Hospital guidance method and device based on voice interaction, equipment, medium and product

Abstract

The invention provides a hospital guide method and device based on voice interaction, equipment, a medium and a product, and the method comprises the steps: determining user information, historical treatment data and a corresponding environment signal-to-noise ratio based on user voice data; determining a voice keyword, a voice text and a corresponding initial confidence coefficient based on the voice data, the historical doctor-seeing data and the user information; based on the voice keyword, the voice text, the corresponding initial confidence coefficient and the environment signal-to-noise ratio, determining a target hospital guide result; and determining the corresponding broadcasting speed and the broadcasting volume, so that the hospital guide equipment broadcasts the target hospital guide result according to the broadcasting speed and the broadcasting volume. According to the method, a user does not need to perform hospital guidance through touch operation, hospital guidance is realized through a voice interaction mode, and the use difficulty of the user is reduced; the weights of the keywords and the texts are dynamically adjusted based on the environmental signal-to-noise ratio, and the hospital guide result is determined based on the adjusted keywords and texts, so that the limitation of a single recognition mode is solved, and the recognition accuracy in the noise environment is improved.

Description

Technical Field

[0001] This application relates to the field of medical information technology, and in particular to a voice-interactive-based triage method, device, equipment, medium, and product. Background Technology

[0002] In current hospital outpatient service scenarios, self-service machines have become essential equipment supporting basic services such as registration, payment, and appointment scheduling. Their application effectively alleviates the service pressure on manual service windows and promotes the efficiency of outpatient service processes. Most current outpatient self-service machines adopt a touchscreen interaction mode; in this mode, patients can complete their transactions through touch operations such as clicking and swiping on the machine.

[0003] However, the aforementioned interaction methods still have significant shortcomings, making it difficult to meet the usage needs of patients of all ages and adapt to complex outpatient environments. On the one hand, for elderly patients, the touchscreen interaction mode is highly complex to operate, and elderly patients generally have insufficient touch accuracy, leading to operational errors during business transactions and significantly reducing the user experience. On the other hand, outpatient areas are usually noisy (such as conversations and announcements), resulting in low recognition accuracy of the voice interaction system and affecting service efficiency.

[0004] Therefore, there is an urgent need for an accurate patient guidance system that can adapt to the complex and noisy environment of outpatient clinics. Summary of the Invention

[0005] This application provides a voice-interactive-based triage method, device, equipment, medium, and product to achieve efficient and accurate triage in complex noisy environments.

[0006] In a first aspect, embodiments of this application provide a voice-interactive-based triage method applied to a triage system, the triage system being used for communication connection with corresponding triage equipment, including:

[0007] In response to the user's voice data, determine the user profile and the corresponding environmental signal-to-noise ratio. The user profile includes user information and historical medical data.

[0008] Based on voice data, historical medical visit data, and user information, candidate triage query data is determined; among which, candidate triage query data includes voice keywords, initial confidence of keywords, voice text, and initial confidence of text.

[0009] Based on voice keywords, voice text, initial confidence of keywords, initial confidence of text, and environmental signal-to-noise ratio, the corresponding target triage result is determined;

[0010] Determine the broadcast speed and volume that match the user information, and send the target triage result to the corresponding triage device so that the triage device broadcasts the target triage result according to the broadcast speed and volume.

[0011] In one possible implementation, the corresponding target triage result is determined based on voice keywords, voice text, initial confidence of keywords, initial confidence of text, and environmental signal-to-noise ratio, including:

[0012] Based on speech keywords, speech text, initial confidence of keywords, initial confidence of text, and environmental signal-to-noise ratio, the weighted confidence of keywords and the weighted confidence of text are determined according to the weighting rules.

[0013] The target triage results are determined based on keyword-weighted confidence, text-weighted confidence, and knowledge graph; among them, the knowledge graph is used to indicate the relationship between symptoms and departments.

[0014] In one possible implementation, based on speech keywords, speech text, initial keyword confidence, initial text confidence, and environmental signal-to-noise ratio, the keyword-weighted confidence and text-weighted confidence are determined according to a weighting rule, including:

[0015] Based on the environmental signal-to-noise ratio, the speech keywords and speech text are weighted according to the weighting rules to obtain the corresponding keyword weights and text weights.

[0016] Based on the initial confidence of keywords, the initial confidence of text, keyword weights, and text weights, the keyword-weighted confidence and text-weighted confidence are determined.

[0017] In one possible implementation, the target triage result is determined based on keyword-weighted confidence, text-weighted confidence, and a knowledge graph, including:

[0018] The weighted confidence difference is determined based on keyword-weighted confidence and text-weighted confidence.

[0019] Determine whether the weighted confidence difference is greater than a preset difference threshold; if the weighted confidence difference is not greater than the preset difference threshold, then select the candidate triage query data with the higher weighted confidence from keyword weighted confidence and text weighted confidence as the target triage query data;

[0020] Based on the target triage query data, a query is performed on the knowledge graph to obtain the target triage results.

