A Bluetooth auscultation system and method based on bone conduction

Abstract

This invention relates to a Bluetooth auscultation system and method based on bone conduction. The system includes at least a stethoscope and a bone conduction sound transmitter. The stethoscope includes at least a sound processing unit; the sound processing unit detects environmental noise in the sound information received by the receiving unit, and filters the sound information and compensates for the sounds of the patient's organs based on stored relevant environmental features. This invention uses a separate stethoscope and bone conduction sound transmitter to perform auscultation without contact with the patient's limbs, preventing cross-infection that may occur in crowded areas. The stethoscope can perform noise reduction processing on the sound, preserving the sounds of the patient's organs, thus improving the efficiency and effectiveness of auscultation. This invention also uses a bone conduction sound transmitter so that medical personnel can hear the sound signal detected by the stethoscope in real time, and without blocking the medical personnel's ears, it allows medical personnel to perform auscultation and communicate with the patient simultaneously.

Description

Technical Field

[0001] This invention relates to the field of auscultation technology, and more particularly to a Bluetooth auscultation system and method based on bone conduction. Background Technology

[0002] Medical staff typically use stethoscopes with a tube and a stethoscope tube design. When wearing protective clothing, they cannot insert the stethoscope into their ears, making auscultation impossible. Furthermore, frequent use of stethoscopes greatly increases the risk of cross-contamination. Even when medical staff use the stethoscope normally, the ear-blocking method prevents them from having real-time conversations with patients, such as guiding them to take deep breaths. Therefore, there is a need for an auscultation system that allows for auscultation without physical contact with the patient and enables free communication with the patient.

[0003] Chinese patent CN111481229B discloses a bone conduction stethoscope, comprising a stethoscope head, a sound tube, and a bone conduction headband connected in sequence. The bone conduction headband includes a head fixation component and a connecting conduction component. The head fixation component includes a head fixing mechanism and a bone conduction mechanism disposed on the head fixing mechanism. The head fixing mechanism is used to fix the stethoscope to the head, and one end of the bone conduction mechanism is used to fit against the skin on the outside of the jawbone or skull. The connecting conduction component is connected to the sound tube and is snapped onto the bone conduction mechanism. The connecting conduction component and the bone conduction mechanism realize solid-state sound conduction for auscultation. The bone conduction stethoscope provided by this patent allows the head fixation component to be pre-fixed to the head, and the connecting conduction component to be snapped onto the head fixation component. This means that it only needs to be fixed once a day, and the connecting conduction component can be snapped onto the head before each use, making it convenient to wear and use. However, this patent does not allow medical personnel to maintain a safe social distance from patients.

[0004] Chinese patent CN112587166A discloses a wireless bone conduction auscultation system that prevents cross-infection and stores auscultation sounds as electronic data. It includes a wireless charging base for wirelessly charging a microphone module and / or earphones; a microphone module for picking up sound from the patient's auscultation site and transmitting the picked-up audio to the wireless charging base and earphones; and earphones for receiving audio signals from one of the microphone module, wireless charging base, or server, converting the signals into sound, and playing it through a speaker for the user to hear. This patent completely separates the sterile area (earphones) from the relatively contaminated area (microphone module), and the outer shell is made of waterproof and antibacterial material for easy disinfection. The bone conduction stethoscope is located outside the ear, eliminating the need for repeated removal and replacement during multiple auscultations, protecting the doctor's ear canal and reducing the risk of infection. The patent employs a catheter-free design, reducing the risk of cross-infection for patients, and the wireless design facilitates doctor-patient interaction. However, this patent does not eliminate ambient noise, resulting in significant background noise in the microphone, affecting the auscultation effect, and the simultaneous transmission to multiple earphones impacts the bandwidth of medical institutions. The auscultation area was not clearly displayed, and the location was not determined using a microphone, making it impossible for others to make an accurate diagnosis even if they heard the sound. Furthermore, the auscultation records did not store corresponding patient vital signs, resulting in poor usability and hindering learning and research.

[0005] Furthermore, on the one hand, there are differences in understanding among those skilled in the art; on the other hand, the inventors studied a large number of documents and patents when making this invention, but due to space limitations, not all details and contents were listed in detail. However, this does not mean that the present invention does not possess the features of these prior art. On the contrary, the present invention already possesses all the features of the prior art, and the applicant reserves the right to add relevant prior art to the background art. Summary of the Invention

[0006] In existing technologies, the accuracy of the stethoscope's position in the auscultation zone cannot be guaranteed due to patients having different body shapes during contactless auscultation. Furthermore, in the event of a sudden large-scale infectious disease outbreak, existing stethoscopes cannot achieve high-quality auscultation under the premise of high-frequency use and avoiding cross-infection. Based on these shortcomings, this invention aims to provide a Bluetooth auscultation system based on bone conduction.

[0007] To address the shortcomings of existing technologies, the present invention provides a Bluetooth auscultation system based on bone conduction, comprising at least: a stethoscope and a bone conduction sound transmitter communicatively connected to the stethoscope. The stethoscope includes at least a receiving unit, a sound processing unit, and a transmitting unit. The sound processing unit detects environmental noise in the sound information received by the receiving unit and filters the sound information and compensates for the sounds of the patient's organs based on stored relevant environmental features, enabling the sound processing unit to output a processed sound signal. The system further includes a server that, based on point positioning and trajectory positioning of the stethoscope, obtains a human body model conforming to the patient's auscultation area and body shape and sends it to a display unit. The present invention uses a separate stethoscope and bone conduction sound transmitter to perform auscultation without contact with the patient's limbs, preventing cross-infection that may occur in crowded areas. The stethoscope can perform noise reduction processing on the sound while preserving the sounds of the patient's organs, thus improving the efficiency and effectiveness of auscultation. This invention also employs a bone conduction sound transmitter, enabling medical personnel to hear the sound signals detected by the stethoscope in real time. This method, which does not block the medical personnel's ears, allows them to perform auscultation while simultaneously conversing with patients. The bone conduction sound transmitter also solves the problem of medical personnel being unable to place the stethoscope earpiece in their ears while wearing protective clothing.

