Drill penetration detection

Abstract

The present disclosure relates to a detection device configured to detect drill penetration and / or penetration tendency when drilling a bone via a surgical drill (2), comprising at least one, in particular a single-axis, piezoelectric acceleration sensor (22) for generating a piezoelectric acceleration signal and an evaluation unit arranged to identify signals or signal variations characteristic of penetration and / or penetration tendency.

Description

[Technical field]

[0001] The present disclosure relates to an operating condition detection device for a surgical drill that detects penetration and / or penetration tendency as an operating condition when drilling into bone, and a method for detecting an operating state / condition, in particular penetration or penetration tendency of a surgical drill (i.e. the imminent penetration of the drill on the side of the bone facing away from the drill). [Background technology]

[0002] Surgical drills play an important role in everyday surgical procedures. They are used, for example, in orthopedic surgery to pre-drill screw holes for fixing metal plates in injured or fractured bones. Furthermore, in neurosurgery, they are also used for trepanation (burr hole) and craniotomy (craniotomy) with safety trepans and craniotomes. A common operation (OP) is the relaxation (decompression) of the spinal cord. Bone growths are removed with the help of drills and pressure on the nerves is relieved. Furthermore, surgical drills are used in tumor surgery in neurospinosurgery and oral and maxillofacial surgery to cut tumor tissue in healthy bone. A further important application is the drilling of channels and bearings for the insertion of cochlear implants next to the auditory canal.

[0003] Such an operation is risky for the patient, since behind the bone that is drilled there is vulnerable healthy tissue, and possibly nerve pathways and blood vessels. Such damage can cause great damage to the patient. The necessary / exact depth of the drill cannot be calculated, but can only be estimated. Studies have shown that drilling into the cortex is, on average, 6.33 mm too deep. Since damage to the spinal cord and especially the brain is irreversible, the safety of the OP, and therefore the safety of the patient, crucially depends on the experience of the surgeon. Summary of the Invention

[0004] Accordingly, it is an object of the present disclosure to provide a device that increases patient safety by preventing excessively deep drilling when drilling into bone and protecting tissue behind the bone from being penetrated or damaged by the drill.

[0005] This object is achieved by a surgical drill as set forth in any one of claims 1 to 11. Moreover, this object is also achieved by a surgical situation detection device as set forth in claim 12.

[0006] It is a further object of the present disclosure to provide a method for reliably detecting a predetermined operating state of a surgical drill.

[0007] This object is solved by a method according to claim 13 or claim 14 and by claim 15.

[0008] In particular, the object of preventing too deep drilling into bone is solved by a detection device configured to detect penetration and / or penetration tendency of a drill during drilling into bone via a surgical drill, the detection device comprising at least one, in particular uniaxial, piezoelectric acceleration sensor for generating a piezoelectric acceleration signal and an evaluation unit configured to identify signals or signal variations characteristic of penetration and / or penetration tendency in the signal received from the piezoelectric acceleration sensor.

[0009] In other words, a detection device is provided in or on the surgical drill, preferably fixed in a force-fit manner. The detection device comprises a piezoelectric acceleration sensor. The piezoelectric acceleration sensor comprises a piezoelectric transducer (e.g. quartz) or ceramic, which is connected at one side to an inertial seismic mass. When the (body) vibrations of the drill act on the piezoelectric acceleration sensor, the mass presses against the piezoelectric transducer. The piezoelectric transducer reacts to the forces caused by the seismic mass, and charges move in the piezoelectric transducer. In this way, the vibrations are converted into a piezoelectric acceleration signal. The piezoelectric acceleration signal thus generated is evaluated by an evaluation unit. The evaluation unit detects penetration or a tendency to penetration based on the characteristic signal and / or on the fluctuations of the characteristic signal.

[0010] The characteristic signal may be, for example, a maximum amplitude, an instantaneous amplitude, an effective value, a vibration velocity, a vibration path, or the like.

[0011] In other words, the detection device comprises an evaluation unit and is configured to detect a material transition of the drill or a tendency of a material transition of the drill, in particular of the drill tip.