[0021] In one possible implementation, the method further includes:

[0022] If the weighted confidence difference is greater than the preset difference threshold, then a query is performed on the knowledge graph based on the candidate triage query data to obtain multiple corresponding candidate triage results; among them, multiple candidate triage results contain the weight value corresponding to each candidate triage result;

[0023] The candidate triage result corresponding to the highest weight value is taken as the target triage result.

[0024] In one possible implementation, the method further includes:

[0025] Obtain user location;

[0026] Based on the user's location, target triage results, and a preset navigation map, a navigation path is generated so that the user can reach the corresponding treatment room.

[0027] In one possible implementation, the method further includes:

[0028] Based on user interaction data, determine the user's status information, which includes: behavioral characteristics and the duration of processing at least one service.

[0029] Determine whether the behavioral characteristics match the preset abnormal characteristics, and determine whether the processing stay duration reaches the preset stay duration;

[0030] If the behavioral characteristics match the preset abnormal characteristics, and / or the processing time reaches the preset processing time, the proactive guidance function will be triggered to guide the user to process the corresponding business.

[0031] Secondly, embodiments of this application provide a voice-interactive-based patient guidance device, comprising:

[0032] The acquisition module is used to respond to the user's voice data to determine the user profile and the corresponding environmental signal-to-noise ratio. The user profile includes user information and historical medical data.

[0033] The processing module is used to determine candidate triage query data based on voice data, historical medical data, and user information; among which, the candidate triage query data includes voice keywords, initial confidence of keywords, voice text, and initial confidence of text.

[0034] The processing module is also used to determine the corresponding target triage result based on voice keywords, voice text, initial confidence of keywords, initial confidence of text, and environmental signal-to-noise ratio;

[0035] The feedback module is used to determine the broadcast speed and volume that match the user information, and send the target triage result to the corresponding triage device so that the triage device broadcasts the target triage result according to the broadcast speed and volume.

[0036] In one possible implementation, the processing module is further configured to:

[0037] Based on speech keywords, speech text, initial confidence of keywords, initial confidence of text, and environmental signal-to-noise ratio, the weighted confidence of keywords and the weighted confidence of text are determined according to the weighting rules.

[0038] The target triage results are determined based on keyword-weighted confidence, text-weighted confidence, and knowledge graph; among them, the knowledge graph is used to indicate the relationship between symptoms and departments.

[0039] In one possible implementation, the processing module is further configured to:

[0040] Based on the environmental signal-to-noise ratio, the speech keywords and speech text are weighted according to the weighting rules to obtain the corresponding keyword weights and text weights.

[0041] Based on the initial confidence of keywords, the initial confidence of text, keyword weights, and text weights, the keyword-weighted confidence and text-weighted confidence are determined.

[0042] In one possible implementation, the processing module is further configured to:

[0043] The weighted confidence difference is determined based on keyword-weighted confidence and text-weighted confidence.

[0044] Determine whether the weighted confidence difference is greater than a preset difference threshold; if the weighted confidence difference is not greater than the preset difference threshold, then select the candidate triage query data with the higher weighted confidence from keyword weighted confidence and text weighted confidence as the target triage query data;

[0045] Based on the target triage query data, a query is performed on the knowledge graph to obtain the target triage results.

[0046] In one possible implementation, the processing module is further configured to:

[0047] If the weighted confidence difference is greater than the preset difference threshold, then a query is performed on the knowledge graph based on the candidate triage query data to obtain multiple corresponding candidate triage results; among them, multiple candidate triage results contain the weight value corresponding to each candidate triage result;

[0048] The candidate triage result corresponding to the highest weight value is taken as the target triage result.

[0049] In one possible implementation, the processing module is further configured to:

[0050] Obtain user location;

[0051] Based on the user's location, target triage results, and a preset navigation map, a navigation path is generated so that the user can reach the corresponding treatment room.

[0052] In one possible implementation, the processing module is further configured to:

[0053] Based on user interaction data, determine the user's status information, which includes: behavioral characteristics and the duration of processing at least one service.

[0054] Determine whether the behavioral characteristics match the preset abnormal characteristics, and determine whether the processing stay duration reaches the preset stay duration;

[0055] If the behavioral characteristics match the preset abnormal characteristics, and / or the processing time reaches the preset processing time, the proactive guidance function will be triggered to guide the user to process the corresponding business.

[0056] Thirdly, embodiments of this application provide an electronic device, including: a memory and a processor;

[0057] The memory stores the instructions that the computer executes;

[0058] The processor executes computer execution instructions stored in memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.

[0059] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible implementations of the first aspect.

