Method, auscultation device, and system for auscultation signal processing

Abstract

The present invention relates to the processing of an auscultation signal that preferably has a periodicity, wherein the auscultation signal is generated by an auscultation device comprising an acoustic pickup for auscultating sounds and a signal converter that is physically and acoustically coupled to the acoustic pickup, for converting the auscultated sounds into the auscultation signal, wherein the auscultation signal is divided into signal intervals, whereby auscultation signal portions are formed, and wherein the auscultation signal portions are overlaid and a result signal is formed by processing and by examining the density of the overlaid auscultation signal portions.

Description

[0001] Method, auscultation device and system for auscultation signal processing

[0002] The present invention relates to a method for processing an auscultation signal preferably having a periodicity, an auscultation device, a system for auscultation signal processing as well as a computer program product and computer-readable storage medium.

[0003] The present invention preferably relates to the auscultation of sounds and the conversion of the auscultated sounds into an auscultation signal using a signal converter. For this purpose, the auscultated sound, which initially exists acoustically, i.e., as a sound wave, can be converted into an electrical and, more preferably, a digital auscultation signal.

[0004] In particular, the sounds are those of internal organs and / or auscultation is performed using a sound pickup, in particular a chestpiece, or an auscultation device, which is preferably a stethoscope. The invention is particularly advantageous when used with a stethoscope, but is also fundamentally applicable and advantageous in other areas.

[0005] The auscultated or auscultable sound, particularly from internal organs, preferably exhibits a periodicity. The auscultated or auscultable sound can originate from organs that perform a periodic movement, for example, the heart or lungs. In particular, the sound or auscultation signal thus comprises at least essentially periodic patterns, such as those generated by breathing or lung movement or the heartbeat of a living being. In other words, the auscultated sound is, for example, a heart murmur, and the auscultation signal is a phonocardiogram, hereinafter abbreviated to PKG or PCG.

[0006] In principle, however, it is also possible for the auscultated or auscultable sound, or the sound converted into an electrical, preferably digital, auscultation signal, to originate from another source. This source is preferably a living being, but can also be a machine or another object, preferably one that moves periodically. The auscultated sound or auscultation signal preferably has a useful component that can and should be used as the basis for further processing. This useful component preferably always exhibits or is formed by periodicity, as can originate from organ movements, but also from other sources.

[0007] Furthermore, the present invention preferably relates to the examination of the auscultation signal. In particular, the auscultation signal is examined for anomalies, i.e., with the aim of identifying anomalies in the signal pattern. Such anomalies, also called features, can be an indicator of a malfunction such as a disease or pathological changes in the auscultated organ. If a (signal) anomaly is identified, a warning can be issued, further examinations initiated, or a diagnosis made.

[0008] Typically, a physician, especially a cardiologist, listens directly to the auscultated sound to identify such abnormalities. However, as is preferred here, the auscultated sound can be evaluated (automatically) for such abnormalities after it has been converted into an auscultation signal.

[0009] It has been shown that automatic detection of anomalies in the auscultation signal is regularly subject to considerable uncertainties and that the significance of a result that can be achieved through automatic evaluation of the auscultation signal is therefore limited.

[0010] Against this background, the object of the present invention is to provide a method, an auscultation device, a system, a computer program product and a computer-readable storage medium, which enables an improved evaluation of the auscultation signal - in particular with regard to the accuracy of the identification of anomalies.

[0011] This object is achieved by a method according to claim 1, by an auscultation device according to claim 10, a system according to claim 12, a computer program product according to claim 14, or a computer-readable storage medium according to claim 15. Advantageous further developments are the subject of the dependent claims. According to the proposal, the auscultation signal is divided into intervals—preferably based on periodicity—to form auscultation signal segments. Furthermore, the auscultation signal segments are superimposed, and a result signal is formed by processing, in particular density analysis, the superimposed auscultation signal segments.

[0012] A density examination in the sense of the present invention is to be understood as a processing or evaluation of the multiple signal values ​​present after superimposing the auscultation signal sections at the respective time, in which the density or the position and / or the distances of the signal values ​​are preferably taken into account.

[0013] The processing or density analysis can, for example, involve the application of preferably nonlinear filters such as ranking filters, adaptive filters, and / or morphological filters, and / or (graphic) high-pass filters, sharpening filters, edge detection filters, or edge localization filters based on image processing methods. Therefore, the processing or density analysis is preferably not a pure arithmetic averaging, but instead preferably, but preferably not necessarily, considers the position and / or distances of the signal values ​​from one another.

[0014] In the context of the present invention, overlaying the auscultation segments preferably means overlaying the signal values ​​(at the respective time point) of the respective intervals. The result of the overlay is therefore that multiple signal values ​​are present at the respective time point.

[0015] In the sense of the present invention, addition of the auscultation sections preferably means the addition of the signal values ​​(at the respective time point) of the respective intervals.

[0016] For addition, the auscultation signal sections are particularly preferably formed in such a way, or the auscultation signals of the auscultation signal sections are added in such a way that features of the auscultation signal of the respective auscultation signal section corresponding to the periodicity are added, in particular by temporal synchronization. Alternatively or additionally, for superimposition, the auscultation signal sections are particularly preferably formed in such a way, or the auscultation signals of the auscultation signal sections are superimposed in such a way that features of the auscultation signal of the respective auscultation signal section corresponding to the periodicity are superimposed, in particular by temporal synchronization.

[0017] In particular, the beginnings of auscultation signal segments are determined based on the periodicity. Alternatively or additionally, the beginnings of auscultation signal segments can be determined, in particular offset, from one another in such a way that features corresponding to the periodicity are or become temporally synchronized with one another.

[0018] It has surprisingly been shown that the periodicity of the auscultation signal is advantageously suitable for improving signal quality. In this context, the proposed superposition makes it possible to determine a more precise course of the auscultation signal, as random components and / or noise components and components of non-periodic interference signals in the auscultation signal are less significant due to the superposition.

[0019] A further, also independently implementable aspect of the present invention relates to an auscultation device. The auscultation device comprises a sound pickup, which can be formed by a chestpiece for auscultating sounds from internal organs. Furthermore, the auscultation device comprises a signal converter physically and acoustically coupled to the sound pickup for converting the auscultated sounds into the auscultation signal, wherein the auscultation device is configured to implement the proposed method. This allows corresponding advantages to be achieved, as explained in connection with the proposed method.

[0020] A further aspect of the present invention, which can also be implemented independently, relates to a system for an auscultation device, in particular according to the proposal. The system has the signal converter for converting the auscultated sounds into the auscultation signal, wherein the signal converter has a connection or connector for physical and acoustic coupling to a sound sensor of the system, such that the auscultated sound recorded by the sound sensor can be converted into the auscultation signal by the signal converter. Furthermore, the system has an analysis device which is designed to divide the auscultation signal into signal intervals, preferably based on a periodicity of the auscultation signal, thereby forming auscultation signal sections. Finally, the analysis device is preferably designed to superimpose the auscultation signal sections and to generate a result signal by processing orDensity study of the superimposed auscultation sections.

[0021] The system can also be used advantageously to upgrade or retrofit existing auscultation devices and provide them with the functionality needed to perform the proposed method. For this purpose, the signal converter can be interposed in the tubing of an auscultation device.

[0022] A further aspect of the present invention, which can also be implemented independently, relates to a computer program product comprising instructions which, when executed on a processor of the analysis device of a proposed system, cause the steps of the proposed method to be carried out.

[0023] A further aspect of the present invention, which can also be implemented independently, relates to a computer-readable storage medium on which the proposed computer program product is stored.