[0012] The essence of the present disclosure therefore consists in identifying, by an evaluation unit of the detection device, signals or signal variations characteristic of penetration and / or penetration tendency, which are generated by a piezoelectric acceleration sensor of the detection device due to vibrations of the surgical drill.

[0013] Such a detection device makes it possible to detect drilling too deep into the bone, in particular penetration and / or a tendency to penetrate, and thus to prevent damage to the tissue located behind the bone. This provides the surgeon with peace of mind and at the same time increases patient safety, as errors and / or incorrect assessments are reduced. Furthermore, the operation can be performed more efficiently, since it is not necessary to analyze X-ray images regarding the drilling depth.

[0014] In a first aspect, the detection device may be configured to output a drilling stop signal when the evaluation unit identifies a signal or a signal variation characteristic of a penetration and / or a penetration tendency.

[0015] In other words, the detection device may cut off the power supply and / or the (drive) air supply to the drive of the surgical drill. Alternatively or additionally, the drive shaft of the surgical drill may be locked by a brake.

[0016] Further, alternatively or additionally, the detection device may be configured to output an alarm signal when the evaluation unit identifies a signal or a signal variation that is characteristic of a penetration and / or a penetration tendency. The alarm signal may be an optical signal, e.g. in the form of a display on a display, a warning light being turned on, an acoustic signal, e.g. in the form of a warning sound, and / or a tactile signal, e.g. in the form of a vibration.

[0017] In a further aspect, a piezoelectric acceleration sensor may detect structure-borne sound signals which are transmitted as analog (acceleration) signals to the evaluation unit.

[0018] In other words, the piezoelectric acceleration sensor may detect vibrations or waves propagating at a finite speed in a solid body, in particular a drill or a drill component and / or a drill bit, and convert them into an analog signal which is then preferably transmitted by a cable or the like to an evaluation unit.

[0019] Structure-borne sound detection can reduce the influence on the detection results of environmental influences, especially noise in the operating room, in the form of machine beeps, voices, noise caused by movements, excessive noise, etc. In this way, the surgeon can be assisted in particularly hectic operating room situations, overdeep drilling can be effectively prevented and the physician can have confidence in the detection results regardless of the environment in which he is located.

[0020] In a further embodiment, the evaluation unit of the detection device may comprise at least one amplifier, in particular an amplifier with a gain greater than / equal to 10, and / or a filter, in particular an n-th order Butterworth filter or a bandpass filter.

[0021] In other words, the analog signal output by the piezoelectric acceleration sensor may be amplified by an amplifier in the evaluation unit and / or filtered by a filter in the evaluation unit, i.e. the amplifier may convert the analog signal into a proportional voltage and amplify it. Subsequent high-pass or low-pass filtering may suppress / filter interference signal components.

[0022] In a further embodiment, the evaluation unit may comprise an analog-to-digital converter sampling the analog signal at a sampling rate of at least 20000 Hz, in particular at a sampling rate of about 25600 Hz.

[0023] In other words, one may "sample" the analog signal. In yet other words, one may read out the amplitude of the waveform of the analog signal at different (defined) points in time. These discrete amplitude values ​​are stored in an array or vector. According to the Nyquist-Shannon theorem, the sampling rate / sampling frequency of an analog signal must be at least twice the highest frequency of the signal to be digitized. Otherwise aliasing effects may occur. The evaluation of the signal can be improved by digitization.

[0024] In other words, the analog-to-digital converter and the piezoelectric acceleration sensor may be matched to each other, especially in terms of sampling rate.

[0025] In a further embodiment, the evaluation unit, preferably comprising a microcontroller, may convert the signal into a spectrogram, preferably a mel-spectrogram, via a Fourier transform, in particular a short-time Fourier transform.

[0026] In other words, the evaluation unit, preferably with a microcontroller provided for this purpose, may decompose a signal, preferably present in the form of a continuous-time signal, into its discrete frequency components. Here, signals whose frequency characteristics change over time may also be processed with a short-time Fourier transform. First, a window and its parameters, window size and step size, are defined. The window is shifted over the signal. A Fourier transform is calculated for each window and a spectrum is formed. The spectra are joined to generate a spectrogram. A spectrogram is a superposition of a series of spectra in a predetermined time interval. The distribution of frequencies over time and their intensity can thus be read from the spectrogram. Such a transformation of the signal makes it possible to optically evaluate the signal.