[0060] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0061] This application provides a voice-interactive-based triage method, apparatus, device, medium, and product. Responding to user voice data, it determines user information, historical medical data, and the corresponding environmental signal-to-noise ratio (SNR). Based on the voice data, historical medical data, and user information, it determines candidate triage query data, including voice keywords and voice text. Based on the voice keywords, voice text, and environmental SNR, it determines the corresponding target triage result. It determines a playback speed and volume matching the user information and sends the target triage result to the corresponding triage device, enabling the triage device to play the target triage result according to the playback speed and volume. This method eliminates the need for user touch operation for triage, achieving triage through voice interaction, thus reducing user difficulty. Simultaneously, based on the environmental SNR, it dynamically adjusts the weights of voice keywords and voice text, and determines the corresponding target triage result based on the adjusted voice keywords and voice text, overcoming the limitations of a single recognition mode and improving recognition accuracy in noisy environments. Attached Figure Description

[0062] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0063] Figure 1 A schematic diagram illustrating a scenario for a voice-interaction-based triage method provided in this application;

[0064] Figure 2 An interactive schematic diagram of a voice-based triage method provided in this application;

[0065] Figure 3 A flowchart illustrating a voice-interaction-based triage method provided in this application;

[0066] Figure 4 A schematic diagram of a voice-interactive-based patient guidance device provided in this application;

[0067] Figure 5 This is a schematic diagram of the structure of an electronic device provided in this application.

[0068] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0069] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0070] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with relevant laws, regulations and standards, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0071] Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover but not exclude inclusion. For example, a product or device that includes a series of components is not necessarily limited to those explicitly listed, but may include other components not explicitly listed or inherent to such product or device. As used in this application, the term "module" means any known or subsequently developed hardware, software, firmware, artificial intelligence, fuzzy logic, or combination of hardware and / or software code capable of performing the functions associated with that element.

[0072] In current hospital outpatient service scenarios, self-service machines have become essential equipment supporting basic services such as registration, payment, and appointment scheduling. Their application effectively alleviates the service pressure on manual service windows and promotes the efficiency of outpatient service processes. Most current outpatient self-service machines adopt a touchscreen interaction mode; in this mode, patients can complete their transactions through touch operations such as clicking and swiping on the machine.

[0073] However, the aforementioned interaction methods still have significant shortcomings, making it difficult to meet the usage needs of patients of all ages and adapt to complex outpatient environments. On the one hand, for elderly patients, the touchscreen interaction mode is highly complex to operate, and elderly patients generally have insufficient touch accuracy, leading to operational errors during business transactions and significantly reducing the user experience. On the other hand, outpatient areas are usually noisy (such as conversations and announcements), resulting in low recognition accuracy of the voice interaction system and affecting service efficiency.

[0074] Therefore, there is an urgent need for an accurate patient guidance system that can adapt to the complex and noisy environment of outpatient clinics.

[0075] This application provides a voice-interactive-based triage method, device, equipment, medium, and product. Responding to user voice data, the method determines user information, historical medical data, and the corresponding environmental signal-to-noise ratio (SNR). Based on the voice data, historical medical data, and user information, it determines candidate triage query data, including voice keywords and voice text. Based on the voice keywords, voice text, and environmental SNR, it determines the corresponding target triage result. It determines the playback speed and volume matching the user information and sends the target triage result to the corresponding triage device, enabling the triage device to play the target triage result according to the playback speed and volume. This method eliminates the need for user touch operation for triage, achieving triage through voice interaction, thus reducing user difficulty. Simultaneously, based on the environmental SNR, it dynamically adjusts the weights of voice keywords and voice text, and determines the corresponding target triage result based on the adjusted voice keywords and voice text, overcoming the limitations of a single recognition mode and improving recognition accuracy in noisy environments.

[0076] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0077] Figure 1 A schematic diagram illustrating a voice-interaction-based triage method provided in this application, such as... Figure 1 As shown, the triage system 100 is connected to the corresponding triage device 200.

[0078] In scenarios where users have a need for triage, the triage system 100 receives the user's voice data and determines the user's information, historical medical data, and the environmental signal-to-noise ratio corresponding to the voice data. Based on the voice data, historical medical data, user information, and environmental signal-to-noise ratio, it determines the target triage result and the broadcast speed and volume that match the user information, and sends the target triage result to the corresponding triage device.

[0079] After receiving the target triage result sent by the triage system 100, the triage device 200 broadcasts the target triage result to the user at a determined broadcast speed and volume.

[0080] Understandably, voice interaction reduces the difficulty of use for users, while dynamic weighting of voice keywords and voice text based on the environmental signal-to-noise ratio improves recognition accuracy in noisy environments. Furthermore, interaction strategies can be dynamically adjusted based on user information, such as playback speed and volume, enhancing adaptability and service efficiency across various scenarios.

