trained structural feature analysis system, comprising: providing or generating a data set comprising a plurality of signals from a plurality of groups of distinct objects, the signals being generated from a series of impact measurements on the distinct objects and grouped based on a common feature of one of the groups of distinct objects; performing a stochastic decomposition of each of said signals to generate a signal collection, each signal including at least one sub-signal; performing an optimization operation to minimize differences between said signal collection and each of said signals to generate an optimized signal collection; identifying and addressing potential errors or imperfections within each of the optimized signal collections; repeating the guess decomposition and optimization operations to regenerate an optimized signal collection after the potential errors or defects have been addressed; selecting at least one desired signal collection from the optimized signal collections for each signal to add to the set of optimized signal collections; and Incorporating the set of optimized signal collections and associated methods for arriving at the set of optimized signal collections into a machine learning trained analysis system (MLTA). generating an optimized signal collection for each of the set of signals; and Connecting the MLTA to a measurement device, the measurement device adapted to generate the clinical signal data by percussing a target object and transmitting the clinical signal data to the MLTA to enable the MLTA to process the clinical signal data to create a clinical optimized signal collection, compare characteristics of the clinical optimized signal collection with characteristics of the set of optimized signal collections, and present the results of the comparison in a human readable format.

[0089] In some exemplary embodiments, the invention includes: Machine learning trained structural feature analysis systems, including: Crash measurement devices, including: a housing having an open front end and a longitudinal axis; an energy application tool mounted within the housing, the energy application tool having a static configuration and an activated configuration; a drive mechanism supported within the housing, the drive mechanism adapted to activate the energy application tool between the rest configuration and the activated configuration to apply a set amount of energy; and a control mechanism connected to provide instructions to said drive mechanism; the drive mechanism varying an amount of energy applied to activate the energy application tool between the rest configuration and the activated configuration based on an input from the control mechanism. A program logic module coupled to said control mechanism, said program logic module being provided by: Providing or generating a data set comprising a plurality of signals from a plurality of groups of distinct objects, the signals being generated from a series of impact measurements on the distinct objects and grouped based on a common feature of one of the groups of distinct objects; generating a set of optimized sub-signal collections for each of the signals, each of the sets of optimized sub-signal collections being generated by: performing a stochastic decomposition of each of said signals to generate a subsignal collection for each signal including at least one subsignal; performing an optimization operation to minimize differences between said sub-signal collection and each of said signals to generate an optimized sub-signal collection; identifying and addressing potential errors or imperfections within each of the optimized sub-signal collections; repeating the guess decomposition and optimization operations to regenerate an optimized sub-signal collection after the potential errors or defects have been addressed; selecting at least one desired subsignal collection from the optimized subsignal collections for each signal for addition to the set of optimized subsignal collections; and incorporating the set of optimized sub-signal collections and associated methods for arriving at the set of optimized sub-signal collections into a machine learning trained analysis system (MLTA); connecting the MLTA to the collision measurement device, the collision measurement device being adapted to generate the clinical signal data by percussing a target object and transmitting the clinical signal data to the MLTA to enable the MLTA to process the clinical signal data to produce a clinical optimized sub-signal collection and a comparison between features of the clinical optimized sub-signal collection and features of the set of optimized sub-signal collections and to determine physical parameters related to the comparison; a control regulator connected to the program logic module and to the control mechanism, the control regulator adapted to output changes to the instructions in response to the MLTA outputting proposed changes due to the physical parameters.

[0090] In some exemplary embodiments, the invention includes: A method for providing a structural feature analysis system, comprising: providing a program logic module (PLM) configured to receive a signal input for generating a signal from a crash measurement by a crash measurement device (PMD); connecting the PLM to the PMD; performing an impact measurement on a tooth-like object using the PMD to generate the signal using the PLM; performing an inference of significant subsignals of said signal by fitting said signal to a basis function; subtracting the prominent side signal from the signal to form a remainder; performing a guess sinusoidal decomposition on the remainder to generate secondary subsignals that, when summed with the significant subsignal, form an approximation of the signal; performing an optimization operation to minimize a difference between the approximation of the signal and the signal to generate an optimized collection of sub-signals; identifying and addressing potential errors or imperfections within each of the optimized sub-signal collections; repeating the guess, guess sinusoidal decomposition and optimization operations of the significant side signals to regenerate the optimized side signal collection after the potential errors or defects have been addressed; selecting at least one desired sub-signal collection from the optimized signal collection; Presenting the desired subsignal collection in a human readable format.

[0091] The present invention with the above and other advantages may be better understood in conjunction with the following detailed description of aspects, embodiments, and examples of the invention, taken together with the description of the drawings. The following description sets forth various aspects, embodiments, and examples of the invention and several specific details thereof, given by way of illustration and not by way of limitation. Many substitutions, modifications, additions, or permutations may be made within the scope of the invention, and the invention includes all such substitutions, modifications, additions, or permutations. [Brief description of the drawings]

[0092] [Figure 1] FIG. 1 shows a diagram of the connection arrangement of the components of the system of the present invention. [Diagram 2] FIG. 2 shows an embodiment of the collision measurement device of the present invention or components thereof. [Figure 2a] FIG. 2a shows an embodiment of the collision measurement device of the present invention or a component thereof. [Figure 2b] FIG. 2b shows an embodiment of the collision measurement device of the present invention or a component thereof. [Figure 2c] FIG. 2c shows an embodiment of the collision measurement device of the present invention or a component thereof. [Figure 2d] FIG. 2d shows an embodiment of the collision measurement device of the present invention or a component thereof. [Figure 2e] FIG. 2e shows an embodiment of the collision measurement device of the present invention or a component thereof. [Figure 2f] FIG. 2f shows an embodiment of the collision measurement device of the present invention or a component thereof. [Diagram 3] FIG. 3 shows an exemplary profile of a signal according to the present invention. [Figure 3a] FIG. 3a shows an exemplary profile of a signal according to the present invention. [Figure 4] 4 and 4a show an example of a missing portion of a signal. [Figure 4b] FIG. 4b shows an example of a sub-signal decomposition of a signal. [Figure 4c] FIG. 4c shows an example of a prominent side signal in a signal. [Figure 4d] FIG. 4d shows the finite element analysis model. [Figure 4e] FIG. 4e shows a periodontal ligament attached to a tooth-like structure. [Diagram 5] FIG. 5 is a diagram showing an example of a heat map of the present invention. [Figure 5a] FIG. 5a is a diagram showing an example of a heat map of the present invention. [Figure 5b] FIG. 5b shows an example of a heat map of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0093] The detailed description set forth below is intended to describe some example systems, devices, and methods provided in accordance with aspects of the invention, and is not intended to represent the only manner in which the invention may be made or utilized. Rather, it should be understood that the same or equivalent functions and components may be accomplished by different embodiments that are also intended to be encompassed within the spirit and scope of the invention.

[0094] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Although any systems, methods, devices, and materials similar or equivalent to the systems, methods, devices, and materials described herein can be used in the practice or testing of the invention, some exemplary systems, methods, devices, and materials are described herein.

[0095] All publications mentioned herein are incorporated herein by reference for the purpose of describing and disclosing, for example, designs or methodologies described therein that may be used in connection with the presently described invention. The publications listed or discussed above, below, or throughout this text are provided solely for their disclosure prior to the filing date of the present application. Nothing herein should be construed as an admission that the inventors are not entitled to antedate such disclosure by virtue of prior invention.

[0096] The present invention relates to a system and method for measuring and evaluating structural characteristics of an object, whether anatomical or non-anatomical, using non-invasive and / or non-destructive methods of measurement. The damping capacity of an object is an important parameter in a wide variety of applications, including anatomical or non-anatomical objects. For example, in the field of dentistry, when a healthy tooth is subjected to an impact force, the mechanical energy associated with the impact is mainly dissipated by the periodontal ligament. Changes in the structure of the periodontal ligament, which reduces its ability to dissipate the mechanical energy associated with the impact force and therefore reduces the overall tooth stability, can be detected by measuring the loss modulus of the tooth.

[0097] Objects, whether anatomical or non-anatomical, such as the dental system, whether natural teeth or implants, can also suffer from defects over time. Some defects require dental restorative treatment to be performed. Such procedures can be invasive, expensive, and require long recovery times, especially if the defects are not easily identifiable until they develop into something more identifiable that may be serious. In order to reduce the risk of ineffective or unnecessary treatment, there is a great need for techniques that can quickly identify and pinpoint the type and location of existing problems before they become serious and / or before destructive treatment.

[0098] As used herein, the following specific definitions shall generally apply unless otherwise indicated.