[0008] According to a preferred embodiment, the point positioning refers to the auscultation area of ​​the stethoscope on the patient's body surface during auscultation, and the trajectory positioning refers to the movement trajectory of the stethoscope during auscultation. The point positioning is obtained by a positioning unit installed in the stethoscope, or by an image recognition unit and / or at least two infrared probes installed in the telescopic rod. Using infrared probes to identify the patient's contour to determine the stethoscope's position on the patient's body surface is more accurate than the method of visual identification of the stethoscope by medical personnel in existing non-contact auscultation techniques. Non-contact auscultation prevents medical personnel from accurately locating the auscultation area through physical contact, and the different body shapes of patients make it difficult for medical personnel to visually determine whether the stethoscope is in the correct auscultation area. To address this, the present invention constructs the patient's body contour using infrared probes, displaying the patient's human body model on a display unit or other terminal device, thereby accurately positioning the stethoscope and confirming whether it is in the correct auscultation area, achieving high-efficiency and high-quality auscultation.

[0009] According to a preferred embodiment, the positioning unit is disposed in the stethoscope in such a way that it can store the current position information of the stethoscope on the patient's body. The storage medium of the positioning unit stores the position information and the sound signal in correspondence, so that the stethoscope can obtain auscultation information that can characterize the patient's auscultation area and body shape by traversing the patient's body surface with its detection points.

[0010] According to a preferred embodiment, the auscultation information includes at least the wavelength, intensity, interval, and background noise of the sound received by the stethoscope, as well as the position of the stethoscope on the patient's body surface. The wavelength, intensity, interval, and background noise can be represented by waveforms and / or spectrograms. The position information is represented by its location on a human body model.

[0011] According to a preferred embodiment, the stethoscope sends the auscultation information to the server, the server sends the sound signal as a waveform and / or spectrogram to the display unit, and the server displays the auscultation area in a highlighted manner on the human body model and sends it to the display unit, so that the sound signal and position information are displayed in correspondence. Through these preferred embodiments of the present invention, the sound signal and position information are combined and displayed in waveform, spectrogram, and / or human body model form, and at least in the display unit. The server records and stores this information during and after the auscultation process performed by medical personnel using the stethoscope, thereby providing an intuitive, low-bandwidth, and low-computing-cost display. This is particularly beneficial for medical personnel, experts, or superiors to remotely retrieve and view the status of the auscultation process during and afterward using a tablet or smartphone from the server, enabling real-time information sharing.

[0012] According to a preferred embodiment, the stethoscope's sound processing modes include at least: a heartbeat mode that emphasizes the heartbeat and weakens the lungs; a lung mode that emphasizes the lungs and weakens the heartbeat; and an extended mode that includes all sound components after filtering. Since the heartbeat and lungs have different auscultation areas and different auscultation sounds, the stethoscope can enhance the targeted auscultation effect by strengthening one while weakening the other, while still retaining the comprehensiveness of all sound components. An extended mode is provided to be applicable to all auscultation areas.

[0013] According to a preferred embodiment, the extended mode is capable of acquiring sound components with frequencies in the range of 20 to 2000 Hz.

[0014] According to a preferred embodiment, the transmitting unit is capable of transmitting the sound signal processed by the sound processing unit to the bone conduction sound transmitter; upon receiving the sound signal, the bone conduction sound transmitter converts the sound signal into a bone conduction signal. The bone conduction sound transmitter includes at least: a transducer vibrator and an attachment surface in contact with the user's skin tissue; the transducer vibrator converts the sound signal into a bone conduction signal and generates vibration based on the bone conduction signal, causing the attachment surface connected to the transducer to generate a resonance peak.

[0015] This invention also relates to a Bluetooth auscultation method based on bone conduction, the method comprising at least: a receiving unit receiving sound information; a sound processing unit detecting environmental noise in the sound information received by the receiving unit, and filtering the sound information and compensating for the sounds of the patient's organs according to stored relevant environmental features, so that the sound processing unit can output a sound signal that has completed sound processing; and a server obtaining a human body model that conforms to the patient's auscultation area and body shape based on point positioning and trajectory positioning of the stethoscope and sending it to a display unit.

[0016] According to a preferred embodiment, the point positioning refers to the auscultation area of ​​the stethoscope on the patient's body surface during auscultation, and the trajectory positioning refers to the movement trajectory of the stethoscope during auscultation; wherein, the point positioning is obtained by a positioning unit disposed in the stethoscope, or the point positioning is obtained by an image recognition unit and / or at least two infrared probes disposed in the telescopic rod.

[0017] Beneficial technical effects of the present invention:

[0018] This invention features a separate stethoscope and bone conduction sound transmitter, allowing for auscultation without contact with the patient's limbs. This prevents potential cross-infection during crowded situations. The stethoscope can perform noise reduction processing to preserve the sounds of the patient's organs, improving the efficiency and effectiveness of auscultation. The bone conduction sound transmitter allows medical personnel to hear the sound signals detected by the stethoscope in real time, and its method of not blocking the medical personnel's ears allows them to both perform auscultation and communicate with the patient simultaneously. The bone conduction sound transmitter also solves the problem of medical personnel being unable to place the stethoscope earpiece in their ears while wearing protective clothing. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating a preferred embodiment of a Bluetooth auscultation system based on bone conduction according to the present invention.