[0024] An auscultation signal within the meaning of the present invention is an electronic, preferably digital, signal that corresponds to, is derived from, or is derived from the auscultated sound. It can be or comprise a signal converted into electrical oscillations using a microphone and subsequently digitized using an A / D converter.

[0025] A periodicity within the meaning of the present invention is preferably a (signal) structure that recurs at certain time intervals. The structure preferably involves certain sound events or corresponding characteristic features of the signal curve. The periodicity preferably results from sound events that recur cyclically and / or at at least substantially constant time intervals in the auscultated sound or auscultation signal.

[0026] An auscultation device, in particular a stethoscope, within the meaning of the present invention is preferably a device for auscultating noises, i.e. vibrations that can be caused, for example, by an organ.

[0027] Auscultation in the sense of the present invention means listening (to an object or the body) using a suitable listening device such as a stethoscope, its chest piece or another sound pickup.

[0028] A sound pickup, in particular a chest piece, within the meaning of the present invention is preferably a device suitable for auscultation, in particular by being placed on the chest of a living being. Irrespective of this, the sound pickup can be brought into contact with other surfaces and be suitable for auscultation there as well.

[0029] The sound sensor is preferably designed to record or convert structure-borne sound into sound propagating in air. For this purpose, the sound sensor can have a contact surface, in particular a membrane, for contact with a body for the purpose of acoustic coupling with it, and a cavity filled with air, wherein vibrations of the body can excite sound propagating in the air via the contact surface (cause air to vibrate), which corresponds to the vibrations of the body. The chest piece is a sound sensor that is designed to record or convert structure-borne sound from a chest of a living being, in particular through its shape and / or structure and / or biocompatibility. Alternatively or additionally, the sound sensor can also transmit the structure-borne sound directly or via a solid-state coupling to the signal transducer.

[0030] A noise within the meaning of the present invention is preferably an acoustic vibration, i.e., in particular, sound, for example, structure-borne noise. The noise can be caused by an organ or its or other movements. The noise therefore preferably corresponds to movements of the organ.

[0031] The noise preferably exhibits a periodicity, particularly preferably in the form of a periodic useful component. The useful component is preferably the component to be analyzed or evaluated. Disturbances or other non-useful components can also, in principle, exhibit a periodicity, but preferably at a different frequency than the periodicity of the useful component. In the following, the terms "periodicity of the useful component" and "periodicity" are used synonymously and interchangeably unless otherwise stated and regardless of any possible periodic components of the noise or disturbance variables.

[0032] A membrane within the meaning of the present invention is preferably a vibratable, thin, flat device that can be set into vibration on one side when stimulated by the sound and can cause air on the opposite side to vibrate accordingly, which is then directed to the earpieces or the signal transducer. This can be done through the tube, which serves as a conduit for acoustic sound waves.

[0033] An organ in the sense of the present invention is preferably a moving body part arranged inside the body of a living being, such as a heart, a lung and the like.

[0034] A signal converter within the meaning of the present invention is preferably a device that is designed to convert acoustic signals into electrical signals. It can be a microphone or the signal converter comprises a microphone. The signal converter can convert the sound into an electrical and preferably digital auscultation signal.

[0035] An interval or signal interval in the sense of the present invention is preferably a time period which is preferably oriented towards or determined by one or more of the periods of the periodicity of the auscultation signal.

[0036] An auscultation signal segment within the meaning of the present invention is an auscultation signal that is limited or clipped to one or more signal intervals. The auscultation signal can therefore be divided into several auscultation signal segments, preferably in such a way that the intersection boundaries are or will be determined based on the periodicity.

[0037] A result signal is a signal determined based on the auscultation signal, which still represents the sounds preferentially originating from the organ, particularly emphasized over interference or with the useful component emphasized over interference.

[0038] An individual value in the sense of the present invention is preferably a signal value that is or has been determined by combining several signal values, for example by superimposing and analyzing or summing signal values ​​of a respective point in time of the respective auscultation signal sections.

[0039] A weighted signal value in the sense of the present invention is preferably a signal value that is or is given greater consideration in determining the individual value compared to other signal values.

[0040] An ambient noise signal within the meaning of the present invention is preferably a signal that represents an ambient noise, in particular an interference noise that acts from outside on the auscultation device, the sound sensor or the signal converter.

[0041] A filter within the meaning of the present invention is preferably a device for frequency-selective attenuation or amplification of a signal based on its filter characteristic, which can be defined as a transfer function or frequency response according to magnitude and phase. It is therefore, in particular, an electrical signal filter. In connection with signal values, a filter is preferably a method for analyzing the signal values ​​for the purpose of determining a result value from the signal values ​​and / or for manipulating the signal values. The filter is preferably a method for sliding (stepwise or continuous over time) processing.

[0042] A microphone in the sense of the present invention is preferably a sensor for converting sound into corresponding electrical oscillations.

[0043] An acoustic separator within the meaning of the present invention can be a preferably passive distributor for acoustic vibrations / sound. However, it is preferably switchable and can, alternatively or in addition to the distribution function, acoustically couple or decouple one or more sound pickups. In particular, it can decouple and recouple the sound pickup from the signal converter, so that sound waves corresponding to the noise no longer reach the signal converter, or can reach it again, so that it does not convert them into the auscultation signal, or converts them back into it.

[0044] A terminal device within the meaning of the present invention is preferably a mobile, battery-operated device with a user interface, such as a smartphone or tablet computer, but can also be a server or other external device. The terminal device is preferably designed or used to establish a data connection to the signal converter, to receive data from the signal converter, and to process the signal, and preferably to output the results.

[0045] A data connection in the sense of the present invention is preferably a particularly wireless coupling for data transmission, in particular with a protocol such as WLAN, WPAN or NFC.

[0046] A user interface within the meaning of the present invention is preferably an input and / or output device. It can be suitable or used for outputting information and / or inputting data and commands.

[0047] An analysis device within the meaning of the present invention is preferably a device for processing and evaluating signals. The analysis device is particularly designed to process and evaluate the auscultation signal in order to generate the result signal and preferably to identify an anomaly therein.

[0048] An anomaly within the meaning of the present invention is preferably a signal structure, in particular a characteristic pattern, a characteristic curve shape, a sequence of gradients, plateaus, characteristic amplitudes, or the like, that indicate a particularity. The anomaly is in particular an indicator, for example, of a pathological change in the organ.

[0049] A storage medium within the meaning of the present invention is preferably a device for non-volatile storage of data such as a program. It can be an internal or external memory such as a flash memory.

[0050] Applicators within the meaning of the present invention are preferably auscultation signal sections arranged in a three-dimensional diagram along the spatial axis. A connector within the meaning of the present invention is, in particular, a tube adapter for acoustically coupling the signal transducer to the sound pickup via a tube in such a way that the membrane vibrations cause the air in the tube to vibrate, and the signal transducer can receive these vibrations and convert them into the auscultation signal.

[0051] A signal value in the sense of the present invention is preferably a value of a signal or the auscultation signal at a specific point in time.

[0052] It is possible and preferred to apply the proposed aspects explained above and below to signals other than auscultation signals and to achieve corresponding advantages, even if the aspects of the present invention have proven to be particularly advantageous with regard to auscultation signals.

[0053] Thus, it is possible to apply the proposed aspects to preferably periodic signals (signals with a periodic component) rather than to auscultation signals in general, and in particular to signals with the previously explained useful component. Examples of these are ECG signals or, more generally, cardiogram signals. In principle, however, the aspects of the present invention can also be applied to other signals, in particular to signals that more generally correspond to preferably periodic mechanical, optical, and / or acoustic oscillations.