[0099] As used herein, a "signal" represents an energy, force, displacement or other physical change value over time returned from an object after impact by an impact measurement device, such as an electrical response generated by a sensing mechanism, such as a piezoelectric sensing element, a strain gauge, a displacement sensor, or the like, force versus time data generated by recording such an electrical response, or a graphical representation of force versus time data (force return graph or FRG), energy versus time data (energy return graph or ERG), displacement versus time data (displacement return graph or DRG), or other graphical representation of applicable data. Unless otherwise indicated, a signal may be generally understood to be in a waveform profile that substantially represents the total force, energy, displacement or other physical change value versus time data recorded by the impact measurement device and received by the system during the measurement that was not adjusted or manipulated by the system. A "signal" may also represent an artificially generated simulated version of the above. Different physical change values ​​may be derived or calculated from the actual measured physical values.

[0100] As used herein, a "sub-signal" refers to a component of a signal that assumes a waveform profile such that the sum of all the sub-signals is the original signal or an approximation thereof.

[0101] As used herein, "sinusoid" refers to a waveform that adopts the general shape of a sine wave, including a sine wave that decays in amplitude over time, such as an exponentially decaying or other damped sinusoid (a "decaying sinusoid"). As used herein, "clinical signal" refers to a signal captured during use of an impact measurement device in a clinical, industrial or other non-training or non-examination environment by an end user (i.e., a "clinician").

[0102] As used herein, a "basis function" refers to a function or waveform for fitting a signal or sub-signal to, for example, a Gaussian distribution curve, a sinusoid (including decaying or damped sinusoids as discussed above), an exponential function, a sinusoid-like curve, and / or any other suitable fitting function or waveform.

[0103] The structural features of the object may be identified based on measurements of the same or other objects previously made using the system and captured by the system. The system may include a device (i.e., impact measurement device) capable of applying energy to the object, which may be an anatomical object or a mechanical object; measuring a response, such as energy reflected from the object as a result of the application of energy, such as percussion of the object, or a response, such as energy return, such as deceleration information of an energy application tool, such as a percussion rod, for a certain time interval; recording or compiling an analysis by a computer system of the measurements, etc.; and creating (i.e., generating a signal) a response profile, such as a signal, return curve or return graph (e.g., time profile or frequency profile), to evaluate the characteristics of the object being measured. The device may be generated from measurements of force / energy / displacement, etc. returned to the impact measurement device, such as by measuring force, energy, displacement or other physical return value at a sensing mechanism over a period of time; or deceleration information from the energy application tool after the energy application process. Several embodiments of the impact measurement device are described above and below in connection with Figs. 2, 2a, 2b, 2c, 2d and 2e.

[0104] FIG. 1 shows one embodiment of the architecture of a system using an end point device (i.e., a collision measurement device as referred to throughout) for tooth measurements, etc. The collision measurement device may be typically installed in, for example, a dental clinic or other location where measurements may be made on a patient's object (e.g., a tooth, an implant, etc.). The collision measurement device may typically include or be connected to a computing device, such as a PC workstation, laptop, tablet, or some other common computing device, which may connect to a larger network, such as the Internet, or a private network, such as a cloud service. At least one device may be attached to the collision measurement device, for example, via a wired data transmission technology, such as USB or Firewire, or via a wireless data transmission technology, such as Bluetooth. The system may further include a base station as shown for interfacing with the collision measurement device and the computing device.

[0105] An impact measurement device suitable for use in inspecting an object may include a housing having a longitudinal axis with an open end and an energy application tool, such as a percussion bar or impact bar, mounted inside the housing for axial movement along the longitudinal axis of the housing, as shown in FIG. 2a, which illustrates one embodiment of the impact measurement device discussed above. In some embodiments, the system may include a hand piece 104 in the form of an impact instrument. The hand piece 104 may have a cylindrical housing 132 having an open end 132a and a closed end 132b. The open end 132a is tapered as illustrated herein, although other configurations are also contemplated. The energy application tool 120, such as a percussion bar 110, may be mounted inside the housing 132 for axial movement, as described above. The hand piece also includes a drive mechanism 160 mounted inside the housing 132 for driving the percussion bar 120 axially within the housing 132 between a retracted position and an extended position during operation. The drive mechanism 160 may include an electromagnetic coil 156, which will be discussed further below. The percussion rod 110 may have a permanent magnetic ensemble 157 mounted on the end remote from the free end. An electromagnetic coil 156 of a drive mechanism 160 may be located behind the other end of the percussion rod 110, resulting in a relatively small outer diameter for the handpiece 104.

[0106] The mounting mechanism for the energy application tool 110, e.g., percussion bar 110, may be formed by bearings 1003 and 1004, as shown in Figures 2a and 2b, for receiving or supporting the percussion bar 110 in a primarily frictionless manner. A magnetic or drive coil 156 may be located within the housing 132 adjacent to the permanent magnet 157 and axially behind the permanent magnet 157. The magnetic coil 156 and the permanent magnet 157 form the drive for the forward and return movement of the percussion bar 110. The drive coil 156 may be an integral component of the housing 130 and may be connected to a supply hose or line 1000.

[0107] The two bearings 1003 and 1004 may be substantially frictionless and may include a number of radially inwardly extending ridges separated by axial openings 1400, as shown in Figures 2a and 2b. The axial openings 1400 in the bearing 1003 allow air movement between a chamber 1600 and a chamber 1500 separated by the bearing 1003, which chamber is formed between the inner wall surface of the housing 132 and the percussion rod 110. The air movement between these chambers 1500 and 1600 may therefore compensate for the movement of the percussion rod 120.

[0108] Referring again to FIG. 2f, the sleeve 108 is positioned toward and extends beyond the end 132a. The sleeve 108 wraps around the end of the housing 132a and is flattened at its end 116 to facilitate positioning against the surface of the object during operation. The sleeve assists in positioning the hand piece 104 on the object to stabilize the hand piece during operation. The sleeve 108 may also include a tab 118, as shown in FIG. 2f, protruding from a portion of its end 116 such that when the open end 116 of the sleeve 108 is in contact with the surface of the object being measured, the tab 118 may rest on a portion of the top of the object. Both the tab 118 and the sleeve 108 assist in stabilizing and repeatably positioning the hand piece 104 against the object, and the tab 118 may be positioned substantially the same distance from the top of the object each time. As discussed above, the object may include an anatomical or physical structure.

[0109] FIG. 2 shows an embodiment of another device (e.g., impact measurement device as referenced throughout) applicable for the present invention. The system may include a handpiece 100 having a housing 102 that houses the energy application tool and the sensing mechanism as shown in the block diagram of FIG. 2 with an energy application tool 110 and a sensing mechanism 111 generally disposed proximal to the end of the energy application tool 110 to receive force or energy from a target. Generally, the handpiece may be referred to as a handheld device, but may also include any other suitable form for a desired application, such as, but not limited to, a mounted device or a tool / mechanical / robot-controlled articulated device. The handpiece 100 may also be interchangeably referred to herein as, for example, a device or an instrument. In some embodiments, the energy application tool 110 as described may be mounted in the housing 102 for axial movement in a direction A toward the object, and such axial movement may be achieved via a drive mechanism 140. The drive mechanism 140 may generally be a linear motor or actuator, such as an electromagnetic mechanism, that may affect the axial position of the energy application tool 110, such as by generating a magnetic field that interacts with at least a portion of the energy application tool 110 to control its position, velocity, and / or acceleration through magnetic interaction. For example, an electromagnetic coil disposed at least partially in the vicinity of the energy application tool 110 may be energized to advance the energy application tool 110 toward the object to be measured, as described with the electromagnetic coil 140. The electromagnetic coil may also be alternately energized to retract the energy application tool 110, for example, to prepare for a subsequent impact. Other elements, such as repulsive magnetic elements, may also be included to assist in repositioning the energy application tool 110 after propulsion via the electromagnetic coil. The drive mechanism 140 and / or other parts of the instrument may generally be powered by a power source, as shown with the power source 146, which may be a battery, a capacitor, a solar cell, a transducer, a connection to an external power source, and / or any suitable combination / plurality thereof.To power the handpiece 100 or to charge an internal power source, such as power source 146, an external connection to a power source, such as power interface 147 of FIG. 2, may be provided, which may include, for example, power contacts for direct conductive charging, or power interface 147 may utilize wireless charging, such as inductive charging.

