Clinical task execution method and device based on voice analysis and electronic equipment
Key information in the conversation between medical staff and patients is extracted through speech analysis technology, solving the problem of inaccurate information understanding, and improving the efficiency of clinical task execution and the quality of medical services.
Patent Information
- Application Number
- CN202411018582.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-29
- Publication Date
- 2024-12-06
AI Technical Summary
In a medical clinical setting, communication between medical staff and patients may lead to inaccurate information understanding and inefficient clinical task execution.
Using a speech analysis method, by obtaining doctor-patient dialogue voice, converting it into text information, identifying key medical information and patient symptom information, and matching it with the execution task library, clinical execution tasks are generated to improve efficiency.
It realizes automated extraction of key information in patient conversations, reduces manual recording of time and human errors by medical staff, improves the quality and efficiency of data collection, and improves the quality of medical services.
Smart Images

Figure CN119092067A_ABST
Abstract
Description
Technical Field
[0001] The present application belongs to the field of data processing technology, and specifically relates to a clinical task execution method, device and electronic equipment based on speech analysis. Background Art
[0002] In a medical clinical setting, guiding patients through various examination and treatment tasks is often time-consuming and complex, relying on effective communication between medical staff and patients.
[0003] Correctly understanding and processing the information provided by patients is crucial to ensure accurate execution of clinical tasks. However, existing communication methods may lead to inaccurate understanding of information and inefficient execution of clinical tasks. Summary of the invention
[0004] The purpose of this application is to provide a clinical task execution method, device and electronic equipment based on speech analysis to improve the task execution efficiency of medical staff.
[0005] According to a first aspect of an embodiment of the present application, a method for performing a clinical task based on speech analysis is provided, and the method may include:
[0006] Obtain the audio of the conversation between doctors and patients;
[0007] Convert conversational speech into text messages;
[0008] Identify key medical information and patient symptom information in text messages;
[0009] Match key medical information and patient symptom information with the execution task library to obtain clinical execution tasks;
[0010] Demonstrate clinical execution tasks to medical staff.
[0011] In some optional embodiments of the present application, obtaining the conversation voice between the doctor and the patient includes:
[0012] Using a cardioid microphone array to capture audio of conversations between doctors and patients;
[0013] The conversation audio is filtered and compressed to form conversation speech.
[0014] In some optional embodiments of the present application, filtering and compressing the conversation audio to form a conversation voice includes:
[0015] Use statistical filtering algorithm to reduce noise of conversation audio to form filtered audio;
[0016] Losslessly compress the filtered audio to form conversational speech.
[0017] In some optional embodiments of the present application, converting the conversation speech into text information includes:
[0018] Use the trained speech recognition model to convert conversational speech into text information.
[0019] In some optional embodiments of the present application, identifying key medical information and patient symptom information in text information includes:
[0020] Use the trained text analysis model to identify key medical information and patient symptom information in text information.
[0021] In some optional embodiments of the present application, the execution task library is constructed in the following manner:
[0022] Gather key medical information and patient symptom information from medical literature, clinical guidelines, and common medical questions;
[0023] For each key medical information and patient symptom information, create a question-answer pair;
[0024] The question part of the question-answer pair includes key medical information and patient symptom information, and the answer part includes the performance task.
[0025] In some optional embodiments of the present application, key medical information and patient symptom information are matched with the execution task library to obtain clinical execution tasks, specifically:
[0026] Match the corresponding question and answer pairs according to key medical information and patient symptom information;
[0027] The answer in the question-answer pair is called and output to obtain the clinical execution task.
[0028] According to a second aspect of an embodiment of the present application, a clinical task execution device based on speech analysis is provided, and the device may include:
[0029] An acquisition module is used to acquire the conversation voice between doctors and patients;
[0030] A conversion module, used to convert the conversation speech into text information;
[0031] A recognition module for identifying key medical information and patient symptom information in text information;
[0032] A matching module is used to match key medical information and patient symptom information with the execution task library to obtain clinical execution tasks;
[0033] The display module is used to display clinical execution tasks to medical staff.
[0034] According to a third aspect of an embodiment of the present application, an electronic device is provided, which may include:
[0035] processor;
[0036] a memory for storing processor-executable instructions;
[0037] Wherein, the processor is configured to execute instructions to implement the clinical task execution method based on speech analysis as shown in any one of the embodiments of the first aspect.
