Efficient mobile terminal for epidemiological investigation

Abstract

The invention discloses an efficient mobile terminal for epidemiological investigation in the technical field of epidemiological investigation and analysis, and the terminal comprises a collection module which is used for collecting basic information of an investigated object, and guiding and collecting voice information of a behavior track of the investigated object; the recognition and analysis module is used for recognizing dialect voice text information; a medical record calling module; the track judgment module is used for judging a behavior track route of the investigated object on a map within a preset time by using a neural network according to the living position and the voice text information; the popular network module is used for calling related traffic information and medical record information according to the behavior track route to construct a popular network; and the result display module is used for displaying the survey result. According to the method, the speech recognition text is analyzed and recognized through the long-short term memory network, the real behavior track route is constructed, interference of dialects and place common names on behavior track route construction is avoided, and the efficiency of epidemiological investigation and the effect in remote areas are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of epidemiological investigation and analysis, and specifically relates to a mobile terminal for efficient epidemiological investigation. Background Art

[0002] Epidemiological investigation refers to a research method that uses investigation methods to collect and analyze data related to diseases in a specific population to understand the epidemic situation of diseases and related factors. Its purpose and significance lie in mastering the epidemic situation of diseases, providing data support for public health decision-making, thereby formulating effective prevention and control strategies to protect public health. Based on the principle of epidemiological investigation and analysis combined with wireless communication devices, it has been possible to conveniently investigate the conditions of epidemic cases, such as the condition, behavior trajectory, and basic information, and conduct statistics and analysis to quickly explore the source of infection, transmission route, and risk population.

[0003] However, in some remote areas, unfamiliarity with the use method of intelligent devices, taboos about infectious diseases, and strong dialect accents have all become obstacles to the integrity and authenticity of epidemiological investigation data. Among them, for the collection of important data for analyzing the behavior trajectory route of the investigation object, due to the difference between the local common names and official place names, even if dialect speech recognition is performed, it is very difficult to correspond the place names with the locations on the map, thereby constructing a complete behavior trajectory route map.

[0004] Therefore, it is necessary to propose a mobile terminal for efficient epidemiological investigation that can perform voice collection and dialect speech recognition, and can identify locations and route descriptions through neural network analysis, thereby constructing a real behavior trajectory route on the map, exploring the source of infection and risk population, and visually displaying the investigation results. Summary of the Invention

[0005] In order to solve the above problems, the purpose of the present invention is to provide a mobile terminal for efficient epidemiological investigation, which uses a long short-term memory network to analyze and identify place names, routes, and road condition descriptions in the speech recognition text, constructs a real behavior trajectory route on the map, avoids the interference of dialects and local common names on the construction of the behavior trajectory route, realizes efficient collection and analysis and identification of epidemiological data, and improves the working efficiency of the mobile terminal for epidemiological investigation and the promotion effect in remote areas.

[0006] In order to achieve the above purpose, the technical solution of the present invention is as follows: A mobile terminal for efficient epidemiological investigation includes a collection module, an identification and analysis module, a medical record calling module, a trajectory judgment module, an epidemic network module, and a result display module.

[0007] The collection module is used to collect the basic information of the investigation object and guide and collect the voice information of the behavior trajectory of the investigation object.

[0008] An identification and analysis module, which is used to identify the residence location of the survey object according to the collected information, identify the dialect speech text information of the survey object, and mark the location of the survey object's address on the map;

[0009] A medical record retrieval module, which is used to retrieve the relevant medical record information of the survey object through a medical database;

[0010] A trajectory judgment module, which is used to use a trained neural network to judge the behavioral trajectory route information of the survey object on the map within a preset time according to the residence location and speech text information of the survey object, make a common saying note for the place names on the map, and call traffic information according to the map route;

[0011] An epidemic network module, which is used to construct an epidemic network based on the behavioral trajectory route of the survey object, call relevant traffic information and medical record information, and obtain the contact information of other potential objects or the infection risk of this survey object;

[0012] A result display module, which is used to display the personal survey results and the epidemic survey results according to the constructed epidemic network.