[0110] In some other embodiments, as illustrated in the block diagram of the handpiece 100 in FIG. 2c, the energy application tool 110 may be utilized to move substantially in a direction A, which may be perpendicular or substantially perpendicular to the longitudinal axis of the housing 102. As illustrated, the energy application tool 110 may be substantially L-shaped, for example, to accommodate interaction with the drive mechanism 140 and to protrude in a direction A substantially perpendicular to the axis of the housing 102. As illustrated in one example, the drive mechanism 140 may act on the energy application tool 110 to rock the pivot 110a and move the tip in the direction A. The drive mechanism 140 may utilize an alternating magnetic element, for example, that may act on the energy application tool 110 to move alternately in two directions, such as up and down. In another example, as shown with bend 110b, the bend portion of the L-shaped energy application tool 110 may include a bendable and / or deformable structure such that a linear force applied by the drive mechanism 140 may push the energy application tool 110 at the tip in direction A by imparting a forward motion around the bend 110b. For example, the bend 110b may include a braided, split, spring-like, and / or otherwise bendable piece that may also impart a motion and / or force around the bend. In general, the shape of the L-shaped energy application tool 110 may generally include other angles other than 90 degrees, such as between about + / - 45 degrees from the rear portion 110d. In some embodiments, the energy application tool 110 may also include multiple portions, such as portions 110c and 110d, that may be separable, such as portions 110c and 110d, such that portion 110c may be removed and disposed of between uses or patients, such as to help prevent cross-contamination. In general, the separable portions may include interfaces for combining them for use in measurements such that they function as a single energy application tool 110, as described below.

[0111] In some embodiments, the L-shaped energy application tool 110 may be rocked about the pivot 110a, such as by using an external force applied from the drive mechanism 140, as shown in Fig. 2d and Fig. 2e. For example, the drive mechanism 140 may apply alternating forces to the energy application tool 110 to rock about the pivot 110a, such as by using an applied force D from the applied portion 140d to the rear portion 110d to cause a rocking motion in a direction A' away from the target object, as shown in Fig. 2d, or by using an applied force E from the applied portion 140c to the rear portion 110d to cause a rocking motion in a direction A'' toward the target object such that the energy application tool 110 is driven in the direction A, as shown in Fig. 2e. The forces D and E may be applied by any suitable method, such as by applying a magnetic force on the energy application tool 110, which may include a magnetic or metallic element that may respond to the application of force from the drive mechanism 140. In general, the shape and arc of the rocking motions A' and A'' may be designed such that the energy application tool 110 impacts the target object in a direction substantially perpendicular to the target object's surface, as shown in FIG. 2e with rocking motion A'' around bend 110b in a substantially vertical orientation in bend 110c. To rest the device 100 for a subsequent measurement, portion 140d may apply a return force D as shown in FIG. 2d to cause rocking motion A' to return the energy application tool 110 to a resting or resting state. In general, the interior of device 100 may be adapted to allow rocking motions A' and A'' without interfering with the energy application tool 110.

[0112] Other examples of endpoint devices may include, by way of example and without limitation, those described in U.S. Patents 6,120,466, 7,008,385, 6,997,887, 9,358,089 9869606, U.S. Pat. No. 10,488,312, PCT / US17 / 69164, PCT Patent Application Serial No. PCT / US20 / 40386, U.S. Patent Publication No. 20190331573, PCT / US2018 / 068083, and / or PCT Publication No. WO2019133946, which are incorporated by reference in their entireties.

[0113] Generally, the system of the present invention includes a program logic module that may be generally utilized to process a signal received from a crash measurement device after measuring an object, as discussed above. The program logic module may generally apply an algorithm to form at least a portion of an initial guess of a decomposition of the signal into at least one of its component sub-signals. The result may generally be an initial guess decomposition. The program logic module may then perform an optimization, such as via an optimization algorithm, to minimize the difference between the initial guess decomposition and the initial signal. This may include, for example, optimizing (i.e., minimizing) the error, absolute error, or squared error between the signal and the guess decomposition. For example, gradient descent (also referred to as steepest descent) is a first-order iterative optimization algorithm for finding a local minimum of a differentiable function that takes repeated steps in the opposite direction of the slope (or approximate slope) of the function at the current point, since this is the direction of steepest descent. Gradient descent optimization may be performed using commercially available or open source artificial intelligence or high-resource computing tools, such as Google TensorFlow. Optimization algorithms can also be used to aid in guess decomposition when the trough or negative amplitude of the signal or sub-signal is not found.

[0114] After the decomposition is optimized, the program logic module (e.g., automatically or in conjunction with an expert operator) may then perform subsequent rounds of guess decomposition and optimization to identify and address potential errors or defects in the decomposition and form an optimized decomposition. Potential errors or defects may, in some embodiments, represent unlikely or impossible physical situations, results that are apparent or likely mathematical errors, an overly complex solution, and / or other results that point to an improper decomposition or optimization. Corrections made to potential errors or defects in the decomposition may be further incorporated into the program logic module such that the program logic module may be better able to identify situations in which such potential errors or defects may arise due to signal characteristics, and therefore may be able to perform a more efficient decomposition without generating solutions with such potential errors or defects.

[0115] After addressing potential errors or imperfections, it may be desirable to repeat the decomposition inference and optimization steps to arrive at a different, modified, or new optimized decomposition that does not include or reduces the potential errors or imperfections that were addressed. The process can be repeated as necessary to eliminate or reduce to a desired level any potential errors or imperfections in the decomposition.

[0116] The optimized decomposition can further be compared to other previous decompositions (e.g., as a whole, by numerical metrics, by common features, by measured physical parameters, etc.) to assist the clinician in making decisions or to reveal information about the structural features of the object.

[0117] In some embodiments, the object may be real, artificial or simulated oral tissue, such as teeth, intraoral restorations, appliances, implants or splints, and / or associated tissues or orthopedic implants. In general, the impact measurement device may apply mechanical energy to the object by percussion and may measure the energy returned to the device, such as by measuring the force / energy / displacement etc. returned to the impact measurement device, such as by measuring force, energy, displacement or other physical return value in a sensing mechanism over a period of time.

[0118] In other embodiments, the object may be a composite, which may include, but is not limited to, mechanical, industrial structures, or polymeric or metallic composite structures including honeycomb or layered honeycomb; industrial structures including, but not limited to, aircraft airframes, automobiles, ships, bridges, tunnels, trains, buildings, power generation equipment, arch structures, or other similar physical structures.

[0119] In general, as described above, the system of the present invention may include a program logic module incorporating a machine learning algorithm trained on a large dataset of signals to arrive at an optimized decomposition of the sub-signals or parts thereof. The training of the program logic module may generally occur in a controlled or pre-production environment, such as a laboratory, manufacturing or development environment, prior to use of the system by an end user, e.g., a dental practitioner or other clinical / industrial clinician in a non-training or non-examination setting. Of course, signals collected from actual operating settings (i.e., clinical signals) may be included in or extend the pre-existing dataset, as discussed below, in order to continue to improve the system through additional training on the extended dataset and make it live. The dataset may generally be based on a single type or related group of types of objects, such that training may result in a program logic module applicable for a particular field or application, such as the dental field. In some embodiments, the objects may be real, artificial or simulated oral tissues, such as teeth, intraoral restorations, appliances, implants or splints, and / or related tissues or orthopedic implants. In general, an impact measurement device may apply mechanical energy on an object by percussion (if not simulated on a computer) and measure the energy returned to the device after impact with the object or deceleration of the impactor, such as by measuring the force / energy / displacement etc. returned to the impact measurement device, such as by measuring the force, energy, displacement or other physical return value in a sensing mechanism over a period of time to form a signal. For computer-simulated objects, the energy return may be simulated, such as via Finite Element Analysis (FEA), as illustrated using the FEA model shown in FIG. 4d, or a constructed signal or sub-signal may be created on the computer or by manipulating / modifying a pre-existing signal or sub-signal.

[0120] The present invention may also include a simulation model component that may be utilized in training the program logic modules. FIG. 9 illustrates an example of an FEA model used as a physical simulation model. This analysis method may include the use of a numerical model to simulate an actual inspection using the devices described herein. In general, modeling and simulation may be desirable to train a system, whose predictive capabilities using a simulated model that may embody an inspection object that has not been inspected are not readily available for actual physical inspection, etc.

[0121] In one example of modeling, a physiologically accurate 3D model of a mandibular second molar was created using a solid modeling computer-aided design program using 3D x-ray computed tomography tooth data, although the same process may be applicable to other teeth as well as other solid objects. The model includes both enamel and dentin along with the pulp cavity, periodontal ligament (PDL) and surrounding bone, an example of which is shown in Figure 4d.

[0122] The solid model was then exported to a computer-aided engineering program for interlocking solids. A nonlinear finite element solver suitable for modeling nonlinear material behavior, such as that reported for PDL, as well as transient environmental conditions including impact, was used. In order to fully analyze the impact event with comparison to experimental data, it was essential to include an impact bar in the simulation model. The elastic modulus of the impact bar, its mass and initial velocity were entered into the program. The resulting impact force was measured by a piezoelectric sensor in the bar.

[0123] The FEA model may contain a large number of elements, for example, about 500,000 to about 1,000,000 elements each. Second-order isoparametric three-dimensional four-node tetrahedrons for the PDL, eight-node isoparametric arbitrary hexahedrons for the impact probe, and linear isoparametric three-dimensional tetrahedrons for the remainder of the model may be used. Boundary conditions may be defined to minimize or prevent free body motion, such as elements on the outer surface of an object, such as bone, may be constrained. The model was run with time increments such as 4 μs.