[0020] Figure 2 This is a schematic diagram of signal transmission of a preferred embodiment of a Bluetooth auscultation system based on bone conduction according to the present invention;

[0021] Figure 3 This is a schematic diagram of a preferred embodiment of the bone conduction sound generator of the present invention;

[0022] Figure 4 This is a cross-sectional view of the transducer and the attachment surface of the bone conduction sound generator of the present invention.

[0023] List of reference numerals

[0024] 1: Stethoscope; 2: Bone conduction sound generator; 3: Server; 4: Display unit; 101: Receiving unit; 102: Sound processing unit; 103: Transmitting unit; 104: Positioning unit; 201: Converter vibrator; 202: Attachment surface. Detailed Implementation

[0025] The following is a detailed explanation with reference to the accompanying drawings.

[0026] Example 1

[0027] This application relates to a Bluetooth auscultation system based on bone conduction, which enables auscultation by medical personnel without contact with the patient's limbs and while simultaneously hearing the patient's voice. It includes at least a stethoscope 1 and a bone conduction sound-emitting component 2 that is communicatively connected to the stethoscope 1.

[0028] according to Figure 1 and 2 In one specific embodiment, the stethoscope 1 includes at least a receiving unit 101, a sound processing unit 102, and a transmitting unit 103. The sound processing unit 102 detects environmental noise in the sound information received by the receiving unit 101, and filters the sound information and compensates for the sounds of the patient's organs based on stored relevant environmental features, enabling the sound processing unit 102 to output a processed sound signal. The system also includes a server 3, which obtains a human body model conforming to the patient's auscultation area and body shape based on point positioning and trajectory positioning of the stethoscope 1 and sends it to the display unit 4. This invention uses a separate stethoscope 1 and bone conduction sound transmitter 2 to perform auscultation without contact with the patient's limbs, preventing cross-infection that may occur in crowded areas. The stethoscope 1 can perform noise reduction processing on the sound while preserving the sounds of the patient's organs, improving the efficiency and effectiveness of auscultation. This invention also uses the bone conduction sound transmitter 2 to allow medical personnel to hear the sound signals detected by the stethoscope 1 in real time without blocking their ears, allowing them to perform auscultation and communicate with the patient simultaneously. The bone conduction sound transmitter 2 also solves the problem that medical staff cannot put the stethoscope earpiece into their ears after wearing protective clothing.

[0029] According to a preferred embodiment, the stethoscope 1 enables wireless auscultation via Bluetooth (or other wireless connection methods, such as 2.4G) with the bone conduction sound transmitter 2, allowing for contactless auscultation when the patient has an infectious disease, thus improving the safety of medical personnel. The stethoscope 1 processes the collected sound information and sends it to the bone conduction sound transmitter 2. The stethoscope 1 can be wirelessly controlled by medical personnel. The stethoscope 1 can be a flattened cylinder, with the side that contacts the patient's skin slightly concave to ensure a close fit. The stethoscope 1 may be equipped with a telescopic rod. Medical personnel fix the stethoscope 1 to one end of the telescopic rod and control the contact position between the stethoscope 1 and the patient by holding the handle at the other end of the rod, giving medical personnel greater freedom in auscultation distance while maintaining a safe social distance. To avoid multiple people using the stethoscope 1, a disposable pouch can be provided, allowing the stethoscope 1 to be placed inside for auscultation. The disposable pouch can be made of plastic. The noise from plastic bags can be eliminated by pre-storing the relevant plastic bag noise, or by using a built-in sensitive sensor to ensure that the auscultation effect is not interfered with by the plastic bag noise.

[0030] Existing stethoscopes, with their complex structure consisting of a stethoscope and tubing, are difficult to disinfect and clean. Frequent use of the stethoscope can easily lead to cross-contamination. The stethoscope 1 of this invention, however, is designed for easy disinfection. Its metal casing allows for cleaning with alcohol wipes, and its waterproof design allows for disinfection by immersing it in an alcohol solution, making it convenient, quick, and highly practical.

[0031] According to a preferred embodiment, the stethoscope 1 further includes a positioning unit 104. The positioning unit 104 is configured within the stethoscope 1 to store information about the current location of the stethoscope 1 on the patient's body. The positioning unit 104 sends the location information to a server 3. The server 3 stores the received location information in correspondence with sound signals, enabling the stethoscope 1 to acquire auscultation information representing the patient's complete physical condition by traversing the patient's body surface with its detection points. The location information stored in the positioning unit 104 can be used for point positioning by a doctor, and the positioning unit 104 has a built-in gyroscope for trajectory positioning. The positioning unit 104 includes a gyroscope, a storage medium, and a transmission module. The storage medium stores the current location information of the stethoscope 1 on the patient's body. The transmission module sends this location information to the server 3.

[0032] According to a preferred embodiment, at least two infrared probes can be installed on the telescopic rod to obtain the contour of the patient's examination area, thereby accurately determining the position information of the stethoscope 1. The positioning unit 104 does not perform point positioning work, but only positions the trajectory of the stethoscope 1. This invention performs point positioning by adding at least two infrared probes, without requiring medical personnel's confirmation as the standard for establishing the human body model, preventing inaccurate point positioning due to medical personnel's misoperation. The infrared probes are respectively placed on both sides of the telescopic rod and aligned with the direction in which the stethoscope 1 is fixed on the telescopic rod. Preferably, the infrared probes are embedded in the telescopic rod and perform thermal radiation monitoring of the position of the stethoscope 1 along the telescopic rod. Preferably, the infrared probes extend from the telescopic rod to perform thermal radiation monitoring of the position of the stethoscope 1 from multiple angles.