[0013] The principle of the basic solution is as follows: First, the collection module collects the basic information of the survey object, including name, age, gender, etc., and guides the survey object to describe their behavioral trajectory within a period of time through voice, such as where they have been, what means of transportation they have taken, etc. Subsequently, the identification and analysis module uses advanced speech recognition technology to convert the voice information of the survey object into text information. After obtaining the behavioral trajectory of the survey object, the medical record retrieval module will retrieve the medical record information of the survey object through the interface with the medical database for subsequent analysis. The trajectory judgment module uses a trained neural network, and this module uses a long short-term memory network (LSTM) to deeply analyze the text information, analyzes and identifies reliable routes and the most authentic behavioral trajectory route according to the road condition information in the voice description text, and excludes the influence of the difference between the common name of the place and the official place name. After the identification is completed, the module will mark the residence address of the survey object and its behavioral trajectory route on the map. This module will make a common saying note for the place names on the map and the traffic information called, so as to construct a more real and accurate behavioral trajectory route. After obtaining the behavioral trajectory, traffic information and medical record information of the survey object, the epidemic network module will use this information to construct an epidemic network. This network can reveal the connection between the survey object and other potential infected objects, so as to evaluate the infection risk of the survey object and find the contact information of other potential objects. Finally, the result display module will visually display the personal survey results and the epidemic survey results for the reference of investigators and decision-makers.

[0014] The beneficial effects of the basic solution are as follows: 1. Traditional epidemiological investigations often rely on manual interviews and records, which is a time-consuming, laborious, and error-prone process. The mobile terminal of the present invention greatly shortens the investigation cycle through automated collection, speech recognition, and analysis, while reducing errors caused by human factors, ensuring the accuracy and integrity of data. Especially during large-scale outbreaks, rapid and accurate data collection is crucial for timely formulating prevention and control strategies. The application of the present invention can quickly gather a large amount of first-hand information, providing strong data support for decision-makers.

[0015] 2. The present invention utilizes the Long Short-Term Memory (LSTM) technology to achieve accurate recognition and understanding of dialect speech texts. This not only solves the dialect barrier but also can analyze and identify the most authentic and reliable behavioral trajectory route by survey respondents describing various road conditions on the route (such as the names of roads passed, road orientations, road conditions, etc.), avoiding being affected by dialects, unclear descriptions, and difficult-to-distinguish local names. Through in-depth analysis of the behavioral trajectories of survey respondents, the trajectory judgment module of the present invention can intelligently identify potential risk areas and populations, providing the possibility for precise prevention and control.

[0016] 3. In remote areas, due to limited resources, traditional epidemiological investigations are often difficult to cover comprehensively. The mobile terminal of the present invention, with its portability and intelligence, can easily penetrate geographical barriers and achieve effective investigations in these areas. In addition, by invoking medical record information in the medical database, the present invention can also comprehensively analyze the health status of survey respondents and further explore potential risk factors.

[0017] 4. Based on the rich data collected, the epidemic network module of the present invention can construct a dynamic model of epidemic transmission, helping decision-makers more intuitively understand the development trend and potential risks of the epidemic. This not only helps formulate more scientific and reasonable prevention and control strategies but also quickly locks in high-risk areas and populations at the initial stage of the epidemic outbreak, providing strong support for timely taking measures such as isolation and testing.

[0018] 5. The application of the present invention can also display the effectiveness and importance of epidemic prevention and control to the public through an intuitive result display module, thereby stimulating the public's enthusiasm for participation and improving health awareness. Through cooperation with departments such as education and publicity, the mobile terminal of the present invention can also be used as an important tool for health education to popularize epidemic prevention and control knowledge and enhance the prevention and control capabilities of the whole society.

[0019] Furthermore, the acquisition module includes a touch screen and a non-touch acquisition unit.

[0020] The touch screen is used to input the basic information of the survey respondent, including name, gender, age, address, and contact information.

[0021] The non-contact collection unit is used to verify the identity information of the survey subject and guide and collect the voice information of the survey subject. Further, the non-contact collection unit includes an RFID reader, a camera, a voice player and a real-time voice collector, and the RFID reader is used to read the identity certificate of the survey subject to verify the real identity of the survey subject;

[0022] Cameras to verify the true identity of survey subjects through facial recognition;

[0023] A voice player is used to play voice information to guide the survey subject to describe the route of the behavior trajectory within a preset time, including the place name, road name, road direction, road conditions and transportation tools;

[0024] A real-time voice collector is used to collect voice information about the behavioral trajectories described by the respondents.

[0025] The beneficial effects of the basic solution are: 1. Through the intuitive interface design, the survey subjects can easily enter basic information such as name, gender, age, address and contact information, etc., without the need for cumbersome paper filling, which greatly improves the efficiency of data collection. At the same time, the touch screen can effectively avoid inaccurate data caused by unclear handwriting or wrong filling.