[0124] A direct integration method can be used to obtain a solution to the equations of motion for the model. In addition, viscous damping can be included in the analysis using classical Rayleigh Damping (RD), which is convenient for an incremental approach to numerical solution. A damping matrix D is defined as a linear combination of the mass and stiffness matrices of the system, and damping coefficients are specified for each element. Rayleigh damping uses coefficients on the element matrices and solves the equations

number

[0125] For example, and without limitation, a particular data set may include data derived from measurements of multiple types of teeth, dental implants / appliances, and / or oral tissues in the dental field, preferably with diverse objects or conditions, so that the trained program logic module may be exposed to multiple possibilities of its optimized decomposition (which may also include simulations of such objects or data derived therefrom). The data set may also be updated or expanded by end users (i.e., clinicians) through continued use of the system in a clinical / industrial setting, so that the program logic module may be trained over time with expanded data sets as it is used to improve its performance. The improved program logic module may then be propagated for utilization by various clinicians via updates (e.g., via updates to the cloud stored / operated portion of the system).

[0126] In an exemplary embodiment of the invention, signals may be measured from an impact measurement device percussing an object or a simulation of such signals and may generally be grouped together based on at least one common feature in a data set. Examples of such features may include object type, object size, location within a given space or environment, physical condition (e.g., damage, amount / type / location of physical repair, etc.), location of measurement with the impact measurement device, object age, treatment or procedure performed on the object, and / or any other applicable physical condition or simulation thereof. Groupings are also not limited and may not need to be exclusive, and multiple, different, overlapping, ad-hoc and / or complex groupings may be used.

[0127] In an exemplary embodiment of the invention, the program logic module may generally apply an algorithm, e.g., a machine learning trained algorithm as described above, to form at least a portion of an initial guess of the decomposition of the signal into at least one of its component sub-signals. Further algorithms of the program logic module, e.g., a machine learning trained algorithm, an FFT or Fourier-like operation, a gradient descent or similar operation, may also be utilized in conjunction to form a remainder of the initial guess decomposition, if applicable. The result may generally be an initial guess decomposition, as shown in FIG. 4b, with the complete signal decomposed into eight component sub-signals and a remaining sub-signal (which may represent, e.g., a noise, very small or insignificant portion of the signal). The program logic module may then perform an optimization, generally as discussed above, to minimize the difference between the initial guess decomposition and the original signal, such as through an optimization algorithm such as a gradient descent or similar algorithm (e.g., as implemented using a commercially available or open source artificial intelligence or high resource computing tool, e.g., Google TensorFlow, etc.). Optimization algorithms may also be utilized to aid in guess decomposition when the trough or negative amplitude of the signal or sub-signal is not found.

[0128] After the decomposition is optimized, as generally discussed above, the program logic module (e.g., automatically or in conjunction with an expert operator) may then perform subsequent rounds of guess decomposition and optimization to identify and address potential errors or defects in the decomposition and form an optimized decomposition. Potential errors or defects may, in some examples, represent unlikely or impossible physical situations, results that are apparent or likely mathematical errors, an overly complex solution, and / or other results that point to an improper decomposition or optimization. Corrections made to potential errors or defects in the decomposition may further be built into the program logic module such that the program logic module may be better able to identify situations where such potential errors or defects may arise due to signal characteristics, and therefore may be able to perform a more efficient decomposition without generating a solution with such potential errors or defects.

[0129] After addressing potential errors or imperfections, it may be desirable to repeat the decomposition inference (e.g., using a machine learning trained algorithm) and optimization steps to arrive at a different, modified, or new optimized decomposition that does not include or reduces the potential errors or imperfections that were addressed. The process can be repeated as necessary to eliminate or reduce to a desired level any potential errors or imperfections in the decomposition.

[0130] The properties and methods used to arrive at the optimized decomposition may then be incorporated into the system by codifying them in an algorithm, e.g., a machine learning or deep learning algorithm, so that the system can more efficiently and accurately decompose new signals encountered, e.g., signals acquired from a crash measurement device generating signals from physical objects. The system may also be able to apply the codified algorithm to new simulated data sets, e.g., data sets simulating hypothetical or new physical features / scenario.

[0131] In some embodiments, the system of the present invention can not only measure and analyze the structural features of the object being measured, but can also be trained to detect when the correct energy application parameters are used in a clinical setting. For example, using machine learning, the system can detect that the detected response indicates that, for example, for a physical tool such as a percussion stick, too much force is being applied, that the duration of each application is too short or too long, that the percussion may not be applied in the correct location, etc., or a combination of the above, and can automatically adjust the device settings to compensate or instruct the clinician to strike the object in a different location to produce a more optimal response. The system can also recognize these situations as a result of a high level of uncertainty or lack of signal detected during the decomposition process.

[0132] In general, without being limited or constrained by any particular theory, the energy or force returned to the crash measurement device may form a signal with only positive amplitude relative to a reference value (e.g., the x-axis of the signal), and the portion of the signal with negative amplitude relative to the reference value may be missing. In general, some sensing mechanisms may be essentially limited to detecting a response signal only in a single direction, e.g., in the direction of the indentation of the piezoelectric force sensor (i.e., since the piezoelectric element usually only generates a signal in response to the indentation). This may generally result in at least some signal loss or other energy profile from the signal, where the amplitude includes some negative portion relative to the reference value, e.g., the application of energy to the object at the onset of the impact, as any response from the object that causes the object to oscillate away from the sensing mechanism, or where mechanical contact or connection is lost and the sensing mechanism may not detect a portion of the signal due to the absence of indentation in the appropriate direction detected by the sensing mechanism. Therefore, the signal generated by the crash measurement device may be partially incomplete. Also, such sensing mechanisms are typically one-dimensional and inherently unable to filter or separate distinct sub-signals that may be present within the overall received signal, as they may typically be superimposed on top of one another. Methods exist for decomposing such signals into their component sub-signals, but the "missing" portions of the signal may present challenges, as standard decomposition methods will not accept signals with missing portions. Also, in some crash measurements, the length of time over which a signal is measurable may be too short to capture sufficient periods (or partial periods) of any signal or its component sub-signals for accurate decomposition, presenting additional challenges.

[0133] In one exemplary embodiment of the invention, the system may be adapted to accept or analyze a signal generated from an impact measurement on an object where the sensing mechanism results in at least a partial limitation or loss of the return signal. Generally, some sensing mechanisms may be inherently limited to detecting a response signal only in a single direction, e.g., in the direction of compression, such as in the arrangements shown for sensing mechanism 111 (e.g., a piezoelectric force sensor) in Figures 2, 2a, 2c, 2d, and 2e. This may generally result in at least some loss of signal, as discussed above, which may result in a signal generated by the impact measurement device that may be partially incomplete.

[0134] In general, without being limited or constrained by any particular theory, in the dental field, recognizing that the PDL generally forms the majority of the attenuation from the impact and that the energy / force return as measured in the signal should appear as a sinusoid (e.g., like a sound wave) and more generally forms only a single peak of an approximately Gaussian shape, it can be assumed that the PDL response may be present in the signal as the first half of a sinusoid with a "negative" amplitude. Therefore, the PDL side signal can generally be assumed as a significant side signal component of the signal and then removed, treating it as a carrier of the remaining side signals and allowing their resolution using generally more standard decomposition methods such as fast Fourier transform or similar Fourier-like operations. This can be particularly useful since the remaining side signals, if any, are generally sinusoids with shorter periods and / or amplitudes (e.g., decaying sinusoids) that may be obtained in the measurement but may be obscured or mischaracterized as other Gaussian shapes due to the presence of a larger single half-period PDL side signal. In some situations, the PDL sub-signal (if any) may be recognized as not forming a significant sub-signal (e.g., when the PDL is damaged, weakened, absent, etc.), and it may be determined or assumed that another physical feature (e.g., a crack or other damage in a tooth or tooth-like structure) may form a significant sub-signal rather than the PDL.

[0135] In some exemplary embodiments, the system may utilize machine learning methods to process and / or analyze signals that are essentially missing negative amplitude portions and attempt to reconstruct or treat the signal as a missing or partial rather than a complete response from the measurement. In the dental field, machine learning algorithms and methods may be trained on a large set of collected impingement data (e.g., ERGs) from a wide range of teeth with different characteristics, e.g., different types (e.g., incisors, bicuspids, canines, molars, etc.), sizes, number of roots, different degrees of physical damage (e.g., fractures, cavities, etc.), degree or type of restoration (e.g., crowns, fillings, etc.), age, etc., to train the algorithms and methods to be able to decompose newly encountered signals into component sub-signals, e.g., a collection of sinusoidal sub-signals that form the signal or an approximation thereof.