[0033] Preferably, the location of the stethoscope 1 can also be determined through image recognition. The positioning unit 104 of the stethoscope 1 can be connected to the image recognition unit. The image recognition unit is, for example, the camera of a medical professional's smartphone or a camera placed independently on the stethoscope table. The medical professional can hold a telescopic rod and a smartphone, and by opening the smartphone's camera, perform image recognition on the real-time position of the stethoscope 1 to form a human body model. If the medical professional cannot temporarily input images through the smartphone's camera, they can place the patient in the image using a camera (camera) placed on the stethoscope table, which is also used for image input, thereby achieving image input. Both the camera and the stethoscope 104 serve as terminal devices. The positioning unit 104 has a built-in gyroscope and storage medium to correspond the trajectory positioning it detects with the point positioning detected by the image recognition unit, thereby obtaining an accurate human body model.

[0034] According to a preferred embodiment, the auscultation information includes at least the wavelength, intensity, interval, and background noise of the sound received by the stethoscope 1, and the position of the stethoscope 1 on the patient's body surface. The wavelength, intensity, interval, and background noise can be represented by waveforms and / or spectrograms. The position information is represented by the position on a human body model.

[0035] According to a preferred embodiment, the stethoscope 1 sends auscultation information to the server 3, and the server 3 displays the sound signal on the display unit 4 in the form of waveform and / or spectrum, and displays the position information on the display unit 4 by highlighting the corresponding position on the human body model, so that the sound signal and position information are displayed in correspondence and visually shared.

[0036] According to a preferred embodiment, the bone conduction sound generator 2 includes at least: a transducer 201 and an attachment surface 202 in contact with the user's skin tissue. The transducer 201 converts sound signals into bone conduction signals and generates vibrations based on these signals, causing the attachment surface 202 connected to the transducer 201 to generate resonance peaks. The aforementioned bone conduction sound generator 2 can be a form of bone conduction application such as bone conduction headphones or bone conduction speakers. In fact, various preferred embodiments of the present invention can be applied to a variety of hearing devices, and are not limited to bone conduction devices. For example, those skilled in the art, after understanding the design principles and usage of the stethoscope 1, can modify and improve the details and implementation methods of the bone conduction sound generator 2 without departing from this principle, including one or more combinations of noise reduction, gain, tinnitus reduction, feedback processing, and volume control. Figure 4 As shown, the principle of the bone conduction sound transmitter 2 is as follows: the bone conduction sound transmitter 2 receives sound signals and, under the action of the transducer 201, converts the sound signals into bone conduction signals; the transducer 201 vibrates according to the bone conduction signals; the vibration of the transducer 201 causes the attachment surface 202 to vibrate, causing the attachment surface 202 to generate resonant peaks, thereby transmitting them to the user's sensory organs. The bone conduction sound transmitter 2 is configured in various shapes that can be placed behind the user's ear, for example, as... Figure 3 As shown. The vibration of the attachment surface 202 is transmitted to the auditory nerve through tissue and bone, allowing the user to hear sound. The attachment surface 202 can be in direct contact with the user's skin or in contact with a transmission layer. The transmission layer includes, but is not limited to, low-density materials such as rubber and plastic. The transducer 201 converts the bone conduction signal (electrical signal) into mechanical vibration, thereby driving the attachment surface 202 to vibrate. The bone conduction sound generator 2 also includes a housing. The housing encloses the transducer 201 and the attachment surface 202. An elastic element can be provided between the transducer 201 and the housing to reduce the vibration amplitude of the housing and reduce sound leakage when the transducer 201, the attachment surface 202, and / or the housing vibrate. The elastic element can be made of materials such as plastic and stainless steel. A certain number of sound-guiding holes can be opened on the housing to guide the sound wave vibration inside the housing to the outside of the housing during the vibration of the transducer 201, interacting with the sound waves leaking out of the housing to suppress the sound leakage of the bone conduction sound generator 2. The housing can also be made of sound-absorbing material.

[0037] Example 2

[0038] This embodiment is a further step and / or supplement to the above embodiments, and repeated content will not be described again. The above system is capable of performing the methods described below.

[0039] According to a preferred embodiment, the filtering process of the sound processing unit 102 includes at least: establishing an acoustic model; optimizing the acoustic model; and filtering the sound information based on the output value of the acoustic model. Specifically, the acoustic model is modeled using a Hidden Markov Model (HMM) and a Gaussian mixture model is established using a Gaussian distribution. The sound information and initial parameter values ​​represent the covariance matrix obtained after calculating the sound information of the target. The acoustic model is optimized based on the maximum likelihood criterion. Under the basic principle of maximum likelihood estimation, the maximum expectation algorithm is used, and through continuous iterative calculation, the final value satisfies the convergence condition and gradually approaches the true parameters.

[0040] 1. Establish an acoustic model

[0041] The acoustic model can be modeled using a Hidden Markov Model (HMM). An HMM is essentially a discrete-time finite-state automaton. The internal states of this HMM are not visible to the outside world; only the output values ​​at each time step are visible. For stethoscope 1, the output values ​​are typically the acoustic features calculated from each frame. These acoustic features are then filtered based on stored relevant environmental features.