[0026] 2. The non-contact collection unit integrates RFID readers, cameras and facial recognition technology, providing multiple guarantees for the identity verification of the survey subjects. The RFID reader can quickly read and verify the identity of the survey subjects to ensure the authenticity of the identity information; the camera further confirms the identity of the survey subjects through facial recognition technology, effectively preventing identity fraud or misidentification, and enhancing the security and credibility of the data.

[0027] 3. The voice player can play preset voice guidance to guide the respondents to describe in detail the route of their behavior within the preset time, including the place name, road name, road direction, road conditions, transportation and other information. This intelligent guidance method not only reduces the communication difficulty of the respondents, but also ensures the integrity and accuracy of the data. At the same time, the voice guidance can be flexibly adjusted according to the actual situation of the respondents, reflecting the humanization of the design.

[0028] 4. The real-time voice collector uses advanced voice recognition technology to collect the voice information of the behavior trajectory described by the survey subjects with high fidelity. This function not only improves the real-time nature of data collection, but also provides a reliable basis for subsequent data analysis and processing. At the same time, real-time voice collection can also effectively avoid information omissions or misunderstandings caused by vague memory or language barriers.

[0029] 5. Through the organic combination of the touch screen and the non-touch acquisition unit, the mobile terminal of the present invention provides a convenient and safe investigation environment for the subjects. The subjects can easily input basic information, verify their identities, and describe their behavior trajectories without performing complex operations or exposing too much personal information. This design not only improves the participation of the subjects but also enhances their comfort and trust.

[0030] Furthermore, the recognition and analysis module includes an acoustic feature extraction unit, an acoustic model unit, and a map location annotation unit.

[0031] The acoustic feature extraction unit is used to filter and extract the acoustic features in the voice information of the behavior trajectory described by the subject.

[0032] The acoustic model unit is used to convert the extracted acoustic features into phonemes and identify the voice text information. The acoustic model unit includes several dialect acoustic pronunciation models.

[0033] The map location annotation unit is used to mark the address of the subject obtained by collection on the map to obtain the address location information.

[0034] The beneficial effects of the basic solution are as follows: 1. The acoustic feature extraction unit is responsible for filtering and extracting the key acoustic features from the voice information of the behavior trajectory described by the subject. This process not only removes background noise and redundant information but also retains the core features that can accurately reflect the voice content. This high-precision and robust feature extraction method provides a solid foundation for subsequent speech recognition.

[0035] 2. The acoustic model unit is built-in with several dialect acoustic pronunciation models and can accurately identify according to the pronunciation characteristics of different dialects. This design enables the mobile terminal of the present invention to be widely applied in different regions, effectively solving the dialect barrier problem. At the same time, through continuous training and optimization of the acoustic model, the recognition accuracy of the present invention has been significantly improved, further enhancing the reliability of the data.

[0036] 3. The map location annotation unit can accurately mark the address location on the map according to the address information of the subject obtained by collection. This function not only provides intuitive geographical location information for epidemiological investigations but also provides strong support for subsequent trajectory analysis and epidemic prevention and control. At the same time, the accuracy of the map location annotation also ensures the accuracy and reliability of data analysis.

[0037] 4. Through the organic combination of functions such as acoustic feature extraction, dialect recognition, speech text conversion, and map location annotation, the mobile terminal of the present invention realizes the rapid and accurate collection and analysis of the behavioral trajectories of the surveyed objects. This improved design not only improves the efficiency of epidemiological investigations but also ensures the accuracy and integrity of the data, providing strong technical support for epidemic prevention and control and public health management.

[0038] Furthermore, the trajectory judgment module includes a judgment unit, a map annotation unit, and a traffic information invocation unit.

[0039] The judgment unit includes a long short-term memory network trained with descriptive text, which is used to judge the true arrival location and route of the surveyed object based on the recognized speech text information to obtain behavioral trajectory route information and display it on the map.

[0040] The map annotation unit is used to annotate the official map names corresponding to the common saying place names in the corresponding speech text information after judgment.

[0041] The traffic information invocation unit is used to invoke the traffic order information after the surveyed object takes the corresponding means of transportation on the smartphone according to the speech text information and add it to the behavioral trajectory route information.

[0042] Furthermore, the long short-term memory network includes an input gate, a hidden layer, a forget gate, and an output gate.

[0043] The input gate is used to select the current input and the hidden layer unit of the previous time step to update the hidden layer.

[0044] The hidden layer is used to encode the input feature vector to capture the long-distance dependencies in the encoded sequence.

[0045] The forget gate is used to select the hidden layer state of the previous time step and discard invalid information.

[0046] The output gate is used to select the hidden layer to generate an output encoded sequence.

[0047] Furthermore, the long short-term memory network is optimized by the backpropagation algorithm to minimize the loss function of the network.