[0027] In a further aspect, the evaluation unit can evaluate the spectrogram with an AI trained network, preferably via the “bounding box principle”, and identify drill penetration or drill penetration tendency when the spectrogram lights up.

[0028] In other words, the evaluation unit may evaluate the spectrogram using a network trained via machine learning. Here, preferably, the spectrogram is evaluated optically. In particular, the "bounding box principle" is used. In the "bounding box principle", at least one rectangle, preferably several rectangles, each with a label, are placed on the spectrogram. Each rectangle is characterized by the coordinates of its upper left corner and its lower right corner. The labels are divided into three classes. Here, the first class is "wait", the second class is "drilling" and the third class is "penetration". In the background, class zero may further be used. In the class "wait", only the fundamental frequency can be seen. In the class "drilling", the frequency division can be seen. In the class "penetration", the spectrogram is again almost back to waiting and still contains harmonics, since the drill has not yet completely left the bone. The rectangle labeled "penetration" starts before the actual penetration of the drill. As an indicator of penetration and / or penetration tendency, lighting in particular at least one harmonic relative to the fundamental frequency is used. Lighting in the spectrogram is caused by an increased occurrence of high amplitudes in the signal. The network can be trained on the basis of training data. Furthermore, the network can be continuously trained further, so that the use of the surgical drill with the detection device according to the invention can improve the discrimination.

[0029] Optical analysis according to the "bounding box principle" is a particularly robust analysis method: reliable identification of penetrations and / or penetration tendencies can be achieved by a network that is trained on training data and continues to learn during use.

[0030] The spectrogram represents the superposition of multiple Fourier transforms and provides precise information on the magnitude of the frequencies and amplitudes during penetration. These features are learned by the AI ​​and also include the value of the spread of the frequency spectrum during penetration. The result is the fundamental frequency, its harmonics and the ± values ​​of the rotation speed. Since the fundamental frequency of the drill and the respective rotation speed are known, the changes in the Fourier transform can be detected as long as they are in the area defined by the AI.

[0031] In further embodiments, the AI ​​trained network may be a cloud network fed with data from compatible sensing devices.

[0032] In other words, the network trained by the AI ​​may be a network used by multiple compatible detection devices for the detection / identification of penetration and / or penetration tendencies.

[0033] In this way, the computationally intensive analysis of the spectrograms can be performed centrally by a corresponding computer, reducing investment costs at the site, e.g., in a hospital. Furthermore, the identification of penetrations and / or penetration trends is significantly improved by bundling multiple data from multiple compatible detection devices. Furthermore, cloud networks offer the advantage that redundancy can be implemented in an economically meaningful way, which increases failure safety.

[0034] In a further embodiment, the piezoelectric acceleration sensor can have the highest sensitivity in the measurement frequency range up to 10000 Hz.

[0035] Furthermore, the object of preventing too deep drilling into bone is solved by a surgical drill, which is formed with a detection device according to one aspect described above.

[0036] In other words, a detection device configured to detect penetration and / or penetration tendency of a drill during drilling into a bone via a surgical drill comprises at least one, in particular a single-axis, piezoelectric acceleration sensor for generating a piezoelectric acceleration signal and an evaluation unit provided for identifying signals or signal variations characteristic of penetration and / or penetration tendency of the surgical drill.

[0037] In one embodiment, the piezoelectric acceleration sensor can be connected to the drive of the surgical drill by a direct metal connection.

[0038] In other words, the piezoelectric acceleration sensor may be mechanically connected to the drive of the surgical drill, either directly or via at least one preferably metallic holder, which may for example be an electric motor or a turbine. In addition to being made of metal, the holder may also be made of other materials that conduct structure-borne sound with as few losses as possible and without attenuation, while still meeting the material requirements of the surgical drill.

[0039] The direct connection ensures that the piezoelectric acceleration sensor can already detect low accelerations and / or changes in acceleration, and also ensures that the influence of vibrations introduced from outside the surgical drill, e.g. due to noise, is filtered out.