[0081] Figure 2 An interactive schematic diagram of a voice-based triage method provided in this application is shown below. Figure 2 As shown, the voice-interaction-based triage method provided in this application includes:

[0082] S201. Send the collected user voice data.

[0083] The patient guidance equipment includes an environmental sensing device, which comprises a sound source localization module and a dynamic noise reduction module. Specifically, the sound source localization module consists of four microphone arrays arranged in a ring, each array containing three directional microphones distributed at 120° intervals, capable of receiving sound source information from multiple directions. The sound source localization module collects the user's voice data, and the patient guidance equipment sends the processed voice data to the patient guidance system. The dynamic noise reduction module, based on noise feature learning and spectrum restoration using a long short-term memory neural network, processes the collected voice data to reduce noise, improving recognition accuracy in noisy environments.

[0084] In addition, the triage equipment also includes an anomaly handling device, which is used to shut down the microphone array when it detects that a user has been away from the triage equipment for more than a preset time, so as to prevent the sound source localization module from continuously collecting irrelevant voice data.

[0085] S202. Based on the user's voice data, determine the user profile and the corresponding environmental signal-to-noise ratio.

[0086] Among them, the environmental signal-to-noise ratio is the power ratio of the useful signal to the background noise in the environment; for example, in the outpatient hall of a hospital, useful information refers to the voice data input by the user into the triage device, and background noise refers to the conversations between people in the hall, the broadcast announcements, etc.

[0087] User profiles include user information and historical medical records. For example, a triage device collects user A's voice data and uploads it to the triage system. Based on the voice data input by user A, the triage system determines the user information corresponding to user A and user A's historical medical records.

[0088] S203. Based on voice data, historical medical data, and user information, determine voice keywords, initial confidence levels of keywords, voice text, and initial confidence levels of text; based on voice keywords, voice text, initial confidence levels of keywords, initial confidence levels of text, and environmental signal-to-noise ratio, determine the corresponding target triage result.

[0089] Among them, voice keywords refer to symptom keywords extracted from user-input voice data and user profiles, while voice-to-text refers to directly converting user-input voice data into text. Specifically, symptom keywords are obtained by matching voice data with a medical-specific thesaurus; then, voice keywords are obtained by combining user profiles; and finally, voice data is converted into text using a voice-to-text model. Furthermore, the voice-to-text model is based on historical voice data. Specifically, historical voice data is input into the model, with intermediate results output every 300ms. The basic model is fine-tuned using hospital-scene corpora until the correct text result is output. Simultaneously, a dialect phoneme mapping table is configured in the model to achieve the conversion and recognition of user dialect speech and the understanding of complex semantics.

[0090] For example, if user A inputs the voice data "I've been feeling dizzy and my ears are ringing for the past few days", based on the voice data and the user profile corresponding to user A, the voice keywords are determined to be "dizziness, tinnitus, elderly user, and no history of seeking medical treatment for dizziness and tinnitus symptoms". Similarly, the voice text can be obtained as "I've been feeling dizzy and my ears are ringing for the past few days; user A is an elderly user; user A has never sought medical treatment for dizziness and tinnitus symptoms".

[0091] In one possible implementation, the corresponding target triage result is determined based on voice keywords, voice text, initial confidence of keywords, initial confidence of text, and environmental signal-to-noise ratio. The specific process is as follows:

[0092] Based on speech keywords, speech text, initial confidence of keywords, initial confidence of text, and environmental signal-to-noise ratio, the weighted confidence of keywords and the weighted confidence of text are determined according to the weighting rules.

[0093] The target triage results are determined based on keyword-weighted confidence, text-weighted confidence, and knowledge graph; among them, the knowledge graph is used to indicate the relationship between symptoms and departments.

[0094] The weighting rule is based on the environmental signal-to-noise ratio (SNR). Specifically, if the SNR is higher than a preset threshold, higher weight is assigned to the speech text; conversely, if the SNR is lower than the preset threshold, higher weight is assigned to the speech keywords. For example, if the preset threshold is 15 dB, and the determined environmental SNR based on the user's speech data is 19 dB, then higher weight is assigned to the speech text. Similarly, if the determined environmental SNR based on the user's speech data is 10 dB, then higher weight is assigned to the speech keywords.

[0095] Understandably, based on voice keywords, voice text, and environmental signal-to-noise ratio, and according to the weighting rules, the calculated keyword-weighted confidence and text-weighted confidence are: keyword-weighted confidence = 0.61, and text-weighted confidence = 0.22.

[0096] Knowledge graphs are used to indicate the relationships between symptoms and departments; specifically, knowledge graphs include symptom-department mapping databases, drug interaction databases, and so on. Furthermore, methods for constructing symptom-department mapping databases include, for example, first establishing symptom feature vectors based on disease codes, then using graph neural networks to calculate the association weights between symptoms and departments within the symptom feature vectors, and finally constructing the symptom-department mapping database based on the calculation results.