[0054] Therefore, the term "auscultation signal" used above and below, as well as in the claims, can be replaced, if necessary, by "preferably periodic signal," "signal with a periodic useful component," "ECG signal," or more generally, "cardiogram signal," and / or "signal corresponding to preferably periodic mechanical, optical, and / or acoustic oscillations." The aspects of acoustic recording are not provided for or are optional, while the focus is on the aspects of processing, with the formation of signal segments rather than auscultation signal segments. The formation and / or processing of the signal segments—preferably to form the result signal—is then advantageously carried out in the same way as the processing of the auscultation signal segments. This represents an independent complex of inventions.In particular, the following represents an independent invention complex: Method for processing a signal which preferably has a periodicity, wherein the signal is divided into signal intervals, preferably based on the periodicity, whereby signal sections are formed, wherein the signal sections are superimposed so that several signal values ​​are present for respective points in time due to the superimposition of the signal sections, and wherein a result signal is formed by processing the superimposed signal sections, wherein the processing comprises determining an individual value for the respective point in time and the result signal is or is formed from the temporal progression of the individual values.

[0055] It is preferred that the processing is or includes an analysis of a distribution of the signal values.

[0056] Alternatively or additionally, it is preferred that the processing or analysis of the distribution is or comprises processing the signal values ​​of superimposed signal sections with a filter, wherein the filter is non-linear, does not form an arithmetic mean and / or takes into account signal values ​​of different points in time to determine the respective individual value.

[0057] Alternatively or additionally, it is preferred that the respective individual value is determined in that signal values ​​which are essentially the same or originate from periodic portions of the signal are weighted more heavily than signal values ​​which are isolated or do not originate from periodic portions of the signal and which have a greater distance to neighboring signal values, and / or wherein the filter is or has a median filter, an adaptive filter and / or morphological filter.

[0058] Further aspects, advantages, and features of the present invention will become apparent from the claims and the following description of exemplary embodiments with reference to the drawings. It shows:

[0059] Fig. 1 shows a proposed system;

[0060] Fig. 2 an auscultation signal;

[0061] Fig. 3 Auscultation signal sections;

[0062] Fig. 4 shows a first result signal;

[0063] Fig. 5 shows a section of the first result signal; Fig. 6 shows an ambient noise signal;

[0064] Fig. 7 a second result signal

[0065] Fig. 8 a third result signal; and

[0066] Fig. 9 superimposed auscultation sections, a result signal based on a moving arithmetic average and an ideal result signal in a diagram of signal values ​​over time;

[0067] Fig. 10 is an enlarged section of the diagram according to Fig. 9;

[0068] Fig. 11 superimposed auscultation sections, a result signal based on a moving median filtering and an ideal result signal in a diagram of signal values ​​over time; and

[0069] Fig. 12 shows an enlarged section of the diagram according to Fig. 11.

[0070] In the figures, the same reference numerals are used for the same or similar elements, whereby the same or similar properties and advantages can be achieved, even if a repeated description is omitted.

[0071] Fig. 1 shows a proposed system 1. The system 1 is designed to process an auscultation signal 2, which is shown as an example in Fig. 2 in a diagram of a signal value U over time t. The auscultation signal 2 preferably has a periodicity 3, which can originate from periodic organ activity, in particular cardiac activity or lung activity.

[0072] Fig. 1 shows a proposed auscultation device 4. The auscultation device 4, preferably a stethoscope, has a sound pickup 5, in particular a chestpiece of the stethoscope, for auscultation (sound recording) of a noise 6.

[0073] For this purpose, the sound sensor 5 can have a membrane 7 that can be set into vibration by the noise 6. This vibration, which corresponds to the noise 6, can be conducted as sound through a tube 8 or other sound conductor, for example via an ear hook 9, to ear parts 10 such as ear tips of the auscultation device 4. The sound sensor 5 is preferably designed to auscultate noises 6 originating from an organ 11, indicated schematically in Fig. 1. The system 1 preferably has a signal converter 12 for converting the auscultated noise 6 into the auscultation signal 2. For this purpose, the signal converter 12 can have a microphone or be designed as a microphone in order to convert the sound that corresponds to the noise 6 or corresponds to the noise 6 into the auscultation signal 2.The microphone is integrated in the signal converter 12 and / or acoustically coupled or can be coupled to the sound pickup 5 via the tube 8.

[0074] The proposed system 1 may, but need not, include the auscultation device 4 or parts thereof, in particular the sound pickup 5. The system 1 may be intended to supplement an auscultation device 4 and thereby configure it to implement one or more of the method aspects of the present invention explained in more detail below. For this purpose, the signal converter 12 of the system 1 can be acoustically coupled to the sound pickup 5 in such a way that it can convert sounds 6 auscultated or auscultable with the sound pickup 5 into the auscultation signal 2.

[0075] The proposed auscultation device 4, in turn, preferably comprises the sound pickup 5 and the signal converter 12, but does not necessarily have to comprise further tubes 8, the ear hook 9 or the ear olives 10, which are therefore optional for the present invention.

[0076] In other words, the auscultation device 4 can be or comprise a so-called auscultation button, i.e., a device for attachment to a body for the purpose of auscultation and obtaining an auscultation signal 2, which can be stored, transmitted, and / or processed by the auscultation button. An auscultation button, on the other hand, preferably has no device for transmitting or emitting sound waves for direct detection by the sense of hearing.

[0077] Referring to Fig. 2, the auscultation signal 2 preferably has the aforementioned periodicity 3. The auscultation signal 2 can be or become divided into signal intervals 13, which are preferably oriented towards the periodicity 3. In particular, the auscultation signal 2 can be divided into the signal intervals 13 based on features of the periodicity 3. The result of dividing the auscultation signal 2 into signal intervals 13 are preferably the auscultation signal sections 14, which are shown as an example in Fig. 3. In principle, it is possible for an auscultation signal section 14 to have multiple signal intervals 13 or periods of the periodicity 3. In the illustrated example, a signal interval 13 extends beyond features of the auscultation signal 3 that represent the periodicity 3, so that an auscultation signal section 14 can also comprise part of another auscultation signal section 14.

[0078] Finally, a result signal 15A is preferably formed. An example of such a result signal 15A is shown in a diagram of a corresponding signal value U over time t in Fig. 4.

[0079] According to an aspect of the present invention that can also be implemented independently, the auscultation signal sections 14 are superimposed. The result signal 15A, as shown in Fig. 4, can be the result of this superimposition. Most preferably, the result signal 15A is formed by processing or density analysis of the superimposed auscultation signal sections 14.

[0080] The processing or density analysis can be carried out graphically or optically or be oriented thereto. The superimposition preferably results in several (individual) signal values ​​U being present per time point 18, as schematically illustrated by way of example in the enlargement in Fig. 5. The (individual) signal values ​​are each one signal value U per superimposed auscultation signal section 14 (and time point 18). Fig. 5 shows, in a diagram of the signal value U over time t, 18 (individual) signal values ​​U for three time points as an example. The time points 18 as well as the (individual) signal values ​​U can be discrete as a result of digitization, as shown in Fig. 5. This is beneficial for simplified representation, but is not mandatory. The (individual) signal values ​​U and time points 18 can be a partial enlargement 16 of the first result signal 15A, as indicated by the enlargement marking 16 in Fig. 5.

[0081] The superposition of the auscultation signal sections 14 preferably results in several (individual) signal values ​​U being present for the respective time points 18. The proposed processing or density analysis can be or include an analysis of a distribution of the (individual) signal values ​​U (per time point t) in order to form the result signal 15A.

[0082] In particular, the processing or density analysis is an analysis in which the signal values ​​U of the respective auscultation signal sections 14 resulting from the superposition are reduced to a single value 17 at the respective time 18. The single value 17, also called the value or signal value U of the highest density, is basically a signal value U at which most of the signal values ​​U of the respective auscultation signal sections 14 lie.