[0038] According to the fourth aspect of the embodiments of the present application, a storage medium is provided. When the instructions in the storage medium are executed by a processor of an information processing device or a server, the information processing device or the server implements a clinical task execution method based on speech analysis as shown in any one of the embodiments of the first aspect.
[0039] The above technical solution of the present application has the following beneficial technical effects:
[0040] The method of the embodiment of the present application converts the conversation speech into text information; then identifies the key medical information and patient symptom information in the text information; and matches the key medical information and patient symptom information with the execution task library to obtain the clinical execution task. It realizes the automatic extraction of key information in the patient conversation, such as symptoms, medical history, etc., reduces the time and possible human errors of manual recording by medical staff, and can more accurately identify and analyze important information in the patient's voice or text, improve the quality and efficiency of data collection, and medical staff can focus more on patient care itself, thereby improving the overall quality of medical services. BRIEF DESCRIPTION OF THE DRAWINGS
[0041] Figure 1 is a flowchart of a method for executing a clinical task based on speech analysis in an exemplary embodiment of the present application;
[0042] Figure 2 is a schematic structural diagram of a clinical task execution device based on speech analysis in an exemplary embodiment of the present application;
[0043] Figure 3 is a schematic structural diagram of an electronic device in an exemplary embodiment of the present application;
[0044] Figure 4 It is a schematic diagram of the hardware structure of an electronic device in an exemplary embodiment of the present application. DETAILED DESCRIPTION
[0045] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below in conjunction with specific implementations and with reference to the accompanying drawings. It should be understood that these descriptions are only exemplary and are not intended to limit the scope of the present application. In addition, in the following description, the description of known structures and technologies is omitted to avoid unnecessary confusion of the concepts of the present application.
[0046] The accompanying drawings show schematic diagrams of layer structures according to embodiments of the present application. These figures are not drawn to scale, and some details are magnified and some details may be omitted for the purpose of clarity. The shapes of various regions and layers shown in the figures and the relative sizes and positional relationships therebetween are only exemplary, and may deviate in practice due to manufacturing tolerances or technical limitations, and those skilled in the art may additionally design regions / layers with different shapes, sizes, and relative positions according to actual needs.
[0047] Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.
[0048] In the description of the present application, it should be noted that the terms "first", "second" and "third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.
[0049] In addition, the technical features involved in the different embodiments of the present application described below can be combined with each other as long as they do not conflict with each other.
[0050] The following, in conjunction with the accompanying drawings, describes in detail the clinical task execution method, device and electronic device based on speech analysis provided by the embodiments of the present application through specific embodiments and their application scenarios.
[0051] like Figure 1 As shown, in a first aspect of an embodiment of the present application, a clinical task execution method based on speech analysis is provided, and the method may include:
[0052] S110: Acquiring the conversation voice between the doctor and the patient;
[0053] S120: Converting the conversation voice into text information;
[0054] S130: Identify key medical information and patient symptom information in the text information;
[0055] S140: Matching key medical information and patient symptom information with the execution task library to obtain clinical execution tasks;
[0056] S150: Demonstrate clinical execution tasks to medical staff.
[0057] The method of this embodiment converts the conversation speech into text information; then identifies the key medical information and patient symptom information in the text information; and matches the key medical information and patient symptom information with the execution task library to obtain the clinical execution task. It realizes the automatic extraction of key information in the patient conversation, such as symptoms, medical history, etc., reduces the time and possible human errors of manual recording by medical staff, and can more accurately identify and analyze the important information in the patient's voice or text, improve the quality and efficiency of data collection, and medical staff can focus more on patient care itself, thereby improving the overall quality of medical services.
[0058] In some embodiments, obtaining the voice of the conversation between the doctor and the patient includes:
[0059] Using a cardioid microphone array to capture audio of conversations between doctors and patients;
[0060] The conversation audio is filtered and compressed to form conversation speech.
[0061] In the embodiment, the scenario in which the conversational speech is generated may be an outpatient conversation, where a doctor talks to a patient and asks the patient about his symptoms, treatment history, and response. For example, the patient may describe a recent increase in blood sugar.
[0062] Record these conversations in real time using a recording device, which can be a professional recording tool or the doctor's mobile device.
[0063] Specifically, the hardware configuration can use directional microphones and select cardioid directional microphones, which are designed to capture sounds from a specific direction while suppressing background noise from other directions. Direct the microphone toward the speaker (doctor and patient) to better capture the conversation. In addition, an array containing multiple microphones is used in combination with a beamforming algorithm (LCMV beamforming) to spatially filter sounds from specific sources. This can further enhance the sound signals from doctors and patients while reducing the impact of noise from other directions.