[0042] Specifically, the input to the acoustic model is the feature extracted from the sound information, which is a multi-dimensional vector. Since features with continuously taking values ​​(the features extracted from the sound information) should be represented by a continuous probability distribution, a Gaussian distribution (or others) can be used to directly describe this. Therefore, a Gaussian mixture model (or Laplace mixture model) is established to fit the distribution of the sound signal. The Gaussian mixture distribution can be viewed as a weighted combination of several Gaussian components, and its specific formula can be expressed as:

[0043]

[0044] Where x represents sound information; N(x|μ k ∑k) represents the k-th component in the Gaussian mixture model; Q k Let k = 1, 2, ..., M, representing the weights of each function component. In the above formula, each function component follows a Gaussian distribution, and its initial value represents the covariance matrix obtained after calculating the sound information of the target. Gaussian mixture models suffer from a training problem, namely, the problem of estimating the number of parameters to achieve the optimal match between the Gaussian mixture model and the training data. This requires further processing by a filtering unit.

[0045] 2. Optimization of the acoustic model based on the maximum likelihood criterion

[0046] The basic principle of maximum likelihood estimation is to derive the parameters P of the model, such that the Gaussian mixture model achieves maximum likelihood for the training corpus, where the training corpus refers to audio information. To this end, the Baum-Welch algorithm is used for parameter estimation, i.e., the expectation-maximization algorithm is employed.

[0047] The Expectation-Maximization (EME) algorithm is a method for finding the maximum likely estimate of parameters. Starting with incomplete datasets, it performs maximum likelihood estimation (MLE) on the parameters and is widely used to process missing or noisy incomplete data. Essentially, it iteratively calculates the parameters until the final value meets the convergence condition and gradually approximates the true parameters. In this invention, the EEM algorithm finds the maximum likely estimate of parameters in a Gaussian mixture model, based on two iterative steps: the expectation step and the maximization step. The expectation step calculates the expectation of the hidden variables using the existing estimates of the Gaussian mixture model (or the parameter estimates from the previous iteration). The maximization step uses the expectations of the hidden variables to perform a maximum likely estimate of the Gaussian mixture model and obtain the parameter estimates. Specifically, for example, there are two parameters x and y, whose initial states are unknown, but the relationship between x and y is known; that is, information about x can be obtained to determine information about y, and vice versa. The expectation-maximization algorithm assigns an initial value to x (or randomly selects a set of parameters) to obtain an estimate of y. Based on the estimate of y, it re-estimates the value of x. Through alternating iterative calculations, the expectation-maximization algorithm gradually improves the parameters of the model, making the likelihood probability of the parameters and training samples gradually increase, and finally terminates at a maximum point.

[0048] The Expectation-Maximization (EM) algorithm provides a simple iterative algorithm for calculating the posterior density function. Its advantages lie in its simplicity and stability; however, it is prone to getting trapped in local optima. A local optimum occurs when, due to the combined effects of multiple factors, different segments of the sound information cannot simultaneously reach their optimal levels, leading to iterative computation until the optimal result of a particular segment is found. To address this, this solution empirically sets the number of iterations. When the EEM algorithm reaches the preset number of iterations, the computation stops, and the output value of the acoustic model is obtained.

[0049] 3. Filtering of sound information based on the output value of the acoustic model.

[0050] There are many different filtering methods, and no specific method is specified here.

[0051] The process of filtering sound information through the above-mentioned filtering unit can effectively eliminate or suppress ambient noise interference.

[0052] In this invention, the stored environmental features include noises such as clothing rubbing, chair movement, computer keyboard typing, computer notification sounds, and medical institution broadcasts that can interfere with the auscultation process of medical personnel. These sounds all have characteristics distinct from auscultation sounds. For example, computer keyboard typing, computer notification sounds, and medical institution broadcasts all have specific characteristics in terms of volume, waveform, frequency, and / or tone; that is, due to differences in the materials and structures of the sound-producing bodies, they each possess their own unique sound characteristics. Pre-storing these sound features and using them to remove the environmental noise received by the stethoscope 1 is a crucial technical measure.

[0053] According to a preferred embodiment, point positioning refers to the auscultation area of ​​the stethoscope 1 on the patient's body surface during auscultation, and trajectory positioning refers to the movement trajectory of the stethoscope 1 during auscultation; wherein, point positioning is obtained by the positioning unit 104 disposed in the stethoscope 1, or point positioning is obtained by the image recognition unit and / or at least two infrared probes disposed in the telescopic rod.

[0054] According to a preferred embodiment, a positioning unit 104 is installed in the stethoscope 1 to store the current position information of the stethoscope 1 on the patient's body. The positioning unit 104 sends the position information to the server 3. The server 3 stores the received position information in correspondence with the sound signal, so that the stethoscope 1 can acquire auscultation information that characterizes the patient's auscultation area and body shape by traversing the patient's body surface with its detection points. The position information stored in the positioning unit 104 can be used by the doctor for point positioning, and the positioning unit 104 has a built-in gyroscope for trajectory positioning. The positioning unit 104 includes a gyroscope, a storage medium, and a transmission module. The storage medium is used to store the current position information of the stethoscope 1 on the patient's body. The transmission module sends the position information to the server 3. When point positioning and trajectory positioning are performed by the positioning unit 104, the medical staff first determines the point positioning of the stethoscope 1, and then the trajectory positioning connects the point positioning to form a human body model. Specifically, the auscultation sites for heart sounds are typically located in the following order: mitral valve auscultation area, at the point of strongest apical impulse, i.e., the junction of the left midclavicular line and the fifth rib; pulmonary valve auscultation area, at the second intercostal space along the left sternal border; aortic valve auscultation area, at the second intercostal space along the right sternal border; the second aortic valve auscultation area, at the third intercostal space along the left sternal border; and tricuspid valve auscultation area, at the lower left sternal border, i.e., the fourth to fifth intercostal spaces along the left sternal border. By auscultating these areas sequentially, the condition of each valve, including valvular stenosis and valvular insufficiency, can be assessed. When the stethoscope 1 reaches a specific auscultation area, the display unit 4 confirms the location, allowing the positioning unit 104 to record this location information. This allows for the acquisition of auscultation information characterizing the patient's auscultation areas and body type by traversing the patient's body surface. The lung auscultation area is much larger. After the stethoscope 1 reaches the auscultation area, medical personnel can click on the corresponding point on the basic human body model through the display unit 4 or other terminal devices to record the real-time position of the stethoscope 1, thus forming the corresponding auscultation information. The trajectory of the medical personnel moving the stethoscope 1 is located by the gyroscope of the positioning unit 104. After the medical personnel determine the first point, they begin recording the movement trajectory of the stethoscope 1, and then connect the second point and so on until the last point, thus forming a complete human body model containing all the patient's points. This human body model can not only reflect the auscultation information, but also reflect the patient's physical characteristics, such as the patient's body shape. This human body model serves as the basic image transmitted by the server 3.