[0083] In the simplest case, the individual value 17 can at least substantially correspond to an average of the signal values ​​U. Alternatively or additionally, the individual value 17 can be determined by determining a distribution of signal values ​​U, in particular by interpolating the distribution over the signal values ​​U at the respective time 18.

[0084] For example, it can be assumed that the distribution of signal values ​​U is based on a certain distribution, such as a normal distribution, which can then be determined and used to determine the individual value 17, where the individual value 17 can correspond to the expected value of the distribution or be derived therefrom.

[0085] The processing, density examination or analysis of the distribution is or preferably comprises a processing of the signal values ​​U of superimposed auscultation signal sections 14 with a filter.

[0086] The filter preferably does not calculate an arithmetic mean. While it is fundamentally possible to generate an arithmetic mean or to consider it additionally, the proposed processing or density analysis has proven advantageous.

[0087] Fig. 9 and a partial enlargement according to Fig. 10 show the example of a moving average calculation in comparison to a proposed processing or density analysis according to Fig. 11 and a partial enlargement according to Fig. 12 using diagrams of signal values ​​U over time t, in which superimposed auscultation signal sections 14, the first result signal 15A-1, 15A-2 and an ideal result signal 16 are plotted. The axis scaling is purely exemplary and is plotted in seconds on the abscissa axis and in mV on the ordinate axis. These are digital signals, whereby seconds can be converted into samples based on a sample rate (here, for example, 16,000 samples per second). For each sample or the corresponding time, the signal values ​​U of the (in particular all) superimposed auscultation signal sections 14 are preferably available (simultaneously).For digital signals such as these, a point in time is preferably the point in time of a specific sample.

[0088] The auscultation signal sections 14 of the examples from Figs. 9 to 12 are derived from the ideal result signal 16 by adding noise so that the result signals 15 can be evaluated in comparison. The moving averaging of the proposed processing or density analysis is based on the same superimposed auscultation signal sections 14. Thus, the diagrams differ by different result signals 15A-1, 15A-2 (shown in dashed lines unless they overlap with the ideal result signal), which are formed, on the one hand, by the moving arithmetic averaging and, on the other hand, by the moving median filtering of the signal values ​​U of the superimposed auscultation signal sections 14.

[0089] Fig. 9 shows the result signal 15A1, which was generated by arithmetic averaging of the signal values ​​U at the respective time. The diagram shown in Fig. 10 shows an enlargement of the diagram shown in Fig. 9 according to the magnification marking 16 there. Here, significant deviations of the result signal 15A1 generated by arithmetic averaging from the ideal result signal 16 are evident, particularly at smaller amplitudes.

[0090] The diagram shown in Fig. 11 shows the result signal 15A2, which was generated by processing the signal values ​​U at the respective time with a median filter. The diagram shown in Fig. 12 shows an enlargement of the diagram shown in Fig. 11 according to the enlargement marking 16 there. Here, hardly any deviations are evident between the result signal 15A2 generated by applying the median filter and the ideal result signal 16, especially at smaller amplitudes, compared to the result signal 15A1 generated by arithmetic averaging. A deviation from the ideal result signal 16 is only evident at larger amplitudes.

[0091] This example illustrates that the proposed processing or density analysis has significant advantages over a moving arithmetic mean calculation. The use of a median filter is particularly preferred, but other filters also offer advantages over the moving arithmetic mean calculation. In addition to the median filter, general ranking filters and other nonlinear filters have proven advantageous. These include, in addition to the median filter and ranking filters, adaptive filters and / or morphological filters.

[0092] Alternatively or additionally, it has surprisingly proven advantageous to consider signal values ​​U from different points in time 18 to determine the respective individual value 17. In particular, signal values ​​U from adjacent samples / points in time can be considered to determine the respective individual value 17. It is therefore possible and preferred to consider signal values ​​U of the different auscultation signal sections 14 from different points in time in a filter function of the filter to determine an individual value 17 at a specific point in time.

[0093] The achievable advantage is visually similar to that of two-dimensional filtering, in particular median filtering, of two-dimensional images, in which the pixels adjacent on all sides are included for the recalculation of a pixel. In the present case, instead, temporally adjacent signal values ​​U (in particular of the immediately adjacent sample(s) / time points and / or all auscultation signal sections 14) are taken into account alternatively or in addition to the signal values ​​U present at the respective time based on the superposition of auscultation signal sections 14. This allows, in particular, high-frequency interference from the respective auscultation signal sections 14 to be effectively suppressed, similar to the suppression of salt-and-pepper noise in image processing.

[0094] It is generally particularly preferred that the filter is or has a median filter, as in the example shown. A median filter can sort the signal values ​​U according to size and returns the middle signal value U as the result (individual value 17) if the number of signal values ​​U is odd, and the arithmetic mean of the two middle signal values ​​U if the number of signal values ​​U is even. Median filters advantageously increase the computational effort by a generally acceptable amount compared to moving averaging. The other filter groups described have, however, also proven advantageous with different ratios of computational effort to accuracy, while for reasons of clarity, further examples are omitted. Since the filters are generally applied successively for each sample or point in time, they are preferably so-calledsliding applications of the filters, even if this is not explicitly stated.

[0095] The individual values ​​17 can then form the first result signal 15A or form the basis for determining the first result signal 15A. For this purpose, the individual values ​​17 can be linked to one another, or a compensation curve can be aligned with them, which then forms the result signal 15A.

[0096] In other words, the processing or density analysis comprises the determination of individual values ​​17 from the signal values ​​U for the respective time 18 and the result signal 15A is or is formed from the temporal course of the individual values ​​17.

[0097] The individual values ​​17 can be determined by giving greater weight to signal values ​​U that are essentially the same or originate from periodic components of the auscultation signal 2 than to signal values ​​U that are at a greater distance from neighboring signal values ​​U (which deviate greatly), are isolated or do not originate from periodic components.

[0098] An example of a more heavily weighted signal value 19 is shown in Fig. 5. A more heavily weighted, for example multiply weighted, signal value 19 can shape or influence the individual value 18 correspondingly more strongly than less heavily weighted signal values ​​U.

[0099] The processing or density analysis may comprise the application of a contrast enhancement method and / or a sheep drawing method to the overlay in order to reduce the signal values ​​per time point 18 to a single value 17 and to form the result signal 15A from the temporal progression of the single values ​​17.

[0100] As previously explained, a diagram of superimposed auscultation signal segments 14 can be interpreted as an image, where a higher density (darkness of pixels) corresponds to, or can be derived from, a higher density (closer distance between signal values ​​U). Visualized to the human eye, this example corresponds to or resembles superimposed transparencies with printouts of the auscultation signal segments 14. This is because an area becomes darker where many auscultation signal segments 14 are closer together, i.e., they approach or intersect.

[0101] As with the previously explained filtering, a contrast enhancement method and / or sheep drawing method known from or based on the processing of two-dimensional images in particular can be used to generate the respective individual value 17 from the signal values ​​U. In this context, high-pass filtering, edge detection, and / or edge localization, for example, have proven advantageous. For details, reference is made to the relevant literature, while it has proven surprisingly advantageous to apply correspondingly known methods in the presently technically remote area.

[0102] As a result, the auscultation signal 2 is preferably divided into signal intervals 13 based on the periodicity 3, whereby the auscultation signal sections 14 are formed, preferably wherein the auscultation signal sections 14 are superimposed and a result signal 15A is formed by processing or density analysis of the superimposed auscultation signal sections 14.

[0103] The course of the individual values ​​17 can be smoothed to form the result signal 15A.