[0064] In addition, voice activity detection is required to distinguish between speech segments and non-speech segments (background noise). This helps reduce the recording of non-dialogue parts and reduces the complexity of subsequent processing; then, advanced denoising techniques such as Wiener filters are used to further clean up background noise.
[0065] The compression in this embodiment is for faster data transmission. Since the data processing capability of the front-end device is limited, more complex processing needs to be performed in the server, so the audio needs to be compressed. In addition, the QUIC protocol needs to be used, and the transmission of the recording file is transmitted through the QUIC protocol. Because QUIC provides a lower connection establishment time and a more efficient packet loss recovery mechanism than traditional TCP, it has built-in TLS encryption to ensure the security and reliability of data transmission. These features make QUIC very suitable for real-time or near real-time audio data transmission, such as medical conversation recording.
[0066] In addition, end-to-end encryption, key management and transmission efficiency optimization are also required.
[0067] In order to ensure the security of the data transmission process, the latest TLS1.3 protocol is used, combined with the EphemeralDiffie-Hellman key exchange mechanism, and a new encryption key is generated for each communication. This solution not only improves the security of the key, but also reduces the handshake time and speeds up data transmission due to the optimization of TLS1.3.
[0068] The NetAdapt system monitors network conditions and automatically adjusts transmission settings. Through a comprehensive assessment of the current network, including factors such as bandwidth utilization and latency, NetAdapt can adjust packet size and select the most suitable transmission bit rate in real time. In addition, NetAdapt can automatically select a downgrade strategy when network conditions are poor, such as switching from QUIC to TCP, to ensure transmission stability and continuity.
[0069] In some embodiments, filtering and compressing the conversation audio to form the conversation speech includes:
[0070] Use statistical filtering algorithm to reduce noise of conversation audio to form filtered audio;
[0071] Losslessly compress the filtered audio to form conversational speech.
[0072] In the filtering process of this embodiment, during the implementation process, the voice signal can be first frame-divided to divide the continuous voice signal into shorter time segments; Fast Fourier Transform (FFT) is applied to each frame to convert it into the frequency domain; Wiener filtering theory is used to estimate the noise power spectrum of each frame, and the power spectrum of the signal is modified accordingly; Inverse Fast Fourier Transform (IFFT) is applied to convert the processed frequency domain signal back to the time domain. This can effectively reduce background noise, improve the clarity of the voice signal, and facilitate subsequent voice recognition and analysis.
[0073] The filtering process can be achieved by the following formula:
[0074]
[0075] Where H(f) is the frequency response of the filter, S xx (f) is the power spectral density of the clean signal, S nn (f) is the power spectral density of the noise.
[0076] Aiming at the speech signal in medical environment, especially in noisy background, an adaptive factor is introduced to optimize the noise estimation, so as to improve the performance of the filter.
[0077] The optimized Wiener filter formula can be expressed as:
[0078]
[0079] Where H(f) is the frequency response of the filter, S xx (f) is the power spectral density of the clean signal, S nn (f) is the power spectral density of the noise, and α is an adaptive factor used to adjust the impact of the noise power spectrum.
[0080] In the optimized Wiener filter formula, the value of α can balance the relationship between noise suppression and signal preservation.
[0081] In an environment with a high noise level (such as a hospital emergency room or a busy clinic), stronger noise suppression is required. In this case, a larger value of α can be selected, α in the range of 1.5 to 2.0.
[0082] If it is important to preserve the details of the speech signal (for example, in a diagnostic conversation), a smaller value of α should be chosen to avoid oversuppressing the speech signal. In this case, α is between 1 and 1.5.
[0083] In addition, the RNNoise noise suppression library can be integrated on mobile devices to improve the quality of voice recordings. By carefully analyzing the audio stream, RNNoise can effectively separate clear voice signals from background noise. This is particularly useful for complex background sounds common in medical environments, ensuring that medical terms can be accurately recognized even under low signal-to-noise ratio conditions. Opus can be selected as an audio compression tool. It is an open source, efficient audio codec standard that supports a wide range of bitrate adjustments. Opus can flexibly switch between almost lossless sound quality and extremely high compression rates to adapt to different network conditions and storage requirements. For medical conversation recordings, this means that while ensuring that the recording is clear and understandable, the file size can be minimized and the transmission process can be accelerated.