[0040] In a further embodiment, the amplifiers and / or filters of the evaluation unit may be formed in / on the surgical drill, preferably on a hermetically sealed circuit board, in the immediate vicinity of the piezoelectric acceleration sensor.

[0041] In other words, a preferably hermetically sealed printed circuit board with components may be formed in / on the surgical drill. Said components may form a filter and / or an amplifier. The printed circuit board may be formed in the immediate vicinity of the piezoelectric acceleration sensor. The gain / power of the amplifier may preferably be at least 10. The filter may limit the signal to a desired signal range and may be independent of interfering signals caused, for example, by the drill and its rotation itself.

[0042] Losses in the transmission from the piezoelectric acceleration sensor to the amplifier and / or filter are minimized due to the close proximity between the piezoelectric acceleration sensor and the amplifier and / or filter, thus improving the detection result of the detection device.

[0043] In a further aspect, the evaluation unit may be fully or partly formed within a base unit (control unit) provided external to the surgical drill.

[0044] In other words, the filters and / or amplifiers may be formed in / on the surgical drill / handpiece of the surgical drill. Further components, such as, for example, an analog-to-digital converter and one or more microcontrollers, may be formed outside the handpiece of the surgical drill in the base unit. The handpiece and the base unit may be connected to each other by cables and / or hoses. The base unit may supply electrical energy to the components formed in the handpiece via the cables.

[0045] In this way, the handpiece of the surgical drill can be constructed particularly compactly and is easy to handle for the operator. In particular, if a filter and / or an amplifier is formed in / on the handpiece, it is ensured that the raw signal of the piezoelectric acceleration sensor is transmitted to the base unit in an appropriate form (amplified, filtered). In the base unit, the signal is sampled by an analog-to-digital converter and then converted by a microcontroller, preferably via a short-time Fourier transform, into a spectrogram. These spectrograms are then transmitted to the AI ​​trained network, preferably by a communication unit formed in the base unit. The AI ​​trained network may be cloud-based. The spectrogram may also be evaluated in the base unit, preferably via a microcontroller further equipped with memory and data processing capabilities. In this way, a modular design can also be implemented, so that several handpieces can be formed with the base unit. This reduces costs and expands the range of treatments by different handpieces adapted to the treatment.

[0046] In a further aspect, a plurality of piezoelectric acceleration sensors may be formed in a handpiece of a surgical drill.

[0047] In a further aspect, a plurality of piezoelectric acceleration sensors may be formed in a handpiece of a surgical drill and spaced apart from one another.

[0048] Furthermore, the object of providing a method for reliably detecting a predetermined operating state of a surgical drill is to provide a method for detecting a predetermined operating state of a surgical (bone) drill, comprising the steps of: - generating a piezoelectric acceleration signal using a piezoelectric acceleration sensor, in particular a single axis, arranged on or in the bone drill; transmitting an acceleration signal to an evaluation unit; identifying a signal or signal variation characteristic of a predetermined operating condition; The problem is solved by a method comprising:

[0049] The above-described method allows early detection of malfunctions of a surgical drill. Thus, for example, malfunctions of the drill bearings of a surgical drill can be detected. Furthermore, non-circular running of the drill, for example due to improper clamping of the drill in a surgical drill, can be detected. If a malfunction is detected, the surgical drill can be stopped and / or an acoustic or tactile alarm signal can be output. The predetermined operating state can be, for example, a normal state and a fault state or a material transition.

[0050] In one embodiment, the acceleration signal is evaluated in an evaluation unit, filtering the acceleration signal with a filter, preferably an nth order Butterworth filter or a bandpass filter; amplifying the acceleration signal with an amplifier, preferably an amplifier having a gain of 10 or greater; Sampling the acceleration signal with an analog to digital converter; - converting the (digital) acceleration signal into a spectrogram, preferably by means of a Fourier transform, particularly preferably a short-time Fourier transform; evaluating the spectrogram via the "bounding box principle" with an AI trained network; Includes.

[0051] Additionally, the present disclosure relates to the use of piezoelectric acceleration sensors, particularly single axis, located or to be located on or within the bone drill to detect predetermined operating conditions.