[0104] The formation of the auscultation signal segments 14 based on the periodic signal components, i.e., the components of the auscultation signal 2 that exhibit or characterize the periodicity 3, i.e., are periodic, is preferably carried out in such a way that the periodic signal components are essentially congruent upon superposition. In particular, periodic structures in the auscultation signal 2 are identified and used to define the signal intervals 13 and to form auscultation signal segments 14 that fall within the signal intervals 13.

[0105] The auscultation signal sections 14 are preferably formed using one or more correlations and / or superimposed substantially congruently.

[0106] Preferably, an autocorrelation (correlation with itself) of the auscultation signal 2 or parts thereof is used to determine, via the periodic similarity of the auscultation signal sections 14, the points in time between which the auscultation signal 2 is divided into the auscultation signal sections 14. Optionally, the auscultation signal 2 can be preprocessed before the autocorrelation. Alternatively or additionally, the auscultation signal 2 can be analyzed, for example, with regard to previously known features of the periodicity 3 of the auscultation signal 2, in order to determine the points in time at which the auscultation signal 2 is divided into the auscultation signal sections 14.

[0107] Alternatively or additionally, the auscultation signal sections 14 are superimposed by bringing them into a relative position to one another at which the cross-correlation of the auscultation signal sections 14 is maximum.

[0108] Alternatively or additionally, subdivision and / or superimposition can be performed based on features of the course of the auscultation signal 2 or the auscultation signal segments 14. However, correlation has proven particularly efficient and effective.

[0109] Ultimately, the amplitude of the periodic component of the auscultation signal 2 formed by periodicity 3 can advantageously be increased relative to non-periodic components of the auscultation signal 2. The proposed method thus enables the selective amplification of the periodic components of the auscultation signal 2 and thus an improvement in the signal-to-noise ratio of useful components to noise components of the auscultation signal 2.

[0110] The overlay can be represented in a common two-dimensional diagram having a time axis for the times t and a value axis for the signal values ​​U. Auscultation sections 14 arranged one after the other in a two-dimensional diagram having a time axis and a value axis—in particular in an (imaginary) three-dimensional diagram along an application axis—can be or will be projected onto the common two-dimensional diagram. It is understood that the graphical representation itself is not absolutely necessary, but can be or will be replaced by corresponding or similar signal-related or mathematical operations.

[0111] In summary, the auscultation signal 2, which is of interest and to be evaluated, is periodic and, if necessary, essentially known in terms of its form or signal waveform. Therefore, it is possible to identify a periodic (useful) portion of the auscultation signal 2 and divide the auscultation signal 2 into periodic sequences, the auscultation signal segments 14, based on the periodicity 3.

[0112] In the present aspect of the invention, the auscultation signal segments 14 are then superimposed. The useful portion of the auscultation signal 2 to be evaluated, i.e., the information-laden portion, is largely identical in its form and progression across the auscultation segments 14. Portions of the auscultation signal 2 originating from unwanted background or interfering noise, on the other hand, have a random occurrence or, preferably, a periodicity that deviates from the useful signal's periodicity 3.

[0113] By superimposing the auscultation signal sections 14 and subsequently evaluating them by the proposed processing or density analysis, the signal-to-noise ratio can be increased, thus emphasizing the useful component compared to noise signals.

[0114] In a further aspect of the present invention, which can also be implemented independently, the auscultation signal sections 14 are added, and a result signal 15A can be formed by the resulting sum signal or can be derived therefrom. The addition of the auscultation signal sections 14 preferably means an addition of the respective signal values ​​U of the respective auscultation section 14 for the respective time 18. The sum signal is the curve of the individual values ​​17 resulting from the addition.

[0115] The formation of the auscultation signal sections 14 is preferably based on the periodic components of the auscultation signal 2 such that the periodic components of the auscultation signal 2 are substantially congruent before or at least during the addition or superposition.

[0116] In principle, the result signal 15A can be or will be standardized. However, it is also possible to provide or evaluate the result signal 15A without such standardization.

[0117] As already explained, the amplitude of the periodic component of the auscultation signal 2 formed by the periodicity 3 is preferably increased in relation to non-periodic components or components with a frequency of the auscultation signal 14 that deviates from the periodicity 3 (of the useful signal), either by processing or density analysis after superposition, by addition or, what is more preferred, by a combination of both procedures.

[0118] This is because when the auscultation signal sections 14 are added and the first result signal 15A is formed therefrom, non-periodic components are eliminated in a different way than by the processing or density analysis after superimposing the auscultation signal sections 14. The addition can lead to an increase in the amplitude 20 of the result signal 15A or the useful component in the result signal 15A compared to that of the auscultation signal sections 14, or the amplitude 20 can be essentially unchanged if a normalization is carried out after the addition, for example the sum of the signal values ​​U is divided by the number of added auscultation signal sections 14 in order to form the result signal 15A.

[0119] However, the respective improvement in the signal-to-noise ratio of the periodic useful component in the auscultation signal 2 or in the resulting result signal 15A can vary between the described methods and, in combination, can therefore be synergistically advantageous. On the one hand, a later selection of the better result signal 15A or the combination of the result signals 15A into a common, even better result signal 15A is possible, in which the periodic useful component of the auscultation signal 2 is emphasized even more strongly than in the individual result signals 15A compared to interference variables such as noise or the like.

[0120] In a further aspect of the present invention, which can also be implemented independently, ambient noise signal components of the auscultation signal 2 that are interspersed by ambient noise are eliminated, which can lead to a second result signal 15B, as shown in Fig. 7. A combination with the superposition or addition of signal intervals 13 can synergistically lead to the third result signal 15C. An example of ambient noise or a corresponding ambient noise signal 21 is shown in Fig. 6. The ambient noise can be noise on the one hand, but can also be artificially generated sound events acting on the system 1 or the auscultation device 4 on the other.

[0121] The ambient noise signal components in the auscultation signal 2 are preferably eliminated by capturing the ambient noise with one or more microphones 23, and thereby converting the ambient noise into the ambient noise signal 21. The ambient noise signal 21 is preferably filtered with a filter 22, in particular wherein the filter characteristic of the filter 22 is derived from a transmission behavior of the auscultation device 4 for the ambient noise to the signal converter 12.

[0122] The filtered ambient noise signal 21 then essentially corresponds to the ambient noise signal components that are converted into the auscultation signal 2 by the signal converter 12 in addition to the auscultated noise 6. Accordingly, the ambient noise signal components are preferably eliminated by subtracting the filtered ambient noise signal 21 from the auscultation signal 2.

[0123] The filter characteristic of filter 22 is preferably determined by masking out the auscultated sound 6 for the signal converter 12, so that the signal converter 12 only converts the portions of the ambient noise transmitted by the auscultation device 4 to the signal converter 12. The result is then compared with the ambient noise signal 21 to form the filter characteristic. In other words, the filter characteristic preferably corresponds at least substantially to a sound transfer function from the environment of the auscultation device 4 to the signal converter 12.

[0124] The operating mode for determining the filter characteristic, in which the auscultated noise 6 is masked out for the signal converter 12, preferably represents a temporary calibration or adjustment process. In normal operation, however, the auscultated noise 6 is not masked out for the signal converter 12. Rather, the auscultated noise 6 is converted into the auscultation signal 2 during normal operation and, if possible, freed from interference.

[0125] Particularly preferably, the signal converter 12 has one or more microphones 23 for recording ambient noise, the microphone(s) 23 is / are therefore arranged for recording ambient noise and is preferably designed to convert airborne noise from the environment into the ambient noise signal 21.