[0084] In some embodiments, converting the conversation speech into text information includes:
[0085] Use the trained speech recognition model to convert conversational speech into text information.
[0086] The speech recognition model in this embodiment can adopt the DeepSpeech2 model. If it is a conversational speech conversion performed by the front end, the text information needs to be compressed here. Before sending it to the server, the text data can be compressed using an efficient text compression algorithm (GZIP). This can not only reduce the size of the transmitted data, reduce the network load, but also speed up the transmission speed. Using GZIP compression can significantly reduce the volume of text data and improve transmission efficiency. In addition, the text data can also be encrypted end-to-end to ensure the security of the data during transmission. Here, a strong encryption standard (AES-256) can be used in combination with a secure key exchange mechanism (Ephemeral Diffie-Hellman key exchange in the TLS1.3 protocol) to dynamically generate an encryption key for each data transmission. In this way, even if the data is intercepted during transmission, it cannot be decrypted without a key, thereby ensuring the confidentiality of the data.
[0087] Use TLS (Transport Layer Security) or SSL (Secure Sockets Layer) protocols to create a secure network connection to ensure that data is encrypted during transmission between the client and the server. TLS / SSL provides security guarantees such as data encryption, server authentication, message integrity verification and key exchange, and is an industry standard for protecting network communication security.
[0088] When establishing a connection, the server presents its SSL certificate to the client, which is issued by a trusted certificate authority (CA). The client verifies the validity of the certificate to ensure that the connection is established with the correct server and to prevent man-in-the-middle attacks.
[0089] In some embodiments, identifying key medical information and patient symptom information in the text message includes:
[0090] Use the trained text analysis model to identify key medical information and patient symptom information in text information.
[0091] The text analysis model in this embodiment can be obtained through the following training:
[0092] Data preparation
[0093] Dataset Preparation: In various departments of the hospital, such as outpatient, emergency, or ward, the conversations between doctors and patients are recorded in real time using dedicated recording equipment.
[0094] Labeling process:
[0095] Key information annotation: Medical experts or trained annotators accurately annotate key medical information in large texts. For example, in the text "blood sugar rises", "blood sugar" is annotated as the key information of the indicator name.
[0096] Relationship annotation: Also annotate the qualifiers or descriptive phrases around the key information, which are essential for understanding the medical significance of the key information. For example, in the sentence "blood sugar rises", "blood sugar" is annotated as the key information of the medical indicator, while "rises" is annotated as a qualifier related to it.
[0097] Model Training
[0098] Model selection: The UIE (Unified Information Extraction) model is selected as the basic model because it is suitable for processing complex texts and key information relationships in the medical field.
[0099] Data input: Use previously annotated data sets as input, which contain specific key information and relationship annotation information.
[0100] Training details: During fine-tuning, the model learns how to accurately identify the key information annotated from the text, and how to understand the relationships and qualifiers between the key information.
[0101] Test data set: Use an independent test set that is not used in the model training process.
[0102] Evaluation indicators: Evaluate the accuracy, recall rate and F1 score of the model in key information identification and relationship extraction to ensure the accuracy and generalization ability of the model.
[0103] Through this detailed training process, the UIE model is able to efficiently and accurately identify key critical information from complex medical texts and extract the relationships between them.
[0104] In some embodiments, the execution task library is constructed in the following manner:
[0105] Gather key medical information and patient symptom information from medical literature, clinical guidelines, and common medical questions;
[0106] For each key medical information and patient symptom information, create a question-answer pair;
[0107] The question part of the question-answer pair includes key medical information and patient symptom information, and the answer part includes the performance task.
[0108] Exemplary question-answer pair compilation: For each key term or symptom, create a question-answer pair. The "question" part contains key information, such as a specific symptom or condition; the "answer" part provides recommended tasks or action guidelines.
[0109] In some embodiments, the key medical information and patient symptom information are matched with the execution task library to obtain clinical execution tasks, specifically:
[0110] Match the corresponding question and answer pairs according to key medical information and patient symptom information;
[0111] The answer in the question-answer pair is called and output to obtain the clinical execution task.
[0112] When key information in the library is identified in the conversation, the corresponding question-answer pair is automatically called, and specific task execution suggestions are provided based on the matched question-answer pair, such as "perform blood pressure monitoring" or "adjust drug dosage."