[0055] According to a preferred embodiment, at least two infrared probes can be installed on the telescopic rod to obtain the contour of the patient's examination area, thereby accurately determining the position information of the stethoscope 1. Point positioning is achieved through the addition of at least two infrared probes, eliminating the need for medical personnel confirmation as the standard for establishing the human body model, thus preventing inaccurate point positioning due to medical personnel's misoperation. The medical personnel's point positioning on the display unit 4 can serve as a calibration for the infrared probe positioning, rather than a decisive factor. At least two infrared probes can detect the contour of the patient's area where the stethoscope 1 is located in the horizontal direction of the telescopic rod (i.e., the horizontal direction relative to the ground) by aligning it with the position of the stethoscope 1. This involves collecting thermal radiation signals from the patient's area using at least two infrared probes and transmitting these signals to the server 3 to generate a contour image. The stethoscope 1 can then be positioned on this contour image. Specifically, during the collection of thermal radiation signals, the at least two infrared probes obtain first and second thermal radiation signals from different perspectives of the patient on the horizontal plane at left and right angles. After the first and second thermal radiation signals are sent to the server 3, a first infrared image and a second infrared image are obtained. Preferably, the first and second infrared images can be registered using multiple data points, thereby fusing and integrating the first and second infrared images, and performing point cloud data network fusion. Specifically, the two-dimensional coordinates of the first and second infrared images are transformed into three-dimensional coordinates using matrices, homogeneous coordinates, and least-squares methods to construct a contour image and even a human body model. The contour image and / or human body model established by the above method has high accuracy, accurately determining the position information of the stethoscope 1 and performing point localization. The arrangement of at least two infrared probes avoids the influence of factors such as unstable handheld operation by medical personnel, and the infrared monitoring method does not involve contact with the patient, eliminating the need for confirmation by medical personnel through the display unit 4, thus shortening the auscultation time. The position information of the stethoscope 1 can be directly displayed in real time on the display unit 4 or other display devices (e.g., smartphones). Medical personnel can calibrate this position information through the display unit 4 or other display devices (e.g., smartphones). This calibration refers to further confirmation of the position information by medical personnel. Due to the high accuracy of infrared monitoring, this calibration can also be omitted. Preferably, three infrared probes can be used to obtain the contour of the patient's detected area. The third infrared probe can be positioned between the first two. For example, when monitoring a patient's back, the first two infrared probes can determine the width and contour of the patient's back. If the patient is large and their height exceeds the monitoring range of the infrared probes, the vertical monitoring of the back may not meet the required accuracy. The third infrared probe can monitor the extended height, enabling the device to monitor the back contour of patients of different body types, increasing the accuracy of the contour image and / or the human model.The aforementioned infrared probe can be activated and deactivated via a switch located on the handle of the telescopic rod, accessible to medical personnel. It is understood that the addition of multiple infrared probes to monitor the patient's complete body contours, and any modifications or improvements to the details and implementation methods of the infrared probes, should fall within the scope of protection of this invention.

[0056] This invention uses an infrared probe to identify the patient's contours to determine the location of the stethoscope 1 on the patient's body surface. Compared to existing contactless auscultation techniques, which rely on medical personnel visually identifying the stethoscope's position, this method is more accurate. Contactless auscultation prevents medical personnel from accurately locating the auscultation area through physical contact, and the varying body shapes of patients make it difficult for medical personnel to visually determine whether the stethoscope is in the correct auscultation area. To address this, the present invention uses an infrared probe to construct the patient's body contours, which are then displayed on the display unit 4 or other terminal devices as a human body model of the current patient. This allows for precise point positioning of the stethoscope 1, confirming whether the stethoscope is in the correct auscultation area, thus achieving high-efficiency and high-quality auscultation.