[0126] However, it is also possible for the auscultation device 4 to have the microphone(s) 23, in particular at a location other than the signal converter 12. The arrangement of a microphone 23 for recording the ambient noise signal 21 in the area between the signal converter 12 and the sound pickup 5 has proven particularly advantageous for the efficiency of suppressing ambient noise signal components in the auscultation signal 2.

[0127] The signal converter 12 may, as shown in Fig. 1, further comprise a function key 24 with which the signal converter 12 can be activated, deactivated and / or a coupling can be initiated.

[0128] Alternatively or additionally, the signal converter 12 has an acoustic filter 25, with which sound waves originating from the sound sensor 5 can be temporarily filtered out for the signal converter 12 in order to determine the filter characteristics. The acoustic filter 25 thus preferably enables a temporary acoustic decoupling of the signal converter 12 from the sound sensor 5. Alternatively or additionally, the acoustic filter 25 can be used to guide the sound 6 both to the signal converter 12 and to the earpieces 10.

[0129] A preferably wireless data connection 27 is or is preferably established between the terminal 26, which is preferably spatially separated from the auscultation device 4, and the signal converter 12, as schematically indicated in Fig. 1.

[0130] To establish the data connection 27, a pairing mode can be activated using the function key 24. The pairing can be established via WLAN, NFC, or WPAN, for example, via Bluetooth, or the data connection 27 can enable communication via corresponding protocols.

[0131] The auscultation signal 2 and / or a result signal 15A, 15B, 15C formed therefrom can be transmitted to the terminal 26 via a data connection 27.

[0132] The terminal 26 may alternatively or in addition to the signal converter 12 or the auscultation device 4 have the or one of the microphones 23 in order to record the ambient noise signal 21.

[0133] The terminal 26 can be or comprise a mobile phone or tablet computer, as shown in Fig. 1. Alternatively or additionally, it can be a server or other central device or a distributed structure comprising the latter. The terminal 26 is preferably mobile (battery-operated or rechargeable), in particular as a smartphone, tablet, or the like. In principle, however, the terminal 26 can also be stationary and / or wired.

[0134] The terminal 26 can have a user interface 28 such as a touch display and / or other input and / or output means. The user interface 28 can be configured to output the result signal 15A, 15B, 15C. Alternatively or additionally, the user interface 28 can receive commands with which the proposed method can be controlled or influenced.

[0135] The microphones 23 of the auscultation device 4 / the signal converter 12 and / or the terminal device 26 can be combined with one another to form the ambient noise signal 21, i.e. can be combined with one another or used in combination to eliminate the ambient noise signal components from the auscultation signal 2.

[0136] The microphone(s) 23 are preferably designed as room microphones for detecting sound from the spatial environment of the auscultation device 4, which is unaffected by the auscultated sound 6. The signal converter 12 can also comprise or be formed by a microphone for converting the sound 26 into the auscultation signal 2, but preferably receives sound waves generated by the membrane 7 directly or via the tube 8. In this respect, the signal converter 12 differs fundamentally from the microphone(s) 23.

[0137] If the ambient noise signal 21 is eliminated from the auscultation signal 2 shown as an example in Fig. 2, as proposed, the second result signal 15B shown schematically in Fig. 7 preferably results.

[0138] It is understood that the combination of the proposed methods for improving signal quality is particularly preferred. In particular, it is particularly preferred to combine the elimination of the ambient noise signal component from the auscultation signal 2 with the formation of the auscultation signal sections 14 and the addition and / or density analysis based on this.

[0139] It is preferable to eliminate the ambient noise signal 21 from the auscultation signal 2 continuously and synchronously, preferably before the auscultation signal sections 14 are formed and further processed. It has been found that it can often be assumed that the ambient noise is at least substantially random or aperiodic, or at least has a periodicity that does not correspond to the periodicity of the noise 6.

[0140] In principle, the ambient noise signal components of auscultation signal 2 are therefore also those components of auscultation signal 2 whose influence can be reduced through density analysis or the addition of auscultation signal segments 14. Surprisingly, however, it has been shown that the combination of these measures, i.e., further processing of auscultation signal 2 by eliminating the ambient noise signal components 21, leads to a better signal-to-noise ratio than the individual measures. In this respect, they are surprisingly synergistic and are therefore preferably used in combination with one another.

[0141] The third result signal 15C, which is or is generated from a combination of the elimination of the ambient noise signal component from the auscultation signal 2 as well as the formation of the auscultation signal sections 14 and the density analysis and / or addition of these, is shown schematically in Fig. 8.

[0142] The system 1 or the auscultation device 4 preferably has / have an analysis device 29 which forms the result signal 15A, 15B, 15C by evaluating the auscultation signal 2.

[0143] The signal converter 12 or the terminal device 26 can, preferably with the analysis device 29, detect an anomaly 30 in the result signal 15A, 15B, 15C or be configured to do so. For this purpose, the result signal 15A, 15B, 15C is evaluated to identify the anomaly 30. An anomaly 30 is considered, in particular, an indicator of a pathological change, thus indicating a disease. The anomaly 30 can be marked and / or an indication thereof can be output, in particular via the user interface 28 of the terminal device 26.

[0144] The detection rate for identifying the anomaly 30 is significantly improved by the proposed processing steps of the auscultation signal 2, in particular in different combinations of the proposed processing steps of the auscultation signal 2 - ambient noise signal elimination, auscultation signal section superposition and density analysis, and / or auscultation section addition.

[0145] The analysis device 29 is preferably structurally separate from the auscultation device 4 and coupled via the preferably wireless data connection 27 to the signal converter 12, which transmits the auscultation signal 2 to the analysis device 29 via the data connection 27. However, the analysis device 29 can also be formed entirely or partially by the signal converter 12.

[0146] The analysis device 29 preferably evaluates the result signal 15A, 15B, 15C to identify the anomaly 30. Optionally, the user interface 18 can also output the result if no anomaly 30 could be identified by evaluating the auscultation signal 2.

[0147] The auscultation device 4 and / or the system 1 is / are preferably designed to carry out one or more aspects of the proposed method.

[0148] In an aspect of the present invention that can also be implemented independently, the system 1 comprises the signal converter 12 for converting the auscultated sounds into the auscultation signal 2 for the auscultation device 4, which has a connection or connector C for physical and acoustic coupling to a sound pickup 5 and / or earpieces 10 of the auscultation device 4, so that the auscultated sounds 6 recorded by the sound pickup 5 can be converted into the auscultation signal 2 by the signal converter 12. Furthermore, the system 1 preferably comprises the terminal device 26 or the analysis device 29, which is / are designed to evaluate the auscultation signal 2. The system 1, the signal converter 12 and / or the terminal device 26 of the system 1, which can preferably be coupled wirelessly or is coupled to the signal converter 12 and is implemented separately, preferably comprises the analysis device 29 completely or partially.

[0149] The analysis device 29 can alternatively or additionally be implemented entirely or partially externally, for example, in the cloud or as a cloud application. In this case, the terminal device 26 can establish the data connection 27 from the signal converter 12 to the analysis device 29. A further aspect of the present invention relates to a computer program product comprising instructions which, when executed on a processor of the analysis device 29 of the proposed system 1, effect implementation of the steps of the proposed method. Furthermore, a further, likewise independently implementable aspect of the present invention relates to a computer-readable storage medium 31 on which the proposed computer program product is stored.

[0150] The computer program product can, in particular, effect the improvement measures for the auscultation signal 2 in the signal converter 12 and / or terminal device 26. Thus, the computer program product can be configured to identify the signal intervals 13 and, based on these, to form the auscultation signal sections 14 with the auscultation signal 2. Preferably, the computer program product is further configured to superimpose and / or add the auscultation signal sections 14, to perform the density analysis if necessary, and / or to effect a normalization of the result signal 15A, 15B, 15C.