[0136] FIG. 3 shows an example of a signal (shown as ERG or FRG) generated by percussing an object (e.g., a tooth) with a device as shown in FIGS. 2, 2a, 2c, 2d and 2e and registering the force returned to a piezoelectric force sensor (e.g., a sensing mechanism 111 as below). As shown, the signal does not register a signal above a given threshold (i.e., the X-axis) due to the loss of a portion of the initial signal created by the object being percussed. In general, the signal returning from the percussed object may form a rather sinusoidal shape (e.g., similar to a sound wave as shown in FIG. 4) that can be rectified or "cut off" by the sensing mechanism (e.g., a piezoelectric force sensor) to generate only the portion of the signal above the threshold (e.g., shown as a cut off signal in FIG. 4a).

[0137] For example, the data set utilized in training machine learning algorithms or in comparison with clinical signals as discussed above and below may also be grouped in other ways, for example by the location of the impact on the object (e.g., of the tooth, such as cheek or mesial, distal or proximal end), by the location of the object relative to other reference points (e.g., mandible vs. maxilla in the oral cavity), by the amount / frequency / number of impacts, or by any other suitable type of grouping. In such an embodiment, the signal may be analyzed and processed as at least one signal of which a portion is missing from the measurement (e.g., the lower half of the sinusoid below a given threshold is missing from the signal, such as the second half of the sinusoid that may extend below the threshold forming a Gaussian-like shape of the first half, as shown in the signal of FIG. 3a). This may result in a better interpretation of the data, since the entire response (or at least an approximation to account for the missing portion of the signal) may then be considered, rather than treating the signal as only being present in the signal (i.e., assuming that the signal is complete without the missing portion of the signal). Furthermore, some data analysis operations may not be able to properly interpret the data if parts of the signal are missing (eg, standard Fast Fourier Transforms or other similar Fourier-like operations).

[0138] In an exemplary embodiment of the invention, the signal may be recognized, analyzed and / or processed as a sum or conglomeration of multiple different sub-signals generated by the interaction of the energy applied to the tooth and the structural features of the tooth and / or surrounding tissues / structures. Without being bound to a particular theory, the signal may represent a number of different sub-signals generated by separate physical structures or features of or around the tooth, respectively, resulting in a sum or conglomeration of sinusoids to form the overall shape of the signal (or a portion (full signal) that is detected, i.e., has no gaps or "bottoms" below a given threshold of detection or measurement, as illustrated by Figs. 3, 4a and 4b), each of which may take an approximation of a sinusoid, as illustrated by the sub-signals in Fig. 4b, such as, but not limited to, a decaying sinusoid (e.g., an exponentially decaying sinusoid in response to dissipation due to the material / structure / movement of the object), as illustrated by the decaying sinusoid-like shape of Figs. 4 and 4b. In some embodiments, the signal may be decomposed into a "sparse" or limited number of sinusoids (i.e., a bounded number of sinusoids to give the original signal or an approximation thereof) to understand, without being bound to a particular theory, that each sinusoid is caused by at least one element of the tooth, its restoration if any, and / or the surrounding tissue. One example of a sparse decomposition is illustrated in FIG. 4b, with the fully unresolved portion shown with the remainder being explained.

[0139] In some embodiments, the prominent Gaussian shape in the signal from the tooth, especially from the initial or intact tooth, may generally be recognized as being mainly due to the response of the PDL (periodontal ligament), shown in FIG. 4e, which absorbs the transfer energy from the energy application tool, which may generally appear similar to the single Gaussian-like shape of FIG. 3a. For an intact healthy tooth, the impact energy generated by chewing is attenuated by the PDL at the interface between the healthy bone and the natural tooth. Even in cases where the PDL may be absent or defective, the portion of the signal referred to as the PDL portion will refer to that produced due to the fixation of the tooth or an implant directly or indirectly inside the bone, which produces a roughly pendulum-like response. In general, the PDL portion of the signal may constitute a majority of the amplitude of the signal, as illustrated with the single peak signal of the PDL portion in FIG. 4c, and may generally be interpreted as a carrier wave that may be used to isolate and / or separate the response from other elements of the tooth or surrounding tissue, such as by subtracting the PDL portion from the signal.

[0140] In some exemplary embodiments, sinusoidal decomposition of a signal using machine learning methods may generally include the steps discussed above with respect to more general signals and their sub-signals, with some characteristics of sinusoids to deal with dental objects such as teeth and implants. For example, the steps may generally include: finding an initial guess for a non-tapering of the signal (e.g., to remove signal deformations near the x-axis of the signal, e.g., deformations due to an energy application tool sticking to the tooth); finding an initial guess for a larger and / or more pronounced sinusoidal component of the signal (e.g., a sinusoid or other prominent side signal typically generated by a PDL); finding an initial guess for the remaining sinusoidal components of the signal (e.g., from cracks, damage, separations between layers, or other features) including a guess at the frequency (e.g., via a Fourier transform-like operation such as a Fast Fourier Transform (FFT) on the signal after subtracting the initial PDL sinusoidal guess, via machine learning or artificial intelligence methods, gradient descent-based methods, etc.); performing an optimization to minimize the difference between the initial signal and the resulting sum of the initial guess sinusoids of the signal to produce a candidate sinusoidal decomposition; identifying / fixing decomposition defects or errors (e.g., unlikely or negatively indicated decomposition results) and returning to the decomposition steps described above as necessary to remove them; and culling the best or other desired candidates from the resulting decompositions. The resulting decomposition and associated data / results / visualizations may then be displayed or output, e.g., in a human readable format, so that a practitioner or other user may use or interpret them for clinical diagnosis, monitoring and / or treatment planning, etc. The decomposition generated by the system may also generally include the determination or calculation of uncertainty measures at various steps of the decomposition, e.g., to calculate the value of the error at various steps or of a particular calculation in the decomposition.

[0141] In some embodiments, the resulting sum of the sub-signals may not approximate the initial signal well or at least not in certain parts. For example, since the initial signal may generally only have positive amplitudes above its reference value, the possible resulting sum of the sub-signals may include some that are negative amplitudes, and during optimization, it may be desirable to arbitrarily select or take the maximum value between 0 and the resulting sum value (i.e., max(0,sum)) in those parts of the negative amplitude to aid in optimization. In other examples, a sensing mechanism or other physical arrangement of the impact measurement device may be utilized that may be able to capture negative amplitude parts that a directionally polarized setup (i.e., a piezoelectric sensing element as discussed above) cannot capture, for example, with a strain gauge-based sensing mechanism or other unmodified sensor, or with the design of the impact measurement device. In such cases, a max(0,sum) operation may generally be undesirable since the resulting signal may not be restricted to positive amplitudes.

[0142] In some exemplary embodiments, the program logic module may be trained using the machine learning or artificial intelligence methods discussed above with respect to training to perform initial and / or subsequent guesses (i.e., after the optimization step) of prominent side signals in the signal to arrive at a complete decomposition of the signal into side signals, where a sinusoidal decomposition of the remainder is performed using the program logic module or other resources of the system's computing device (e.g., a local computer or cloud service) to enable the system to generate a remainder of the signal without prominent side signals. In general, without being bound by a particular theory, a number of signals may contain prominent side signals that, when determined and removed from the signal, produce a remainder of the side signals that can be determined without complex machine learning or artificial intelligence methods, for example, by utilizing FFT or Fourier-like operations, gradient descent-based methods, etc., so that intensive resources for machine learning or artificial intelligence operations can be conserved. For example, removal of prominent side signals (e.g., PDL side signals) may result in more easily identifiable side signals (e.g., sinusoids, etc.) by not overlooking significant portions, "negative" amplitude portions, or other features of the side signals, which may require or benefit from machine learning or artificial intelligence methods, which may be used as needed.

[0143] In some embodiments, a program logic module or another component of the system (e.g., a local computer or cloud service) may perform an initial and / or subsequent guess (i.e., after an optimization step) of significant side signals, such as by performing a fit of at least a portion of the signal to a basis function (e.g., a Gaussian curve, a sinusoid, a sinusoid-like curve, etc.).