[0097] In one possible implementation, the target triage result is determined based on keyword-weighted confidence, text-weighted confidence, and a knowledge graph, including:

[0098] The weighted confidence difference is determined based on keyword-weighted confidence and text-weighted confidence.

[0099] Determine whether the weighted confidence difference is greater than a preset difference threshold; if the weighted confidence difference is not greater than the preset difference threshold, then select the candidate triage query data with the higher weighted confidence from keyword weighted confidence and text weighted confidence as the target triage query data;

[0100] Based on the target triage query data, a query is performed on the knowledge graph to obtain the target triage results.

[0101] The preset difference threshold can be set according to requirements, for example, to 0.4. For instance, if the keyword weighted confidence score is 0.61 and the text weighted confidence score is 0.22, the weighted confidence score difference is 0.61 - 0.22 = 0.39. Comparison shows that the weighted confidence score difference is less than the preset difference threshold. The voice keyword corresponding to the keyword weighted confidence score of 0.61 is selected as the target triage query data. Based on the target triage query data, a query is performed on the knowledge graph to obtain the target query result, which is "It is recommended to prioritize the ENT department; Dr. Wang has appointments available this afternoon."

[0102] This step determines the target triage query data by comparing the weighted confidence difference, avoiding the limitations of a single recognition mode and improving the accuracy of triage query results.

[0103] S204. Determine the broadcast speed and broadcast volume that match the user information.

[0104] Specifically, based on user A's user information, if user A is determined to be an elderly user, then the broadcast speed will be reduced by 30% and the broadcast volume will be increased by 25%.

[0105] S205, Receive the target triage result and the broadcast speed and volume that match the user information.

[0106] S206. Broadcast the target diagnosis results to the user according to the broadcast speed and volume.

[0107] The triage system sends the retrieved target triage results, along with the determined broadcast speed and volume, to the triage device so that the triage device can broadcast the triage results to user A at the specified broadcast speed and volume.

[0108] This application provides a voice-interactive-based triage method that, in response to user voice data, determines user information, historical medical data, and the corresponding environmental signal-to-noise ratio (SNR). Based on the voice data, historical medical data, and user information, candidate triage query data is determined, including voice keywords and voice text. Based on the voice keywords, voice text, and environmental SNR, a corresponding target triage result is determined. A playback speed and volume matching the user information are determined, and the target triage result is sent to the corresponding triage device, causing the triage device to play the target triage result according to the playback speed and volume. This method eliminates the need for users to perform triage via touch operation, achieving triage through voice interaction, thus reducing user difficulty. Simultaneously, based on the environmental SNR, the weights of voice keywords and voice text are dynamically adjusted, and based on the adjusted voice keywords and voice text, the corresponding target triage result is determined, overcoming the limitations of a single recognition mode and improving recognition accuracy in noisy environments.

[0109] Figure 3 The flowchart of a voice-interaction-based triage method provided in this application is as follows: Figure 3 As shown, in this embodiment... Figure 2 Based on the embodiments, the voice-interaction-based triage method is described in detail, which includes:

[0110] S301, responding to the user's voice data, determines the user profile and the corresponding environmental signal-to-noise ratio.

[0111] Step S301 is similar to step S202 above, and will not be described again here.

[0112] S302. Based on voice data, historical medical records, and user information, determine voice keywords, initial confidence levels of keywords, voice text, and initial confidence levels of text.

[0113] Step S302 is similar to step S203 above, and will not be described again here.

[0114] S303. Based on speech keywords, speech text, initial confidence of keywords, initial confidence of text, and environmental signal-to-noise ratio, determine the keyword-weighted confidence and text-weighted confidence according to the weighting rules.

[0115] In one possible implementation, based on speech keywords, speech text, initial keyword confidence, initial text confidence, and environmental signal-to-noise ratio, the keyword-weighted confidence and text-weighted confidence are determined according to a weighting rule, including:

[0116] Based on the environmental signal-to-noise ratio, the speech keywords and speech text are weighted according to the weighting rules to obtain the corresponding keyword weights and text weights.

[0117] Based on the initial confidence of keywords, the initial confidence of text, keyword weights, and text weights, the keyword-weighted confidence and text-weighted confidence are determined.

[0118] The weighting rule is based on the environmental signal-to-noise ratio (SNR). Specifically, if the SNR is higher than a preset threshold, higher weight is assigned to the speech text; conversely, if the SNR is lower than the preset threshold, higher weight is assigned to the speech keywords. For example, if the preset threshold is 15 dB, and the determined environmental SNR based on the user's speech data is 19 dB, then higher weight is assigned to the speech text. Similarly, if the determined environmental SNR based on the user's speech data is 10 dB, then higher weight is assigned to the speech keywords.