[0151] Alternatively or additionally, the computer program product can be configured to filter ambient noise signals 21 originating from one or more of the microphones 23 and to eliminate the filtered ambient noise signal 21 from the auscultation signal 2. The computer program product preferably performs the latter before forming the auscultation signal sections 14.

[0152] In principle, the computer program product can be executed entirely on the terminal device 26, while the auscultation signal 2 and, if applicable, the ambient noise signal 21 are transmitted from the signal converter 12 to the terminal device 26.

[0153] However, it is preferred that the computer program product for eliminating the ambient noise component from the auscultation signal 2 is stored on a storage medium 31 of the signal converter 12 and is executable or executed on the signal converter 12, while the further evaluation of the result, i.e., the auscultation signal 2 with the ambient noise component eliminated, is or can be carried out by a processor of the terminal device 26, with the computer program product being stored on the storage medium 31 of the terminal device 26. However, other solutions are also possible here. Further aspects of the present invention, which can also be combined with the aspects described above, are:

[0154] 1. Method for processing an auscultation signal 2, which preferably has a periodicity 3, wherein the auscultation signal 2 is or is generated with an auscultation device 4, in particular a stethoscope, having a sound sensor 5, in particular a chest piece of the stethoscope, for auscultating noises 6, in particular internal organs 11 and a signal converter 12 physically and acoustically coupled to the sound sensor 5 for converting the auscultated noises 6 into the auscultation signal 2, wherein the auscultation signal 2 is preferably divided into signal intervals 13 based on the periodicity 3, whereby auscultation signal sections 14 are formed, and wherein the auscultation signal sections 14 are added and a result signal 15A is or is formed by the resulting sum signal or is derived therefrom.

[0155] 2. Method according to aspect 1, characterized in that the formation of the auscultation signal sections 14 based on the signal intervals 13 takes place in such a way that the auscultation signal sections 14 are substantially congruent before the addition.

[0156] 3. Method according to aspect 1 or 2, characterized in that the result signal 15A is or is normalized.

[0157] 4. Method according to one of the preceding aspects, characterized in that the amplitude of the periodic component of the auscultation signal 2 formed by the periodicity 3 is increased in relation to non-periodic components of the auscultation signal 2.

[0158] 5. Method according to one of the preceding aspects, characterized in that the auscultation device 4 further comprises an analysis device 29, which forms the result signal 15A by evaluating the auscultation signal 2. 6. Method aspect 5, characterized in that the analysis device 29 is structurally separate from the auscultation device 4 and is coupled via a preferably wireless data connection 27 to the signal converter 12, which transmits the auscultation signal 2 to the analysis device 29 via the data connection 27.

[0159] 7. Method according to aspect 5 or 6, characterized in that the analysis device 29 evaluates the result signal 15A in order to identify an anomaly 30.

[0160] 8. Method according to one of the preceding aspects, characterized in that the auscultation signal sections 14 are superimposed and the result signal 15A is optimized by a density analysis of the superimposed auscultation signal sections 14.

[0161] 9. Method according to one of the preceding aspects, characterized in that ambient noise signal components of the auscultation signal 2 which are scattered by ambient noise are eliminated.

[0162] 10. Auscultation device 4, in particular a stethoscope, which has a sound pickup 5, in particular the chest piece of the stethoscope, for auscultating sounds 6, in particular internal organs 11, and which has a signal converter 12 physically and acoustically coupled to the sound pickup 5 for converting the auscultated sounds 6 into an auscultation signal 2, wherein the auscultation device 4 is designed to carry out a method according to one of the preceding claims.

[0163] 11. Auscultation device according to aspect 10, characterized in that the auscultation device 4 further comprises one or more ear olives 10 physically and acoustically coupled to the sound sensor 5 and / or an acoustic switch 25, in particular wherein the switch 25 enables a temporary acoustic decoupling of the signal transducer 12 from the sound sensor 5.

[0164] 12. System 1 for an auscultation device 4, preferably according to aspect 10 or 11, wherein the system 1 has a signal converter 12 for converting the auscultated sounds 56 into an auscultation signal 2, which has a connector C for physical and acoustic coupling to a sound pickup 5, in particular a chest piece of a stethoscope, so that the auscultated sound 6 recorded with the sound pickup 5 can be converted by the signal converter 12 into an auscultation signal 2, and wherein the system 1 has an analysis device 29 which is designed to divide the auscultation signal 2 into signal intervals, preferably based on the periodicity 3, whereby auscultation signal sections 14 are formed, to add the auscultation signal sections 14 and thereby to form a result signal 15A.

[0165] 13. System according to aspect 12, characterized in that the signal converter 12 and / or a terminal device 26 of the system 1, which can preferably be coupled wirelessly to the signal converter 12 and is implemented separately, has the analysis device 29.

[0166] 14. A computer program product comprising instructions which, when executed on a processor of an analysis device 29 of a system according to claim 12 or 13, cause the steps of the method according to any one of aspects 1 to 9 to be carried out.

[0167] 15. Computer-readable storage medium on which the computer program product according to aspect 14 is stored.

[0168] Further aspects of the present invention, which can also be combined with the aspects described above, are:

[0169] 1. Method for processing an auscultation signal 2, wherein the auscultation signal 2 is or is generated with an auscultation device 4, in particular a stethoscope, having a sound pickup 5, in particular a chest piece, for auscultating noises 6, in particular internal organs, and a signal converter 12 physically and acoustically coupled to the sound pickup 5 for converting the auscultated noises 6 into the auscultation signal 2, and wherein ambient noise signal components of the auscultation signal 2 interspersed by ambient noise are eliminated.

[0170] 2. Method according to aspect 1, characterized in that the ambient noise signal components are eliminated by detecting the ambient noise with one or more microphones 23, converting it to an ambient noise signal 21 and filtering the ambient noise signal 21 with a filter 22, wherein the filter characteristic of the filter 22 is derived from a transmission behavior of the auscultation device 4 for the ambient noise to the signal converter 12.

[0171] 3. Method according to aspect 2, characterized in that the filter characteristic is or is determined by masking out the auscultated noise 6 for the signal converter 12, so that the signal converter 12 only converts the components of the ambient noise transmitted by the auscultation device 4 to the signal converter 12 and the result is then compared with the ambient noise signal 21 in order to form the filter characteristic.

[0172] 4. Method according to aspect 2 or 3, characterized in that the auscultation device 4 has the or one of the microphones 23, preferably the signal converter 12.

[0173] 5. Method according to one of aspects 2 to 4, characterized in that a preferably wireless data connection 27 is or is established between a terminal 26 spatially separated from the auscultation device 4 and the signal converter 12, preferably wherein the terminal 26 has the or one of the microphones 23.

[0174] 6. Method according to aspect 5, characterized in that the auscultation signal 2 and / or a result signal 15A, 15B, 15C formed therefrom is transmitted to the terminal 26 via the data connection 27.

[0175] 7. Method according to one of aspects 2 to 6, characterized in that the microphone(s) 23 is / are designed as room microphones for detecting ambient noise from the spatial environment of the auscultation device 4, which is / are unaffected by the auscultated noise.

[0176] 8. Method according to one of the preceding aspects, characterized in that the result signal 15A, 15B, 15C is formed and / or further analyzed by the terminal 26, preferably wherein the terminal 26 evaluates the result signal 15A, 15B, 15C to identify an anomaly 30.

[0177] 9. Method according to one of the preceding aspects, characterized in that the auscultation signal 2 is preferably divided into signal intervals 13 based on the periodicity 3, whereby auscultation signal sections 14 are formed, preferably wherein the auscultation signal sections 14 are superimposed and a result signal 15A, 15B, 15C is formed by a density analysis of the superimposed auscultation signal sections 14; and / or wherein the auscultation signal sections 14 are added and a result signal 15A, 15B, 15C is formed by the resulting sum signal or is or is derived therefrom.