[0057] Preferably, the system further includes an image recognition unit. The image recognition unit determines the location of the stethoscope 1 using image recognition. The positioning unit 104 of the stethoscope 1 can be connected to the image recognition unit, so that the point positioning obtained by the image recognition unit corresponds to the trajectory positioning obtained by the positioning unit 104. The image recognition unit can be, for example, the camera of a medical staff member's smartphone or a camera placed independently on the stethoscope table. The medical staff member can hold the telescopic rod and the smartphone. When the stethoscope 1 on the telescopic rod is located on the patient's body surface, the image is obtained by opening the smartphone camera, and the real-time position of the stethoscope 1 is image-recognized to form a human body model. This image can be obtained from a photograph or a frame of a video. If the medical staff member cannot temporarily input images using the smartphone camera (i.e., it is inconvenient to operate the telescopic rod and the smartphone with both hands at the same time), the patient can be placed in the image using a camera (camera) placed on the stethoscope table, which is also used for image input, thereby realizing image input. Both the camera and the video camera serve as terminal devices. There are various existing technologies that can achieve image recognition. For example, Person Re-identification (ReID) technology can be used to identify people in images captured by cameras and / or video cameras, thereby obtaining a human body model corresponding to the patient's body shape. Another example is obtaining point cloud data of the surrounding environment through a camera. Based on distance and normal vector information, the point cloud data is segmented into multiple blocks, and point cloud blocks that conform to the length, width, and curves of a human body are selected. HOD features are then extracted to obtain a human body model. After obtaining the human body model corresponding to the patient using the above methods, the positioning unit 104 can obtain the point positioning of the stethoscope 1, thereby achieving accurate auscultation of multiple locations on the patient. Infrared probes can also obtain accurate point positioning; however, the drawback of infrared probes is their high cost. If all telescopic rods are equipped with infrared probes, it will increase manufacturing costs, and the infrared probes need to be charged at high frequencies to ensure that the obtained patient contour and human body model remain within the allowable error range. The image recognition unit can use the smartphone carried by medical personnel to take videos or photos, and send the obtained images to the server 3 for image recognition to obtain the human body model. This solution enables auscultation to be performed in various complex environments, such as temporary medical stations. It requires minimal investment, as a large portion of the image recognition work is done on server 3, with smartphones or cameras serving only as image acquisition tools, making it widely applicable in auscultation procedures within medical institutions.

[0058] According to a preferred embodiment, the auscultation information includes at least the wavelength, intensity, interval, and background noise of the sound received by the stethoscope 1, and the position of the stethoscope 1 on the patient's body surface. The wavelength, intensity, interval, and background noise can be represented by waveforms and / or spectrograms. The position information is represented by the position on a human body model.

[0059] According to a preferred embodiment, the stethoscope 1 sends auscultation information to the server 3, and the server 3 displays the sound signal on the display unit 4 in the form of a waveform and / or spectrum. The server 3 displays the auscultation area in a highlighted manner on the human body model and sends it to the display unit 4 so that the sound signal is displayed in correspondence with the position information.

[0060] Through these preferred embodiments of the present invention, sound signals and location information are combined and displayed, at least in the display unit 4, in the form of waveforms, spectrograms, and / or human body models. The server 3 records and stores this information during and after the auscultation process performed by medical personnel using the stethoscope 1, thereby providing an intuitive, low-bandwidth, and low-computing-cost display. This is particularly beneficial for medical personnel, experts, or superiors to remotely access and view the auscultation process and its aftermath using devices such as tablets or smartphones from the server 3, enabling real-time information sharing. During auscultation, when encountering situations where the patient's condition is difficult to determine, the ability to transmit auscultation information requiring expert or superior decision-making in an intuitive manner is a crucial technical measure. In the present invention, while the display unit 4 provides an intuitive and replayable display to the on-site medical personnel in the aforementioned situations, it is often necessary for the medical personnel to report to experts or superiors to determine the patient's condition, which is difficult to assess. Since verbal descriptions cannot intuitively explain the auscultation content, it is difficult for experts or superiors to provide more scientific decisions remotely on-site. Using the aforementioned design scheme of the present invention, sound signals and location information are combined in the form of waveforms, spectrum diagrams, and / or human body models and displayed at least in the display unit 4. The server 3 can respond to a receiving request, such as from a smartphone, by transmitting the combined image and audio to the smartphone. The image can display auscultation information in a one-to-one correspondence manner. For example, the server 3 provides a human body model with all points and trajectories of the stethoscope 1 during auscultation. Initially, the receiving end, such as the smartphone, only receives the image, rather than all data including audio, thereby reducing data communication volume. This reduction in computational load leads to a geometric reduction in parameter calculations. Receiving only the image shortens data reception time, achieving faster real-time status display. Subsequently, the user clicks on points and / or trajectories in the image, and the corresponding audio data is then transmitted to the smartphone for auditory and visual display.

[0061] According to a preferred embodiment, the sound processing modes of the stethoscope 1 include at least: a heartbeat mode that emphasizes the heartbeat and weakens the lungs; a lung mode that emphasizes the lungs and weakens the heartbeat; and an extended mode that includes all sound components after filtering. Since the heartbeat and lungs have different auscultation areas and different auscultation sounds, the stethoscope can enhance the targeted auscultation effect by strengthening one while weakening the other, while still retaining the comprehensiveness of all sound components. An extended mode is provided to be applicable to all auscultation areas.

[0062] According to a preferred embodiment, the extended mode is capable of capturing sound components with frequencies between 20 and 2000 Hz.

[0063] Throughout the text, the features indicated by “preferred” are only optional and should not be construed as mandatory. Therefore, the applicant reserves the right to abandon or delete the relevant preferred features at any time.

[0064] It should be noted that the specific embodiments described above are exemplary, and those skilled in the art can devise various solutions inspired by the disclosure of this invention. These solutions all fall within the scope of this invention and its protection. Those skilled in the art should understand that this specification and its accompanying drawings are illustrative and not intended to limit the scope of the claims. The scope of protection of this invention is defined by the claims and their equivalents.