[0119] For example, based on user A's voice data, the environmental signal-to-noise ratio (SNR) is determined to be 12dB. This SNR is lower than a preset threshold, therefore, higher weights are assigned to the voice keywords. Thus, according to the weight allocation rules, weights are assigned to the voice keywords and voice text, resulting in the following keyword weights and text weights: Keyword weight = 0.72, Text weight = 0.28. Specifically, according to the weight allocation rules, the assigned keyword weight is = fixed weight × age weight = 0.72, and the text weight is = 1 - 0.72 = 0.28; where the fixed weight is determined based on the environmental SNR, specifically, the fixed weight = The age weight can be set according to the needs, for example, set to 0.8.

[0120] Furthermore, if a user's age exceeds a preset age threshold, the keyword weight is increased. For example, if the preset age threshold is 60 years old, and user A is older than 60, the keyword weight is increased by 20%. Therefore, keyword weight = fixed weight × age weight (1 + 20%). Specifically, based on user A's voice data, the initial confidence scores for keywords and text are determined as follows: initial keyword confidence score = 0.85, initial text confidence score = 0.78. Keyword weighted confidence score = initial keyword confidence score × keyword weight = 0.85 × 0.72 = 0.61; text weighted confidence score = initial text confidence score × text weight = 0.78 × 0.28 = 0.22.

[0121] This step dynamically adjusts the weights of speech keywords and speech text based on the environmental signal-to-noise ratio and a weight allocation mechanism to adapt to different environmental noise levels and user characteristics, thereby improving the accuracy and adaptability of speech recognition.

[0122] S304. Determine the target triage result based on keyword weighted confidence, text weighted confidence, and knowledge graph; whereby the knowledge graph is used to indicate the relationship between symptoms and departments.

[0123] Step S304 is similar to step S203 above, and will not be described again here.

[0124] In one possible implementation, the method further includes:

[0125] If the weighted confidence difference is greater than the preset difference threshold, then a query is performed on the knowledge graph based on the candidate triage query data to obtain multiple corresponding candidate triage results; among them, multiple candidate triage results contain the weight value corresponding to each candidate triage result;

[0126] The candidate triage result corresponding to the highest weight value is taken as the target triage result.

[0127] Similarly, if the difference between the calculated keyword-weighted confidence score and the text-weighted confidence score is greater than a preset difference threshold, then a query is performed on the knowledge graph based on the candidate triage query data to obtain multiple corresponding candidate triage results. For example, based on the voice keywords, a query on the knowledge graph yields candidate triage result 1 "ENT, weight value 0.63"; based on the voice text, a query on the knowledge graph yields candidate triage result 2 "Neurology, weight value 0.57". Based on the candidate triage results, candidate triage result 1 with a weight value of 0.63 is selected as the target triage result.

[0128] When the weighted confidence difference between the voice keywords and the voice text is greater than the preset difference threshold, both the voice keywords and the voice text will be used as the data for the triage query to avoid missing or incorrect queries due to reliance on a single recognition result, and to improve the query accuracy in noisy environments.

[0129] S305. Determine the broadcast speed and volume that match the user information; send the target triage result and the broadcast speed and volume that match the user information; broadcast the target triage result to the user according to the broadcast speed and volume.

[0130] Step S305 is similar to steps S204, S205, and S206 above, and will not be described again here.

[0131] In one possible implementation, the method further includes:

[0132] Obtain user location;

[0133] Based on the user's location, target triage results, and a preset navigation map, a navigation path is generated so that the user can reach the corresponding treatment room.

[0134] The preset navigation map refers to the hospital's indoor point cloud map. After determining the user's target triage result, a navigation path will be generated for the user based on the user's location and the target triage result, so that the user can reach the treatment department more quickly for treatment.

[0135] Specifically, for example, the system obtains the location of user A in the outpatient hall; based on user A's location and the location of the ENT department, combined with a pre-set navigation map, it generates a navigation route and displays it to the user through the patient guidance device; in addition, it uses 3D rendered navigation arrows overlaid on the navigation route to make it clearer for the user. Simultaneously, after generating the navigation route, the patient guidance device interface vibrates to indicate the location of the "Confirm Registration" button, guiding the user to click the button to register.

[0136] In one possible implementation, the method further includes:

[0137] Based on user interaction data, determine the user's status information, which includes: behavioral characteristics and the duration of processing at least one service.

[0138] Determine whether the behavioral characteristics match the preset abnormal characteristics, and determine whether the processing stay duration reaches the preset stay duration;

[0139] If the behavioral characteristics match the preset abnormal characteristics, and / or the processing time reaches the preset processing time, the proactive guidance function will be triggered to guide the user to process the corresponding business.

[0140] Among them, behavioral characteristics refer to a user's body movements, facial expressions, heart rate status, etc.