[0178] 10. Auscultation device 4, in particular a stethoscope, which has a sound sensor 5, in particular a chest piece, for auscultating sounds 6, in particular internal organs, and which has a signal converter 12 physically and acoustically coupled to the sound sensor 5 for converting the auscultated sounds 6 into an auscultation signal 2, wherein the auscultation device 4 is designed to carry out a method according to one of the preceding claims.

[0179] 11. Auscultation device according to aspect 10, characterized in that the auscultation device 4 further comprises one or more ear parts 10, in particular ear olives, physically and acoustically coupled to the sound sensor 5 and preferably an acoustic switch 25, in particular wherein the switch 25 enables a temporary acoustic decoupling of the signal transducer 12 from the sound sensor 5.

[0180] 12. System 1 for an auscultation device 4, preferably according to aspect 10 or 11, wherein the system 1 has a signal converter 12 for converting the auscultated sounds 6 into an auscultation signal 2, which has a connector C for physical and acoustic coupling to a sound sensor 5, so that the auscultated sound 6 recorded with the sound sensor 5 can be converted into an auscultation signal 2 by the signal converter 12, and wherein the system 1 has an analysis device 29 which is designed to divide the auscultation signal 2 into signal intervals 13, preferably based on the periodicity 3, whereby auscultation signal sections 14 are formed, and wherein a result signal 15A, 15B, 15C is formed by adding and / or superimposing the auscultation signal sections 14.

[0181] 13. System according to aspect 12, characterized in that the signal converter 12 and / or a terminal device 26 of the system 1 that can be wirelessly coupled to the signal converter 12 and is structurally separate comprises the analysis device 29. 14. A computer program product comprising instructions that, when executed on a processor of an analysis device 29 of a system 1 according to aspect 12 or 13, effect implementation of the steps of the method according to one of aspects 1 to 9.

[0182] 15. Computer-readable storage medium on which the computer program product according to aspect 14 is stored.

[0183] The different aspects of the present invention can be advantageously combined with one another, but can also be advantageously implemented separately, even if individual combinations are not explicitly explained.

[0184] List of reference symbols: System 18. Time point Auscultation signal 19. Weighted signal value Periodicity 20. Amplitude Auscultation device 21. Ambient noise signal Sound pickup 22. Filter Noise 23. Microphone Diaphragm 24. Function key Tube 25. Acoustic crossover Ear hook 26. Terminal . Earpiece 27. Data connection . Organ 28. User interface . Signal converter 29. Analysis device . Signal interval 30. Anomaly . Auscultation signal section 31. Storage medium A. First result signal B. Second result signal A Applicants C. Third result signal C Connector D. Ideal result signal t Time . Magnification marker U Signal value . Single value

Claims

Patent claims:

1. A method for processing an auscultation signal (2), which preferably has a periodicity (3), wherein the auscultation signal (2) is generated by an auscultation device (4), preferably a stethoscope, having a sound pickup (5), in particular a chest piece, for auscultating sounds (6), in particular internal organs (11), and a signal converter (12) physically and acoustically coupled to the sound pickup (5) for converting the auscultated sounds (6) into the auscultation signal (2), wherein the auscultation signal (2), preferably based on the periodicity (3), is divided into signal intervals (13), whereby auscultation signal sections (14) are formed, wherein the auscultation signal sections (14) are superimposed, so that by superimposing the auscultation signal sections (14) for respective points in time (18), several signal values (U) are present, and wherein a result signal (15A, 15C) by processing,in particular density examination, the superimposed auscultation signal sections (14) are formed, wherein the processing comprises the determination of an individual value (17) for the respective time (18) and the result signal (15A, 15C) is or is formed from the temporal course of the individual values (17).

2. Method according to claim 1, characterized in that the processing is or comprises an analysis of a distribution of the signal values (U).

3. Method according to claim 1 or 2, characterized in that the processing or analysis of the distribution is or comprises processing of the signal values (U) of superimposed auscultation signal sections (14) with a filter, wherein the filter is non-linear, does not form an arithmetic mean and / or takes into account signal values (U) of different times (18) for determining the respective individual value (17).

4. Method according to claim 3, characterized in that the respective individual value (17) is determined by substantially equal signal values (u) or signal values originating from periodic portions of the auscultation signal (14) being are weighted as signal values (U) which are at a greater distance from neighbouring signal values (U), which are individual or do not originate from periodic components of the auscultation signal (14), and / or wherein the filter is or has a median filter, an adaptive filter and / or morphological filter.

5. Method according to one of the preceding claims, characterized in that the processing comprises the application of a contrast enhancement method and / or a sheep drawing method to the overlay in order to reduce the signal values (U) per time point (18) to the individual value (17), preferably wherein the contrast enhancement method and / or sheep drawing method is or comprises high-pass filtering, edge detection and / or edge localization.

6. Method according to one of the preceding claims, characterized in that the course of the individual values (17) is smoothed in order to form the result signal (15A, 15C).

7. Method according to one of the preceding claims, characterized in that the formation of the auscultation signal sections (14) based on the periodicity of the auscultation signal (2) takes place in such a way that features of the periodicity (3) are substantially congruent during the superposition.

8. Method according to one of the preceding claims, characterized in that the auscultation signal sections (14) are formed by correlation and / or are superimposed substantially congruently.

9. Method according to one of the preceding claims, characterized in that the superposition can be represented in a common two-dimensional diagram having a time axis and a value axis, in that - in particular in a three-dimensional diagram with along an axis for applicates a - auscultation sections (14) arranged one after the other in a two-dimensional diagram having a time axis and a value axis are or are projected onto the common two-dimensional diagram.

10. Auscultation device (4), in particular a stethoscope, comprising a sound pickup (5), in particular a chest piece, for auscultating sounds (6), in particular internal organs (11), and which has a signal converter (12) physically and acoustically coupled to the sound sensor (5) for converting the auscultated sounds (6) into an auscultation signal (2), wherein the auscultation device (4) is designed to carry out a method according to one of the preceding claims.

11. Auscultation device according to claim 10, characterized in that the auscultation device (4) has one or more ear olives (10) physically and acoustically coupled to the sound sensor (5) and / or an acoustic switch (25), in particular wherein the switch (25) enables a temporary acoustic decoupling of the signal converter (12) from the sound sensor (5).

12. System (1) for an auscultation device (4), preferably according to claim 10 or 11, wherein the system (1) comprises a signal converter (12) for converting the auscultated sounds (6) into an auscultation signal (2), which has a connector (C) for physical and acoustic coupling with a sound pickup (5), in particular a chest piece, so that the auscultated sound (6) recorded with the sound pickup (5) can be converted into an auscultation signal (2) by the signal converter (12), and wherein the system (1) comprises an analysis device (29) which is designed to divide the auscultation signal (2) into signal intervals (13), preferably based on the periodicity (3), whereby auscultation signal sections (14) are formed, to superimpose the auscultation signal sections (14) and to generate a result signal (15A, 15C) by processing of the superimposed auscultation signal sections (14).

13. System according to claim 12, characterized in that the signal converter (12) and / or a terminal device (26) of the system (1) which can be wirelessly coupled to the signal converter (12) and is structurally separate has the analysis device (29).

14. A computer program product comprising instructions which, when executed on a processor of an analysis device (29) of a system according to claim 12 or 13, cause the steps of the method according to any one of claims 1 to 9 to be carried out.

15. Computer-readable storage medium (31) on which the computer program product according to claim 14 is stored.

Images