[0113] In addition, professional medical staff and NLP experts are required to collaborate to complete the initial construction of the execution task library. The library is regularly updated and adjusted based on user feedback and medical development to ensure the accuracy and timeliness of the information. Through such an execution task library, clear action guidance can be quickly provided when a specific medical condition is detected, helping medical staff to respond to clinical situations efficiently.
[0114] The above embodiments extract key information from patient conversations, such as symptoms and medical history, in an automated manner, reducing the time and possible human errors of medical staff in manual recording. Using advanced natural language processing technology, the system can more accurately identify and parse important information in the patient's voice or text, improving the quality and efficiency of data collection. By improving efficiency and accuracy through automation and intelligent means, medical staff can focus more on patient care and treatment itself, thereby improving the overall quality of medical services.
[0115] In general, the above-mentioned embodiment methods aim to improve the efficiency and quality of information processing in medical practice through the combination of natural language processing and data analysis tools, thereby improving the workflow of medical staff and the treatment experience of patients.
[0116] It should be noted that the clinical task execution method based on speech analysis provided in the embodiment of the present application can be performed by a clinical task execution device based on speech analysis, or a control module in the clinical task execution device based on speech analysis for executing the method based on speech analysis for executing a clinical task. In the embodiment of the present application, the method based on speech analysis for executing a clinical task execution device based on speech analysis is taken as an example to illustrate the device for executing a clinical task based on speech analysis provided in the embodiment of the present application.
[0117] like Figure 2 As shown, in a second aspect of an embodiment of the present application, a clinical task execution device based on speech analysis is provided, and the device may include:
[0118] An acquisition module 210 is used to acquire the conversation voice between the doctor and the patient;
[0119] A conversion module 220, used to convert the conversation speech into text information;
[0120] An identification module 230 for identifying key medical information and patient symptom information in the text information;
[0121] A matching module 240 is used to match the key medical information and the patient symptom information with the execution task library to obtain a clinical execution task;
[0122] The display module 250 is used to display the clinical execution tasks to the medical staff.
[0123] The clinical task execution device based on speech analysis in the embodiment of the present application can be a device, or a component, integrated circuit, or chip in a terminal. The device can be a mobile electronic device or a non-mobile electronic device. Exemplary, the mobile electronic device can be a mobile phone, a tablet computer, a laptop computer, a palmtop computer, an in-vehicle electronic device, a wearable device, an ultra-mobile personal computer (ultra-mobile personal computer, UMPC), a netbook or a personal digital assistant (personal digital assistant, PDA), etc., and the non-mobile electronic device can be a server, a network attached storage (Network Attached Storage, NAS), a personal computer (personal computer, PC), a television (television, TV), a teller machine or a self-service machine, etc., and the embodiment of the present application is not specifically limited.
[0124] The clinical task execution device based on speech analysis in the embodiment of the present application may be a device having an operating system. The operating system may be an Android operating system, an iOS operating system, or other possible operating systems, which are not specifically limited in the embodiment of the present application.
[0125] The clinical task execution device based on speech analysis provided in the embodiment of the present application can achieve Figure 1 To avoid repetition, the various processes implemented by the method embodiment are not described here.
[0126] Alternatively, if Figure 3As shown, the embodiment of the present application also provides an electronic device 300, including a processor 301, a memory 302, and a program or instruction stored in the memory 302 and executable on the processor 301. When the program or instruction is executed by the processor 301, each process of the above-mentioned clinical task execution method embodiment based on speech analysis is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.
[0127] It should be noted that the electronic devices in the embodiments of the present application include the mobile electronic devices and non-mobile electronic devices mentioned above.
[0128] Figure 4 A schematic diagram of the hardware structure of an electronic device to implement an embodiment of the present application.
[0129] The electronic device 400 includes but is not limited to components such as a radio frequency unit 401 , a network module 402 , an audio output unit 403 , an input unit 404 , a sensor 405 , a display unit 406 , a user input unit 407 , an interface unit 408 , a memory 409 , and a processor 410 .
[0130] Those skilled in the art will appreciate that the electronic device 400 may also include a power source (such as a battery) for supplying power to each component, and the power source may be logically connected to the processor 410 through a power management system, thereby implementing functions such as managing charging, discharging, and power consumption management through the power management system. Figure 4 The electronic device structure shown in the figure does not constitute a limitation on the electronic device. The electronic device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently, which will not be described in detail here.