[0141] In practical applications, in addition to guiding users to register, the triage system can also handle services such as payment and inquiries. If the triage system detects any abnormal user interaction while the user is using the device to complete any service, it will trigger an active intervention mechanism to guide the user through the process.

[0142] For example, after completing their treatment in the department, User B pays through the triage device. The triage device collects User B's interaction data in real time. Based on this data, User B's status information is determined. If any abnormality is detected, the proactive guidance function is triggered. Specifically, based on User B's interaction data, User B's status information is determined. If an increased heart rate is detected, or if User B stays on any interface for more than 30 seconds or fails to pay multiple times while using the triage device, it indicates an abnormality in User B's use of the triage device. The proactive guidance function is then triggered, and the triage device broadcasts a voice message to User B: "Hello, do you need help completing your payment?" If User B confirms the need, the device guides User B to complete the payment process.

[0143] This step improves the efficiency of users' medical visits and reduces labor costs by triggering proactive guidance functions through real-time monitoring of user interaction data.

[0144] Figure 4 A schematic diagram of a voice-interactive-based patient guidance device provided in this application is shown below. Figure 4 As shown, the voice-interactive-based triage device 400 provided in this embodiment includes:

[0145] The acquisition module 401 is used to determine the user profile and the corresponding environmental signal-to-noise ratio in response to the user's voice data. The user profile includes user information and historical medical data.

[0146] The processing module 402 is used to determine candidate triage query data based on voice data, historical medical data and user information; wherein, the candidate triage query data includes voice keywords, initial confidence of keywords, voice text and initial confidence of text;

[0147] The processing module 402 is also used to determine the corresponding target triage result based on voice keywords, voice text, initial confidence of keywords, initial confidence of text, and environmental signal-to-noise ratio;

[0148] Feedback module 403 is used to determine the broadcast speed and volume that match the user information, and send the target triage result to the corresponding triage device so that the triage device broadcasts the target triage result according to the broadcast speed and volume.

[0149] In one possible implementation, the processing module 402 is further configured to:

[0150] Based on speech keywords, speech text, initial confidence of keywords, initial confidence of text, and environmental signal-to-noise ratio, the weighted confidence of keywords and the weighted confidence of text are determined according to the weighting rules.

[0151] The target triage results are determined based on keyword-weighted confidence, text-weighted confidence, and knowledge graph; among them, the knowledge graph is used to indicate the relationship between symptoms and departments.

[0152] In one possible implementation, the processing module 402 is further configured to:

[0153] Based on the environmental signal-to-noise ratio, the speech keywords and speech text are weighted according to the weighting rules to obtain the corresponding keyword weights and text weights.

[0154] Based on the initial confidence of keywords, the initial confidence of text, keyword weights, and text weights, the keyword-weighted confidence and text-weighted confidence are determined.

[0155] In one possible implementation, the processing module 402 is further configured to:

[0156] The weighted confidence difference is determined based on keyword-weighted confidence and text-weighted confidence.

[0157] Determine whether the weighted confidence difference is greater than a preset difference threshold; if the weighted confidence difference is not greater than the preset difference threshold, then select the candidate triage query data with the higher weighted confidence from keyword weighted confidence and text weighted confidence as the target triage query data;

[0158] Based on the target triage query data, a query is performed on the knowledge graph to obtain the target triage results.

[0159] In one possible implementation, the processing module 402 is further configured to:

[0160] If the weighted confidence difference is greater than the preset difference threshold, then a query is performed on the knowledge graph based on the candidate triage query data to obtain multiple corresponding candidate triage results; among them, multiple candidate triage results contain the weight value corresponding to each candidate triage result;

[0161] The candidate triage result corresponding to the highest weight value is taken as the target triage result.

[0162] In one possible implementation, the processing module 402 is further configured to:

[0163] Obtain user location;

[0164] Based on the user's location, target triage results, and a preset navigation map, a navigation path is generated so that the user can reach the corresponding treatment room.

[0165] In one possible implementation, the processing module 402 is further configured to:

[0166] Based on user interaction data, determine the user's status information, which includes: behavioral characteristics and the duration of processing at least one service.

[0167] Determine whether the behavioral characteristics match the preset abnormal characteristics, and determine whether the processing stay duration reaches the preset stay duration;

[0168] If the behavioral characteristics match the preset abnormal characteristics, and / or the processing time reaches the preset processing time, the proactive guidance function will be triggered to guide the user to process the corresponding business.

[0169] The voice-interactive triage device provided in this embodiment can execute the methods provided in the above-described method embodiments. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0170] Figure 5 This is a schematic diagram of the structure of an electronic device provided in this application. Figure 5 As shown, the electronic device 50 provided in this embodiment includes at least one processor 501 and a memory 502. Optionally, the device 50 further includes a communication component 503. The processor 501, memory 502, and communication component 503 are connected via a bus 504.