[0144] For example, the program logic module may generally apply an algorithm to fit the signal to basis functions, such as by utilizing gradient descent optimization, Levenberg-Marquardt optimization, and / or any other similar or suitable method or combination / multiple thereof. The system may then generate a residual of the signal without significant side signals, where a sinusoidal decomposition of the residual is performed on the program logic module of the system or other resources of the computing device (e.g., a local computer or cloud service) to arrive at a complete decomposition of the signal into side signals. In general, without being bound by a particular theory, a number of signals may contain significant side signals that, when determined and removed from the signal, produce a residual of the side signals that can be determined without complex machine learning or artificial intelligence methods, for example, by utilizing FFT or Fourier-like operations, gradient descent-based methods, etc., so that intensive resources for machine learning or artificial intelligence operations can be conserved. For example, removal of prominent side signals (e.g., PDL side signals) may result in more easily identifiable side signals (e.g., sinusoids, etc.) by not overlooking significant portions, "negative" amplitude portions, or other features of the side signals, which may require or benefit from machine learning or artificial intelligence methods. The result may generally be an initial guess decomposition, as illustrated in FIG. 4b, in which the complete signal is decomposed into eight component sub-signals and the remaining sub-signals (which may represent, for example, noise, very small, or insignificant portions of the signal). The program logic module may then perform optimization, generally as discussed above, to minimize the difference between the initial guess decomposition and the original signal, such as through an optimization algorithm, e.g., gradient descent or similar algorithm (e.g., as implemented using a commercially available or open source artificial intelligence or high resource computing tool, e.g., Google TensorFlow, etc.). If the bottom or negative amplitude of the signal or sub-signal is missing, an optimization algorithm may also be utilized to aid in the guess decomposition.

[0145] After the decomposition is optimized, as generally discussed above, the program logic module (e.g., automatically or in conjunction with an expert operator) may then perform subsequent rounds of guess decomposition and optimization to identify and address potential errors or defects in the decomposition and form an optimized decomposition. In some embodiments, the potential errors or defects may represent unlikely or impossible physical situations, results that are apparent or likely mathematical errors, an overly complex solution, and / or other results that point to an improper decomposition or optimization. Corrections made to potential errors or defects in the decomposition may further be built into the program logic module such that the program logic module may be better able to identify situations in which such potential errors or defects may arise due to signal characteristics, and therefore perform a more efficient decomposition without generating a solution with such potential errors or defects.

[0146] The decompositions derived from the acquired signals, e.g., clinical signals from measurements in a clinical setting, may be compared in various ways to and analyzed in conjunction with a dataset (e.g., a large dataset of signals or information / metrics derived from signals from objects similar to the target object in the clinical signal). The dataset may be similar in form or content to the datasets discussed above with respect to training machine learning algorithms, or may be adapted, truncated, extended, or formed from different sources. In some embodiments, the objects embodied in the signals of the dataset may be real or artificial, or may be simulated, such as with oral tissues such as teeth, intraoral restorations, appliances, implants or splints, and / or associated tissues or orthopedic implants. In general, as discussed above, the impact measurement device may apply mechanical energy on the object by percussion (if not simulated on a computer) and measure the energy returned to the device after impact with the object or deceleration of the impactor, such as by measuring the force / energy / displacement etc. returned to the impact measurement device, such as by measuring the force, energy, displacement or other physical return value on a sensing mechanism over a period of time to form a signal. For computer-simulated objects, the energy return may be simulated, such as via Finite Element Analysis (FEA), as illustrated using the FEA model shown in FIG. 4d, or a constructed signal or sub-signal may be created on the computer or by manipulating / altering a pre-existing signal or sub-signal. For example, and without limitation, a particular data set may also include or have available data derived from measurements on multiple types of teeth, dental implants / appliances, and / or oral tissues in the dental field (which may also include simulations of such objects or data derived therefrom), preferably in a variety of subjects or conditions, so that comparisons or other analyses can be performed with multiple possibilities of optimized decomposition from the clinical signal.The dataset may also be updated or expanded through continued use of the system by end users (i.e., clinicians) in clinical / industrial settings to provide additional data for comparison or analysis. Improved datasets may be propagated for use by various clinicians via updates (e.g., via updates to cloud-stored / operated portions of the system).

[0147] In an exemplary embodiment of the invention, signals may be measured from an impact measurement device percussing an object or a simulation of such signals and may generally be grouped based on at least one common feature in the data set. Examples of such features may include object type, object size, location within a given space or environment, physical condition (e.g., amount / type / location of damage, physical repair, etc.), location of measurement with the impact measurement device, object age, treatment or procedure performed on the object, and / or any other applicable physical condition or simulation thereof. Without limitation, groupings may also not need to be exclusive and multiple, different, overlapping, ad-hoc and / or complex groupings may be used.

[0148] In another exemplary aspect of the invention, the system may generate or calculate various numerical metrics from the decomposition of signals that show statistically significant differences in a dataset such that these numerical metrics may be utilized in a probability distribution or heat map to aid in the prediction or detection of different physical features or attributes by comparison with the numerical metrics derived from the decomposition in a clinical setting. For example, a numerical metric generated from a dataset containing known physical features (e.g., dental damage type) may be used for comparison using the same type of numerical metric derived from clinical measurements, and the probability or degree of matching may be determined by the system to output the likelihood of a match with that particular physical feature.

[0149] The types of numerical metrics may include, but are not limited to, normal fit error (NFE), which is the total error (difference) between an ideal curve (e.g., produced by a defect-free object) and the actual inspection data. These results may be calculated from the signal. In the case of decomposition of sinusoids or other side signals from the initial complete signal, the NFE may be calculated for the decomposition (i.e., sinusoid decomposition NFE or SDNFE) as the error between the prominent sinusoid being decomposed (e.g., PDL sinusoid) and the complete signal. Other numerical metrics may include PDL sub-periods or frequencies, frequencies or periods of other sinusoids, caries percentage of exponentially decaying sinusoids, sinusoid amplitudes, number of periods in a sinusoid, other statistically significant metrics, and / or weighted versions / combinations of any of the above (i.e., to account for tapered weighting of nearby values ​​or distributions of values).

[0150] 5, 5a and 5b illustrate examples of heat maps showing the distribution of numerical metrics across populations. As shown, with circles showing weighted averages ordered according to size by standard deviation, darker areas indicate higher totals and lighter areas indicate lower totals. Such heat maps may be desirable or useful in assessing the likelihood of a measurement matching some physical feature or combination of features based on a cohort in a dataset of a system that uses those features or combinations to identify. For example, for illustrative purposes, FIG. 5 may describe a general population in a dataset, while FIG. 5a may describe a cohort (e.g., a group of teeth identified as having a particular type of damage) overlaid on the general population, while FIG. 5b may describe another cohort (e.g., a group of "good" or healthy / intact teeth overlaid on the general population). The placement of measurements applied to these heat maps may then help determine whether a measured object has a likelihood of fitting into one population or another.

[0151] In general, heatmaps may be of any suitable dimension (e.g., one-dimensional, two-dimensional, three-dimensional, etc.). Some higher dimensions may be difficult for humans to visualize or interpret, and therefore computer-controlled interpretation or reduction to a numerical or simplified graphical representation may be used to aid the human user's interpretation.

[0152] In another aspect of the invention, the machine learning algorithms and methods of the system, which may be generally separated or segregated by use, training or purpose from the machine learning algorithms discussed elsewhere, for example, with respect to signal decomposition or other uses, may be trained on large sets of collected impaction data that may be annotated with features (which may be determined by an "expert" or other trusted characterizer or via a machine learning algorithm) to identify, infer with a degree of probability and / or associate with a particular feature the measured signal, or to select an appropriate method of further analysis or algorithmic manipulation to produce a useful output for a user to utilize or interpret for selection of an appropriate plan of further diagnosis, monitoring, treatment, etc. For example, signals from a data set may be correlated with certain physical features as determined by an expert, such as a dental practitioner who identifies the type or extent of tooth damage by physical examination, x-ray imaging, deconstruction, etc., which the system may utilize as additional correlation data for the signal. Correlations with training and annotations on these groupings can be further exploited in machine learning assisted analysis of heat maps of numerical metrics, as discussed above, such as by using machine learning algorithms trained from such to find correlations and possible matches to populations. This may be desirable in heat maps or comparisons where dimensionality or other complexities pose challenges to human interpretation.

[0153] The comparison may also be performed using machine learning algorithms to help increase efficiency and the probability of matching with known data sets. For example, basic machine learning methods such as kernel density estimation and / or calibration curve fitting Bayesian networks may be used. More advanced comparison methods may also be generated using deep learning methods after the machine learning system has been fully developed and / or trained.

[0154] In some embodiments, the system may detect numerical metrics in the clinical measurement that may instruct or suggest the user to change some physical parameters of the clinical measurement, such as by changing parameters of the impact measurement device, to generate better or more accurate data. For example, some numerical metrics may instruct or suggest the use of different impact forces, frequencies or positions on the impacting object to reveal additional information or to improve the quality of the measurement with the impact measurement device. The system may use information not directly derived from the signals in its data set, such as the location of the impact on the object, the impact force, the impact measurement device settings during the measurement, and / or other relevant data or factors. The system may then provide feedback for the control of the impact measurement device to suggest or implement a repeat measurement with the different settings, locations, timing, etc. The detection of such suggestions / indications by the system may be trained into a program logic module using machine learning methods similar to those discussed above. This helps to minimize subjective decisions made by the clinician that may not be possible without the present invention.