[0052] It is here again pointed out that the identification of such signals or signal variations characteristic of a predefined operating state can also be performed, for example, by comparing the signal or variation pattern with specific patterns pre-stored for matching in the simplest embodiment. Alternatively or additionally, it is also conceivable to assign absolute or relative signal values / amplitudes / duty cycles / signal durations etc. to specific operating states. Finally, it is in principle possible to provide detection algorithms comparable to image detection algorithms. [Brief description of the drawings]

[0053] [Figure 1] FIG. 2 is an explanatory diagram of a surgical drill including a first handpiece and a base unit. [Diagram 2] FIG. 4 is an explanatory diagram of a second handpiece. [Diagram 3] FIG. 4 is an enlarged view of the second handpiece. [Figure 4] FIG. 13 is an explanatory diagram of a sleeve for the first handpiece. [Diagram 5] FIG. 2 is an explanatory diagram of a piezoelectric acceleration sensor. [Figure 6] FIG. 13 shows a graph and spectrogram of detected vibration. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0054] Embodiments of the present disclosure are described below with reference to the associated figures.

[0055] 1 shows a surgical drill 2 comprising a first configuration example first handpiece 4 and a base unit 6. The first handpiece 4 is connected to the base unit 6 by a cable 8.

[0056] FIG. 2 shows the second handpiece 10 of the second configuration example. FIG. 3 is an enlarged view of the second handpiece 10. The second handpiece will be described in more detail below with reference to FIGS. 2 and 3. The second handpiece 10 is connected to the base unit 6 by a cable 8. The second handpiece 10 has a handle portion 12 in the form of a pistol grip. The handle portion 12 is provided and configured so that it can be gripped by a practitioner, such as a doctor, so as to reliably guide the second handpiece 10. An operation knob 14 is formed on the front side of the handle portion 12. The operation knob 14 can be used, for example, to operate the rotation direction and rotation speed of a surgical drill. An electric drive portion 16 is formed on the end of the handle portion 12 away from the cable 8. In other words, the electric drive portion 16 is formed above the handle portion 12 of the second handpiece 10. A bar chuck 18 configured to receive a drill 20 is formed on the front side of the electric drive portion 16. Here, the bar chuck 18 protrudes beyond the handle portion 12 and the operation knob 14. The drive unit 16 and the bar chuck 18 are preferably connected directly or via a gear mechanism (not shown). The piezoelectric acceleration sensor 22 is formed by being directly connected to the drive unit 16. In other words, the piezoelectric acceleration sensor 22 is directly metal-connected to the drive unit 16. In other words, the piezoelectric acceleration sensor 22 is in direct contact with the drive unit 16. Alternatively, it is also conceivable that the piezoelectric acceleration sensor 22 is connected to the drive unit 16 via a metal connection member. In this way, the structure-borne sound of the surgical drill, particularly the structure-borne sound of the drive unit 16, can be directly received by the piezoelectric acceleration sensor 22. The piezoelectric acceleration sensor 22 transfers the detected acceleration signal to a filter and an amplifier. The filter and the amplifier are formed on a hermetically sealed printed circuit board 24. Here, the piezoelectric acceleration sensor 22 and the hermetically sealed printed circuit board 24 are accommodated in close proximity to each other in the handpiece 10 of the surgical drill 2. The piezoelectric acceleration sensor 22 and the hermetically sealed printed circuit board 24 are connected to a sensor cable 26. In the embodiment shown, a piezoelectric acceleration sensor 22 and hermetically sealed printed circuit board 24 are received entirely within the handle portion 12 of the handpiece 10 .