Claims

1. A Bluetooth auscultation system based on bone conduction, characterized in that, include: A separate stethoscope and a bone conduction sound transmitter that communicates with the stethoscope. A stethoscope includes a receiving unit, a sound processing unit, and a transmitting unit; The sound processing unit detects environmental noise in the sound information received by the receiving unit, and performs filtering processing on the sound information and compensation for the patient's organ sounds based on the stored relevant environmental features, so that the sound processing unit outputs a sound signal that has completed sound processing. The server detects the contour of the patient's body part where the stethoscope is located based on at least two infrared probes and generates a contour image. It performs point localization of the stethoscope on the contour image and matches the trajectory localization with the point localization detected by the image recognition unit. By connecting the trajectory localization with the point localization, the server obtains a human body model that conforms to the patient's auscultation area and body shape characteristics and includes all points of the patient, and sends it to the display unit. Point positioning refers to the auscultation area on the patient's body surface where the stethoscope is located during auscultation, while trajectory positioning refers to the movement trajectory of the stethoscope during auscultation. The positioning unit is installed in the stethoscope to store the current position information of the stethoscope on the patient's body, and sends the position information to the server. The server stores the received position information in correspondence with the sound signal, so that the stethoscope can obtain auscultation information that represents the patient's complete physical condition by traversing the patient's body surface with its detection points. The server highlights the auscultation area on the human body model and sends it to the display unit so that the sound signal is displayed in correspondence with the position information. Then, the sound signal and position information are combined in the form of waveform, spectrum and / or human body model and displayed in the display unit. In response to the smartphone's receiving request, the server sends the combined image and audio to the smartphone. The image displays the auscultation information in a one-to-one correspondence. The human body model provided by the server has all the points and trajectories of the stethoscope's position during the auscultation. In the initial reception, the smartphone only receives the image. Afterwards, the user clicks on the points and / or trajectories in the image, and the corresponding audio data is transmitted to the smartphone for auditory and visual display.

2. The Bluetooth auscultation system based on bone conduction as described in claim 1, characterized in that, The point positioning is obtained by the positioning unit (104) disposed in the stethoscope (1), or The point positioning is obtained by an image recognition unit and / or at least two infrared probes installed in the telescopic rod.

3. The Bluetooth auscultation system based on bone conduction as described in claim 2, characterized in that, The storage medium of the positioning unit (104) stores the location information and the sound signal in correspondence, so that the stethoscope (1) can obtain auscultation information that can characterize the patient's auscultation area and body shape by traversing the patient's body surface with its detection points.

4. The Bluetooth auscultation system based on bone conduction as described in claim 3, characterized in that, The auscultation information includes at least the wavelength, intensity, interval, and noise of the sound received by the stethoscope (1) and the location information of the stethoscope (1) on the patient's body surface.

5. The Bluetooth auscultation system based on bone conduction as described in claim 4, characterized in that, The stethoscope (1) sends the auscultation information to the server (3), and the server (3) sends the sound signal to the display unit (4) in the form of waveform and / or spectrum.

6. The Bluetooth auscultation system based on bone conduction as described in claim 5, characterized in that, The sound processing modes of the stethoscope (1) include at least: Emphasizing the heartbeat and weakening the lungs' heartbeat pattern; A lung pattern that emphasizes the lungs and weakens the heartbeat; and An extended mode that includes all sound components after filtering.

7. The Bluetooth auscultation system based on bone conduction as described in claim 6, characterized in that, The extended mode is capable of acquiring sound components with frequencies ranging from 20 to 2000 Hz.

8. The Bluetooth auscultation system based on bone conduction as described in claim 7, characterized in that, The transmitting unit (103) can transmit the sound signal processed by the sound processing unit (102) to the bone conduction sound transmitter (2); after receiving the sound signal, the bone conduction sound transmitter (2) converts the sound signal into a bone conduction signal; The bone conduction sound generator (2) includes at least a transducer (201) and an attachment surface (202). The transducer (201) converts the sound signal into a bone conduction signal and generates vibration based on the bone conduction signal, causing the attachment surface (202) connected to the transducer (201) to generate a resonance peak.

9. A Bluetooth auscultation method based on bone conduction, characterized in that, include: The receiving unit receives audio information; The sound processing unit detects environmental noise in the sound information received by the receiving unit, and performs filtering processing on the sound information and compensation for the patient's organ sounds based on the stored relevant environmental features, so that the sound processing unit outputs a sound signal that has completed sound processing. The server detects the contour of the patient's body part where the stethoscope is located based on at least two infrared probes and generates a contour image. It performs point localization of the stethoscope on the contour image and matches the trajectory localization with the point localization detected by the image recognition unit. By connecting the trajectory localization with the point localization, the server obtains a human body model that conforms to the patient's auscultation area and body shape characteristics and includes all points of the patient, and sends it to the display unit. Point positioning refers to the auscultation area on the patient's body surface where the stethoscope is located during auscultation, while trajectory positioning refers to the movement trajectory of the stethoscope during auscultation. The positioning unit is installed in the stethoscope to store the current position information of the stethoscope on the patient's body, and sends the position information to the server. The server stores the received position information in correspondence with the sound signal, so that the stethoscope can obtain auscultation information that represents the patient's complete physical condition by traversing the patient's body surface with its detection points. The server highlights the auscultation area on the human body model and sends it to the display unit so that the sound signal is displayed in correspondence with the position information. Then, the sound signal and position information are combined in the form of waveform, spectrum and / or human body model and displayed in the display unit. In response to the smartphone's receiving request, the server sends the combined image and audio to the smartphone. The image displays the auscultation information in a one-to-one correspondence. The human body model provided by the server has all the points and trajectories of the stethoscope's position during the auscultation. In the initial reception, the smartphone only receives the image. Afterwards, the user clicks on the points and / or trajectories in the image, and the corresponding audio data is transmitted to the smartphone for auditory and visual display.

10. The Bluetooth auscultation method based on bone conduction as described in claim 9, characterized in that, The point positioning is obtained by the positioning unit (104) disposed in the stethoscope (1), or The point positioning is obtained by an image recognition unit and / or at least two infrared probes installed in the telescopic rod.

Images