[0155] 2a in the housing 132, the drive mechanism 140 supported within the housing may receive commands from the system to activate the energy application tool 110 between a rest configuration and an activated configuration to apply a set amount of energy in a horizontal orientation, typically using an inclinometer adapted to measure the inclination of the energy application tool 110 relative to the horizontal. For a given object, the drive mechanism 140 may vary the amount of energy applied to activate the energy application tool 110 between the rest configuration and the activated configuration based on the inclination up to at least approximately the set amount of energy at an inclination other than horizontal. Therefore, the same drive mechanism 140 described above for varying the amount of energy applied (e.g., by varying the voltage, current, or both) may be applicable for the different types of objects described above to vary the coil drive time (by varying the length of time the coil is energized or activated), vary the coil delay time (by varying the time between drive activities), vary the number of coil energizations (i.e., by varying the number of drive pulses applied), vary the coil polarity and / or combinations thereof, and adjust the energy application process to simulate a substantially horizontal position during measurement. These factors, including varying the power, drive time, polarity, and delay time, may be managed through changing firmware settings for the power, drive time, number of drives, number of drive pulses, polarity, and drive delay of the coil energization for the desired result. Without wishing to be bound by any particular theory, it is believed that numerous variations may be used to achieve the desired result and the firmware may be designed to select a particular solution, or in some cases, to select the optimal solution. Transformations may be suggested, adjusted or otherwise indicated to the user by the system based on program logic modules that detect or calculate possible changes in the physical parameters of the measurement to help improve accuracy or reveal more information about the object. The system may perform some or all of these functions automatically, or may alert or prompt the user to adjust the parameters.

[0156] Generally, when natural teeth are replaced with implants due to injury or disease, the ligaments are largely lost. However, as described above, the systems and methods of the present invention can also be used to evaluate the structural characteristics of implant structures that use abutments. Some materials used for abutments, such as composites, gold, and zirconia, can produce side signals that are somewhat similar to the PDL response.

[0157] Furthermore, the system and method may be useful to measure the dynamic response when a force is applied to the abutment material, and may also be useful in predicting the suitability or compatibility prior to implant treatment, or in selecting appropriate materials to protect the natural teeth adjacent to the implant and making better selections of materials to minimize differences between the implant and the natural teeth in response to impacts, which may improve the effectiveness of the abutment construction, increasing the selection of materials or material combinations that may be suitable, leading to better patient care.

[0158] In some embodiments, device measurements and / or expert annotations may be stored using a distributed computing environment, such as the cloud. Storage, such as in the cloud, may allow multiple expert annotations to be collected simultaneously, reducing the time to accumulate an expert annotation dataset to increase prediction accuracy. In some embodiments, device measurements and / or expert annotations may be collected at multiple instances of the system and integrated into one or more of the instances. In some embodiments, device measurements and / or expert annotation entries may be encrypted.

[0159] As previously mentioned, machine learning techniques may include regression (e.g., logistic, linear), clustering (e.g., k-means), neural networks (e.g., deep learning), classifiers (e.g., support vector machines, decision trees, random forests), deep learning, etc. While the basic machine learning techniques utilized may themselves be standardized techniques and may not be unique in themselves, the inventors have discovered certain unique adaptations to the types of data stored in case files to make them useful for machine learning algorithms. For example, signals produced by systems and methods as described herein above and below may be used to measure and evaluate structural features of objects, whether anatomical or non-anatomical, in a non-invasive manner and / or using non-destructive methods of measurement. Structural features of an object may be identified based on measurements of the same or other objects previously made and captured using a system using devices such as those illustrated in Figures 2, 2a, 2c, 2d, and 2e and their corresponding descriptions, or those described in U.S. Patent Nos. 6,120,466, 7,008,385, 6,997,887, 9,358,089 9869606, U.S. Pat. No. 10,488,312, PCT / US17 / 69164, PCT Patent Application Docket No. PCT / US20 / 40386, U.S. Patent Publication Nos. 20190331573, PCT / US2018 / 068083, and / or PCT Publication WO2019133946, which are incorporated by reference in their entireties, and may be filtered and converted into a spectrogram for use in deep learning. The models can then be trained, versioned, and stored in a secure database running on a set of centralized cloud-based servers.

[0160] Example of a sinusoidal decomposition optimization algorithm An algorithm may be utilized to optimize the sinusoidal decomposition, such as by utilizing gradient descent optimization, which may utilize an optimization tools package (e.g., Google TensorFlow, etc.) in some exemplary embodiments. One example of a gradient optimization algorithm for use with such an optimization tools package is shown below: Optimization Algorithms Minimize the squared error {α i}, {p i}, {o i}, {d i} find:

number

number

[0161] Examples of how to detect and address potential errors / defects The table below describes examples of potential problems with the decomposition performed by the program logic module, such as during training or when decomposing clinical / industrial measurement signals, the uncertainties such problems create, how they can potentially be detected, and how they can be addressed by the system.

[0162] [Table 1]

[0163] Although the invention has been described with respect to specific aspects, embodiments, and examples thereof, these are illustrative only and do not limit the invention. The description herein of the illustrated embodiments of the invention, including the description in the Abstract and Summary of the Invention, is not intended to be exhaustive or to limit the invention to the precise form disclosed herein (in particular, the inclusion of any particular embodiment, feature, or function in the Abstract or Summary of the Invention is not intended to limit the scope of the invention to such embodiment, feature, or function). Rather, the description is intended to describe the illustrated embodiments, features, and functions to provide those skilled in the art with a context for understanding the invention, without limiting the invention to any specifically described embodiment, feature, or function, including such embodiment, feature, or function described in the Abstract or Summary of the Invention. Specific embodiments and examples of the invention are described herein for illustrative purposes only, and various equivalent modifications are possible within the spirit and scope of the invention, as those skilled in the art will understand and appreciate. As indicated, in light of the foregoing description of the illustrative embodiments of the invention, these modifications can be made to the invention and should be included within the spirit and scope of the invention. Thus, while the invention has been described herein with reference to specific embodiments thereof, it will be understood that in the foregoing disclosure, liberal modification, various changes and substitutions are contemplated, and that in some instances, some features of the embodiments of the invention may be employed without a corresponding use of other features without departing from the scope and spirit of the invention as described. Accordingly, many modifications may be made to adapt a particular situation or material to the essential scope and spirit of the invention.

[0164] In general, references to the "cloud" can include both Internet-connected computing services and / or resources as well as those that may reside on a smaller or private network.

[0165] In general, the "program logic modules" and software elements are generally capable of being configured, run, stored, processed, and / or executed in any of the above, individually on different computer processors and / or memories, in combination with one another on the same computer processor and / or memory, and / or in various time arrangements, where applicable. Nothing herein should be implied or construed as requiring that any program logic module and / or software element be run on any one or combination of computing processors and / or memories, any suitable combination or unit may be utilized.

[0166] References throughout this specification to "one embodiment," "an embodiment," or "a specific embodiment," or similar terminology mean that a particular feature, structure, or characteristic described in connection with an embodiment is included in at least one embodiment, and may not necessarily be present in all embodiments. Thus, the individual appearances of the phrases "in one embodiment," "in an embodiment," or "in a specific embodiment," or similar terminology in various places throughout this specification do not necessarily refer to the same embodiment. Furthermore, the particular features, structures, or characteristics of any particular embodiment may be combined in any suitable manner with one or more other embodiments. It should be understood that other variations and modifications of the embodiments described and illustrated herein are possible in light of the teachings herein, and should be considered as part of the spirit and scope of the invention.

[0167] In the description herein, numerous specific details are provided, such as examples of components and / or methods, to provide a thorough understanding of embodiments of the invention. However, one of ordinary skill in the art will recognize that the embodiments may be practiced without one or more of the specific details, or with other devices, systems, assemblies, methods, components, materials, and / or parts, etc. In other instances, well-known structures, components, systems, materials, or operations have not been shown or described in particular detail to avoid obscuring aspects of the embodiments of the invention. Although the invention may be described using specific embodiments, this does not limit the invention to any particular embodiment, and one of ordinary skill in the art will recognize that additional embodiments are readily discernible and part of the invention.

[0168] As used herein, the terms "comprises," "comrising," "includes," "including," "has," "having," or any other variation thereof, are intended to cover a non-exclusive inclusion. For example, a list of elements includes a process, product, article, or device is not necessarily limited to those elements, but may also include other elements not expressly listed or inherent to such process, product, article, or device.

[0169] Furthermore, as used herein, the term "or" generally contemplates "and / or" unless otherwise indicated. For example, condition A or B satisfies any one of the following: A is true (or exists) and B is false (or does not exist), A is false (or does not exist) and B is true (or exists), or both A and B are true (or exist). As used herein, including in the claims that follow, terms preceded by "a" or "an" (and "the" if the antecedent is "a" or "an") include both the singular and the plural of such terms, unless clearly indicated otherwise in the claim (i.e., unless the reference "a" or "an" clearly indicates only the singular or only the plural). Also, as used in the description of this specification, the meaning of "in" includes "in" and "on," unless the context clearly dictates otherwise.