[0057] FIG. 4 shows an alternative embodiment of the first configuration, in particular for the first handpiece 4. In particular, FIG. 4 shows a retaining sleeve 28, which is mounted on the first handpiece 4 and is configured to receive and optionally guide the drill 20 in the longitudinal direction of the retaining sleeve 28, in particular in the central direction. The retaining sleeve 28 comprises a grooved retaining part 30 and a fastening part 32. The fastening part 32 is configured to receive a piezoelectric acceleration sensor 22 (not shown in this configuration). The piezoelectric acceleration sensor 22 may be fixed on the fastening part 32 of the retaining sleeve 28, in particular via a fastening compound, adhesive, a latch mechanism or similar fastening means. That is, the piezoelectric acceleration sensor 22 may be fixed on the retaining sleeve 28 in a materially adhesive or force-fit manner. The main focus here is on a direct connection between the piezoelectric acceleration sensor 22 and the retaining sleeve 28. That is, the piezoelectric acceleration sensor 22 is fixed on the retaining sleeve 28 in such a way that the vibrations of the retaining sleeve 28 are transmitted to the piezoelectric acceleration sensor 22 as directly as possible without attenuation. In contrast to the second embodiment of Figures 2 and 3, the piezoelectric acceleration sensor 22 of the first embodiment is arranged outside the first handpiece 4. In this embodiment, the piezoelectric acceleration sensor 22 and the hermetically sealed printed circuit board 24 are preferably arranged in a sealed identification element (not shown). The embodiment shown in Figure 4 is also conceivable as a retrofit solution, for example.

[0058] FIG. 5 shows a schematic representation of the piezoelectric accelerometer 22. The piezoelectric accelerometer 22 comprises a seismic mass 34 and a piezoelectric transducer / piezoceramic 36. The piezoelectric accelerometer 22 is configured to convert a mechanical vibration acceleration a acting on the piezoelectric accelerometer 22 into a voltage measurement signal U. In other words, vibrations act on the piezoelectric accelerometer 22, pressing the seismic mass 34 against the piezoelectric transducer 36. Piezoelectric elements such as the piezoelectric transducer 36 react to the acting force with a charge transfer q in the transducer. In this way, the vibration acceleration acting on the piezoelectric accelerometer 22 is converted into a charge. This charge is proportional to the acting force. The seismic mass 34 and the piezoelectric transducer 36 form a spring-mass system with low-pass behavior and a linear frequency range.

[0059] The mode of operation of the detection device for the surgical drill 2 is described below. When drilling into a bone, vibrations are generated in the surgical drill 2 due to the interaction of the drill 20 with the bone. These vibrations occur in particular in the drive 16 of the surgical drill 2. The piezoelectric acceleration sensor 22 detects the vibrations of the drive 16 and transmits an analog voltage signal on the sensor cable 26 to a filter and amplifier located on a hermetically sealed printed circuit board 24. The analog voltage signal is filtered by the filter in order to filter out interfering signal parts. This is done, for example, by high-pass or low-pass filtering. In the amplifier, the signal is amplified, preferably by at least a factor of 10. The signal thus preprocessed (38 in FIG. 6) is transmitted to the base unit 6 via the cable 8 connecting the handpiece 10 to the base unit 6.

[0060] The evaluation of the preprocessed signal 38 is explained below with reference to FIG.

[0061] The signal 38, preprocessed by the filters and amplifiers of the hermetically sealed printed circuit board 24, is present in WAV format. The signal 38 is a mapping of the acceleration detected by the piezoelectric acceleration sensor over time t (here in seconds). This signal is sampled in the base unit 6 by an analog-to-digital converter formed in the base unit 6 and is subsequently decomposed into frequency components by the microcontroller. The decomposition into frequency components is performed by performing a short-time Fourier transform over a defined time window. The final signal obtained in the form of a spectrogram 40 is represented on the horizontal axis as time t (here in seconds), on the vertical axis as frequency f (here in Hertz) and as amplitude as intensity. In the present disclosure, the spectrogram 40 is shown inverted for better illustration. In other words, areas that are dark in the real spectrogram are shown here as light and areas that are bright in the real spectrogram are shown here as dark in the spectrogram 40 shown. The spectrogram 40 is then evaluated via a microcontroller included in the base unit 6 or via a network trained via AI. The evaluation is performed according to the "bounding box principle", i.e. the spectrogram 40 is evaluated graphically. The AI ​​trained network places rectangles with different labels on the spectrogram 40. The AI ​​trained network places the rectangles for example based on the brightness of the spectrogram 40. The labels used are "wait L", "pierce B", "penetrate D". In this case the rectangles may overlap.