[0131] It should be understood that in the embodiment of the present application, the input unit 404 may include a graphics processor (GPU) 4041 and a microphone 4042, and the graphics processor 4041 processes the image data of a static picture or video obtained by an image capture device (such as a camera) in a video capture mode or an image capture mode. The display unit 406 may include a display panel 4061, and the display panel 4061 may be configured in the form of a liquid crystal display, an organic light emitting diode, etc. The user input unit 407 includes a touch panel 4071 and other input devices 4072. The touch panel 4071 is also called a touch screen. The touch panel 4071 may include two parts: a touch detection device and a touch controller. Other input devices 4072 may include, but are not limited to, a physical keyboard, a function key (such as a volume control button, a switch button, etc.), a trackball, a mouse, and a joystick, which will not be repeated here. The memory 409 may be used to store software programs and various data, including but not limited to applications and operating systems. The processor 410 may integrate an application processor and a modem processor, wherein the application processor mainly processes an operating system, a user interface, and applications, and the modem processor mainly processes wireless communications. It is understandable that the above-mentioned modem processor may not be integrated into the processor 410.
[0132] An embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the various processes of the above-mentioned clinical task execution method embodiment based on speech analysis are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.
[0133] The processor is a processor in the electronic device described in the above embodiment. The readable storage medium includes a computer readable storage medium, such as a computer read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.
[0134] An embodiment of the present application further provides a chip, which includes a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the various processes of the above-mentioned clinical task execution method embodiment based on speech analysis, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.
[0135] It should be understood that the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.
[0136] It should be noted that, in this article, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.
[0137] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a computer software product, which is stored in a storage medium (such as ROM / RAM, a disk, or an optical disk), and includes a number of instructions for a terminal (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present application.
[0138] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.
Claims
1. A clinical task execution method based on speech analysis, characterized in that: include: Obtain the audio of the conversation between doctors and patients; Converting the conversation speech into text information; identifying key medical information and patient symptom information in the text information; Matching the key medical information and the patient symptom information with an execution task library to obtain a clinical execution task, wherein the execution task library is a matching library constructed based on the key information of the question and answer and the execution task; Demonstrate the clinical execution tasks to medical staff.
2. The clinical task execution method based on speech analysis according to claim 1, characterized in that: Acquire the audio of the conversation between doctors and patients, including: Using a cardioid microphone array to capture audio of conversations between doctors and patients; The conversation audio is filtered and compressed to form conversation speech.
3. The clinical task execution method based on speech analysis according to claim 2, characterized in that: The conversation audio is filtered and compressed to form a conversation voice, including: Using a statistical filtering algorithm to reduce noise on the conversation audio to form filtered audio; The filtered audio is losslessly compressed to form a conversational speech.
4. The clinical task execution method based on speech analysis according to claim 1, characterized in that: The step of converting the conversation speech into text information comprises: The conversation speech is converted into text information using a trained speech recognition model.
5. The clinical task execution method based on speech analysis according to claim 1, characterized in that: The identifying key medical information and patient symptom information in the text information includes: The trained text analysis model is used to identify key medical information and patient symptom information in the text information.
6. The clinical task execution method based on speech analysis according to claim 1, characterized in that: The execution task library is constructed in the following way: Gather key medical information and patient symptom information from medical literature, clinical guidelines, and common medical questions; For each key medical information and patient symptom information, create a question-answer pair; The question part of the question-answer pair includes key medical information and patient symptom information, and the answer part includes performing tasks.
7. The clinical task execution method based on speech analysis according to claim 6 is characterized in that: The key medical information and the patient symptom information are matched with the execution task library to obtain the clinical execution task, specifically: Matching corresponding question-answer pairs according to the key medical information and the patient symptom information; The answer in the question-answer pair is called for output to obtain a clinical execution task.
8. A clinical task execution device based on speech analysis, characterized in that: include: An acquisition module is used to acquire the conversation voice between doctors and patients; A conversion module, used for converting the conversation speech into text information; an identification module, used for identifying key medical information and patient symptom information in the text information; A matching module, used for matching the key medical information and the patient symptom information with an execution task library to obtain a clinical execution task; The display module is used to display the clinical execution task to medical staff.
9. An electronic device, characterized in that: include: A processor, a memory, and a program or instruction stored in the memory and executable on the processor, wherein the program or instruction, when executed by the processor, implements the steps of the clinical task execution method based on speech analysis as described in any one of claims 1 to 7.
10. A readable storage medium, characterized in that: The readable storage medium stores a program or instruction, and when the program or instruction is executed by the processor, the steps of the clinical task execution method based on speech analysis as described in any one of claims 1-7 are implemented.