Claims

1. 1. A method for providing a machine learning trained structural feature analysis system, comprising: providing or generating a data set comprising a plurality of signals from a plurality of groups of different objects, the signals being generated from a series of impact measurements on the different objects and grouped based on a common characteristic of one of the groups of different objects; generating a set of optimized signal collections for each of the signals, each set of optimized signal collections comprising: performing a stochastic decomposition of each signal to generate a signal collection for each signal including at least one sub-signal; performing an optimization operation to minimize differences between the signal collection and each of the signals to generate an optimized signal collection; identifying and addressing potential errors or imperfections within each of the optimized signal collections; repeating the guess decomposition and the optimization operations to recreate an optimized signal collection after the potential errors or defects have been addressed; selecting at least one desired signal collection from the optimized signal collections for each signal to add to the set of optimized signal collections; and Incorporating the set of optimized signal collections and associated methods for arriving at the set of optimized signal collections into a machine learning trained analysis system (MLTA). said generating being generated by connecting the MLTA to a measurement device adapted to generate clinical signal data by percussing a target object and transmitting the clinical signal data to the MLTA to enable the measurement device to process the clinical signal data to produce a clinical optimized signal collection, compare characteristics of the clinical optimized signal collection with characteristics of the set optimized signal collection, and present results of the comparison in a human readable format; A method comprising:

2. 1. A machine learning trained structural feature analysis system, comprising: a housing having an open front end and a longitudinal axis; an energy application tool mounted within the housing, the energy application tool having a static configuration and an activated configuration; a drive mechanism supported within the housing, the drive mechanism adapted to activate the energy application tool between the rest configuration and an activated configuration to apply a set amount of energy; and a control mechanism connected to provide commands to said drive mechanism; the drive mechanism varying the amount of energy applied to activate the energy application tool between the rest configuration and the activated configuration based on input from the control mechanism. a collision measurement device comprising: a program logic module coupled to said control mechanism, said program logic module comprising: providing or generating a data set comprising a plurality of signals from a plurality of groups of different objects, the signals being generated from a series of impact measurements on the different objects and grouped based on a common characteristic of one of the groups of different objects; generating an optimized sub-signal collection for each of said sets of signals, wherein the optimized sub-signal collection for each of said sets comprises: performing a stochastic decomposition of each signal to generate a subsignal collection for each signal including at least one subsignal; performing an optimization operation to minimize differences between said sub-signal collection and each of said signals to generate an optimized sub-signal collection; identifying and addressing potential errors or imperfections within each of the optimized sub-signal collections; repeating the guess decomposition and the optimization operations to recreate an optimized sub-signal collection after the potential errors or defects have been addressed; selecting at least one desired sub-signal collection from the optimized sub-signal collections for each signal to add to the set of optimized sub-signal collections; and Incorporating the set of optimized sub-signal collections and associated methods for arriving at the set of optimized sub-signal collections into a Machine Learning Trained Analysis System (MLTA). said generating being generated by connecting the MLTA to the collision measurement device, the collision measurement device adapted to generate the clinical signal data by percussing a target object and transmitting the clinical signal data to the MLTA to enable the MLTA to process the clinical signal data to produce a clinical optimized sub-signal collection and a comparison between characteristics of the clinical optimized sub-signal collection and characteristics of the set of optimized sub-signal collections and to determine physical parameters associated with the comparison; the program logic module provided by a control regulator connected to the program logic module and the control mechanism, the control regulator adapted to output changes to the instructions in response to the MLTA outputting proposed changes due to the physical parameters; A machine learning trained structural feature analysis system comprising:

3. 1. A method for providing a structural signature analysis system, comprising: providing a program logic module (PLM) configured to interface with a crash measurement device (PMD) and to receive signal inputs for generating signals from crash measurements by said PMD; sending the input of a signal including data from impact measurements on one of a plurality of different objects using the PMD to the PLM and generating the signal using the PLM; performing an estimation of significant subsignals of said signal by fitting said signal to basis functions; subtracting the prominent side signal from the signal to form a remainder; performing a guess decomposition on the remainder to generate a secondary subsignal that sums with the significant subsignal to form an approximation of the signal; performing an optimization operation to minimize the difference between the approximation of the signal and the signal to generate an optimized collection of sub-signals; identifying and addressing potential errors or imperfections within each of the optimized sub-signal collections; and repeating the guessing, guess decomposition, and optimization operations of the significant side signals to regenerate the optimized side signal collection after the potential errors or defects have been addressed; selecting at least one desired sub-signal collection from the optimized signal collection; presenting said desired subsignal collection in a human readable format; A method comprising:

4. 4. The method of claim 3, wherein performing an optimization operation comprises minimizing a difference between each of the signals and a maximum of zero and the sum of each of the sub-signal collections to generate an optimized signal collection.

5. 3. The system of claim 2, wherein performing an optimization operation comprises minimizing a difference between the signal and a maximum of sums and zeros of each of the sub-signal collections to generate an optimized sub-signal collection.

6. The method of claim 1 , wherein the at least one sub-signal comprises a waveform from a periodontal ligament (PDL) attenuation response.

7. The method of claim 3 , wherein the inferential decomposition is performed using a machine learning algorithm for at least one member of the sub-signal collection.

8. The method of claim 1 , wherein the signal collection includes at least one sinusoidal subsignal.

9. 4. The method of claim 1, wherein the signal is generated from a crash measurement device that records only positive amplitude return signals relative to a threshold on a sensing element, and at least one sinusoidal waveform is determined by the inferential decomposition that interprets the signal as a waveform that may include negative amplitude missing signals from the signal.

10. The method of claim 1 or 3, wherein the guess decomposition further comprises performing a frequency guess of the at least one sub-signal.

11. The method of claim 10 , wherein the frequency estimation comprises a Fourier transform or a Fourier-like operation.

12. 10. The method of claim 9, wherein the guess decomposition further comprises performing a frequency guess of the at least one subsignal using a Fourier transform or a Fourier-like operation that is not optimized for sinusoids having missing subsignal portions below the threshold.

13. 2. The method of claim 1, wherein the common features of the group of different objects are selected from the group consisting of tooth type, tooth size, tooth age, degree or type of tooth restoration, degree or type of tooth damage, mandibular location, maxillary location, number of tooth roots, dental or orthodontic treatment, and location of impaction measurements on the objects.

14. 2. The method of claim 1, wherein the features of the clinical optimized signal collection and of the set optimized signal collection are selected from the group consisting of a periodontal ligament decay response (PDLP) period, a frequency of at least one of the sub-signals, a caries percentage of an exponentially decaying sinusoid, an amplitude of the sinusoid, a number of periods within the sinusoid, and combinations thereof.

15. The method of claim 9 , wherein the at least one sinusoidal waveform comprises an exponentially decaying sinusoid.

16. The method of claim 1 or 3, wherein the optimization operation comprises a gradient descent-based or related optimization.

17. The method of claim 1 , wherein the data set is subjected to a signal filtering operation to remove signal distortions.

18. The method of claim 1 or 3, wherein the stochastic decomposition comprises a sparse sinusoidal decomposition.

19. The method of claim 1 or 3, wherein the speculative decomposition further comprises calculating an uncertainty value for the signal collection.

20. The method of claim 1 , wherein the MLTA is adapted to use a machine learning algorithm to generate the results of the comparison.

21. 21. The method of claim 20, wherein the machine learning algorithm comprises a deep learning algorithm, a kernel density estimation, or a calibrated Bayesian network.

22. 4. The method of claim 1, wherein the different objects are selected from the group consisting of natural teeth, artificial or replica teeth, simulated teeth or oral tissue, dental restorations, oral tissue, dental appliances or implants, and dental splints.

23. 4. The method of claim 3, further comprising: performing estimation of the prominent side signals using a machine learning algorithm (MLA) on the PLM, the MLA being trained on a dataset including a plurality of signals with known prominent side signals and known sinusoidal decomposition.

24. A method described in any one of claims 1 and 3, wherein the guess synthesis is generated by gradient descent-based or related optimization, Fourier-like operations, machine learning algorithms or a combination of two or more thereof.

25. The method of claim 3 , wherein the basis functions are selected from the group consisting of Gaussian, exponential, sinusoidal, exponentially decaying sinusoidal, and sinusoid-like curves.

26. The method described in claim 3, wherein the prominent secondary signal includes a waveform from a periodontal ligament (PDL) attenuation response.

27. ​​A system as described in any one of claims 2 and 5, wherein the guess synthesis is generated by gradient descent based or related optimization, Fourier-like operations, machine learning algorithms or a combination of two or more thereof.

Images