[0171] In a specific implementation, at least one processor 501 executes computer execution instructions stored in memory 502, causing at least one processor 501 to perform the above-described method.

[0172] The specific implementation process of processor 501 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0173] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0174] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0175] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0176] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0177] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0178] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0179] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0180] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0181] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0182] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0183] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0184] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0185] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A voice-interactive-based patient guidance method, characterized in that, Applied to a patient guidance system, the patient guidance system is used for communication connection with corresponding patient guidance equipment, including: In response to the user's voice data, a user profile and corresponding environmental signal-to-noise ratio are determined, wherein the user profile includes user information and historical medical data; Based on the voice data, the historical medical records, and the user information, candidate triage query data is determined; wherein, the candidate triage query data includes voice keywords, initial confidence of keywords, voice text, and initial confidence of text; Based on the voice keywords, the voice text, the initial confidence of the keywords, the initial confidence of the text, and the environmental signal-to-noise ratio, the corresponding target triage result is determined; Determine the broadcast speed and volume that match the user information, and send the target triage result to the corresponding triage device so that the triage device broadcasts the target triage result according to the broadcast speed and volume.

2. The method according to claim 1, characterized in that, The process of determining the corresponding target triage result based on the voice keywords, the voice text, the initial confidence level of the keywords, the initial confidence level of the text, and the environmental signal-to-noise ratio includes: Based on the speech keywords, the speech text, the initial confidence of the keywords, the initial confidence of the text, and the environmental signal-to-noise ratio, the keyword weighted confidence and the text weighted confidence are determined according to the weight allocation rules. The target triage result is determined based on the keyword weighted confidence score, text weighted confidence score, and knowledge graph; wherein, the knowledge graph is used to indicate the association between symptoms and departments.

3. The method according to claim 2, characterized in that, The process of determining the keyword-weighted confidence and text-weighted confidence based on the speech keywords, the speech text, the initial confidence of the keywords, the initial confidence of the text, and the environmental signal-to-noise ratio, according to a weighting allocation rule, includes: Based on the environmental signal-to-noise ratio, the speech keywords and speech text are weighted according to the weight allocation rules to obtain the corresponding keyword weights and text weights. Based on the initial confidence of the keywords, the initial confidence of the text, the keyword weights, and the text weights, the weighted confidence of the keywords and the weighted confidence of the text are determined.

4. The method according to claim 3, characterized in that, The process of determining the target triage result based on the keyword-weighted confidence score, text-weighted confidence score, and knowledge graph includes: Based on the weighted confidence scores of the keywords and the weighted confidence scores of the text, the weighted confidence score difference is determined; Determine whether the weighted confidence difference is greater than a preset difference threshold; if the weighted confidence difference is not greater than the preset difference threshold, then select the candidate triage query data with the higher weighted confidence from the keyword weighted confidence and the text weighted confidence as the target triage query data; Based on the target triage query data, a query is performed on the knowledge graph to obtain the target triage result.

5. The method according to claim 4, characterized in that, The method further includes: If the weighted confidence difference is greater than the preset difference threshold, then based on the candidate triage query data, a query is performed on the knowledge graph to obtain multiple corresponding candidate triage results; wherein, the multiple candidate triage results include the weight value corresponding to each candidate triage result; The candidate triage result corresponding to the highest weight value is taken as the target triage result.

6. The method according to claim 1, characterized in that, The method further includes: Obtain user location; Based on the user's location, the target triage result, and the preset navigation map, a navigation path is generated so that the user can reach the corresponding treatment room based on the navigation path.

7. The method according to any one of claims 1-6, characterized in that, The method further includes: Based on the user's interaction data, the user's status information is determined, including: behavioral characteristics and the processing time for at least one service. Determine whether the behavioral characteristics match preset abnormal characteristics, and determine whether the processing stay duration reaches the preset stay duration; If the behavioral characteristics match the preset abnormal characteristics, and / or the processing time reaches the preset time, the active guidance function is triggered to guide the user to process the corresponding business.

8. A voice-interactive-based patient guidance device, characterized in that, include: The acquisition module is used to determine the user profile and the corresponding environmental signal-to-noise ratio in response to the user's voice data, wherein the user profile includes user information and historical medical data; The processing module is used to determine candidate triage query data based on the voice data, the historical medical data, and the user information; wherein, the candidate triage query data includes voice keywords, initial confidence of keywords, voice text, and initial confidence of text; The processing module is further configured to determine the corresponding target triage result based on the voice keywords, the voice text, the initial confidence of the keywords, the initial confidence of the text, and the environmental signal-to-noise ratio; The feedback module is used to determine the broadcast speed and volume that match the user information, and send the target triage result to the corresponding triage device so that the triage device broadcasts the target triage result according to the broadcast speed and volume.

9. An electronic device, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-7.

Images