[0062] The detection of penetration and / or penetration tendency is performed by the AI ​​trained network based on the illumination A of the spectrogram 40. As already mentioned above, the illumination A is shown in this disclosure in inverted colors for better illustration. Also, the fundamental frequency GF of the handpiece 10 is clearly visible in the spectrogram 40. The illumination A in the spectrogram 40 represents an increased occurrence of high amplitudes in the signal. The AI ​​trained network is trained to detect such optical changes in the spectrogram 40. When a penetration is detected, the AI ​​trained network can output, for example, a motor stop command or an alarm. The alarm can be an acoustic alarm in the form of a warning sound, an optical alarm in the form of a flash of light, etc., or a tactile alarm in the form of a vibration of the handpiece 4, 10.

[0063] In summary, the piezoelectric acceleration sensor 22 detects vibration accelerations of the drive 16 of the surgical drill 2. These vibration accelerations are pre-processed by filters and amplifiers in or on the handpiece 4,10 and then sampled by an analog-to-digital converter in the base unit 6. The signal is then converted by the microcontroller into a spectrogram 40. The spectrogram 40 is finally analyzed / evaluated by an AI trained network to detect penetration or penetration tendency based on optical changes in the spectrogram 40. [Explanation of symbols]

[0064] 2 Surgical drills 4. First Handpiece 6 Base Unit 8 Cable 10 Second Handpiece 12 Handle section 14 Operation knob 16 Drive unit 18. Virchuck 20 Drill 22 Piezoelectric acceleration sensor 24 Printed Circuit Board 26 Sensor cable 28 Retaining sleeve 30 Holding part 32 Fastening part 34 Earthquake-resistant block 36 Piezoelectric vibrator 38 (preprocessed) signals 40 Spectrogram L Idol B. Learning D Penetration A. Lighting GF Reference Frequency t time f frequency

Claims

1. A surgical drill having a detection device formed thereon, at least one piezoelectric acceleration sensor configured to detect breakthrough and / or breakthrough tendency of the drill when drilling a bone via the surgical drill, the piezoelectric acceleration sensor generating a piezoelectric acceleration signal; an evaluation unit arranged to identify signals or signal variations characteristic of said breakthroughs and / or breakthrough trends; Equipped with The piezoelectric acceleration sensor detects structure-borne sound signals which are transmitted as analog signals to the evaluation unit.

2. The surgical drill according to claim 1 , wherein the evaluation unit comprises at least one amplifier and / or filter.

3. 3. The surgical drill of claim 2, wherein the evaluation unit comprises an analog-to-digital converter that samples the analog signal at a sampling rate of at least 20,000 Hz.

4. 4. The surgical drill according to claim 3, wherein the evaluation unit converts the signal into a spectrogram via a Fourier transform.

5. 5. The surgical drill of claim 4, wherein the evaluation unit evaluates the spectrogram with an AI-trained network and identifies a breakthrough of the drill or a breakthrough tendency of the drill when the spectrogram illuminates.

6. 6. The surgical drill of claim 5, wherein the AI-trained network is a cloud network fed with data from compatible detection devices.

7. 2. The surgical drill of claim 1, wherein the piezoelectric acceleration sensor has the highest sensitivity in a measurement frequency range up to 10,000 Hz.

8. 2. The surgical drill of claim 1, wherein the piezoelectric acceleration sensor is connected to a drive portion of the surgical drill by a direct metal connection.

9. 2. The surgical drill according to claim 1, wherein the amplifier and / or filter of the evaluation unit are formed on a circuit board, the circuit board being formed in / on the surgical drill in the vicinity of the piezoelectric acceleration sensor.

10. 2. The surgical drill of claim 1, wherein the evaluation unit is formed completely or partly in a base unit provided external to the surgical drill.

11. 1. A detection device configured to detect breakthrough and / or breakthrough tendency of a drill when drilling a bone via a surgical drill, comprising: at least one piezoelectric acceleration sensor for generating a piezoelectric acceleration signal; an evaluation unit arranged to identify signals or signal variations characteristic of said breakthroughs and / or breakthrough trends; Equipped with The piezoelectric acceleration sensor detects structure-borne sound signals which are transmitted as analog signals to the evaluation unit.

Images