[0048] Furthermore, the descriptive text for training the long short-term memory network includes text information on locations, routes, road conditions, and directions.

[0049] The beneficial effects of the basic solution are as follows: 1. The long short-term memory network (LSTM) in the judgment unit is trained with descriptive texts and can accurately understand and judge the real arrival location and route of the survey object. The unique structure of the LSTM, including the input gate, hidden layer, forget gate, and output gate, enables it to capture long-distance dependencies in the encoded sequence and effectively handle complex and variable trajectory descriptions. This intelligent judgment mechanism not only improves the accuracy of trajectory recognition but also reduces the error rate of ordinary program judgments.

[0050] 2. The map annotation unit can automatically convert the common-place names in the voice text information into the official map names and make annotations. This function not only solves the ambiguity problem in place name recognition but also makes the trajectory display more intuitive and accurate. The survey object does not need to worry about trajectory misjudgment caused by unclear place name expressions, greatly improving the convenience and practicality of the survey.

[0051] 3. The traffic information calling unit can intelligently call the transportation tool ride order information in the survey object's smartphone and integrate it into the behavior trajectory route information. This function not only enriches the dimension of trajectory data but also ensures the comprehensiveness and real-time nature of the trajectory information. By calling the traffic order information, the investigators can more accurately understand the travel mode and time of the survey object, providing more accurate data support for epidemic prevention and control.

[0052] 4. The LSTM network is optimized by the backpropagation algorithm, which can minimize the loss function and improve the recognition accuracy and generalization ability of the network. At the same time, by training descriptive texts containing text information such as locations, routes, road conditions, and directions, the LSTM network can better adapt to the trajectory description habits in different regions, enhancing the adaptability and flexibility of the system.

[0053] 5. Through the introduction and refined design of the trajectory judgment module, the mobile terminal of the present invention can accurately capture and comprehensively display the behavior trajectory of the survey object. This improved design not only improves the depth and breadth of epidemiological investigations but also provides more detailed data support for epidemic prevention and control. The investigators can quickly lock down high-risk areas and populations based on the trajectory information, providing a strong basis for timely taking prevention and control measures.

[0054] Furthermore, the epidemic network module includes an epidemic classification unit, a route network unit, and an antenna network unit.

[0055] The epidemic classification unit is used to classify the epidemic network according to the types of epidemics based on the relevant medical record information of the survey object obtained by calling.

[0056] The route network unit is used to overlap the routes of several survey objects of the same type of epidemic on the map according to the behavior trajectory route information to construct a behavior trajectory route network.

[0057] An antenna network unit is used to obtain pre-infected objects with infection risks according to the behavior trajectory route network and traffic order information, call the order information to obtain the contact information of the pre-infected objects, and construct an antenna network.

[0058] The beneficial effects of the basic solution are as follows: 1. The epidemic classification unit can accurately classify the epidemic network according to the relevant medical record information of the surveyed objects obtained by the call, by epidemic type. This function not only helps the investigators quickly identify the types and characteristics of the epidemics, but also provides a scientific basis for the subsequent construction of the behavior trajectory route network. Through classified management, the investigators can formulate corresponding prevention and control strategies for different epidemics, improving the pertinence and effectiveness of the prevention and control measures.

[0059] 2. The route network unit can overlap and construct the behavior trajectory route network of the surveyed objects of the same epidemic type on the map according to the behavior trajectory route information. This function enables the investigators to intuitively see the spread of the epidemic in different regions and different time periods, providing visual data support for epidemic prevention and control. At the same time, the behavior trajectory route network is also dynamic and can be continuously updated with the addition of new survey data, ensuring the timeliness and accuracy of the data.

[0060] 3. The antenna network unit can quickly identify pre-infected objects with infection risks according to the behavior trajectory route network and traffic order information, call the order information to obtain their contact information, and construct an antenna network. This function not only expands the coverage of epidemic prevention and control, but also can, through an efficient information transmission mechanism, timely convey the prevention and control information to potential infected persons, reducing the risk of epidemic spread. The construction of the antenna network helps to form a closed-loop management of epidemic prevention and control, improving the overall efficiency of the prevention and control work.

[0061] 4. By introducing the epidemic network module and refining the design of its internal functional units, the mobile terminal of the present invention realizes the intelligent construction and management of the epidemic transmission network. The investigators can quickly obtain accurate and comprehensive epidemiological data without cumbersome manual operations, providing a scientific basis for epidemic prevention and control. This improved design not only improves the intelligent level of epidemiological investigations, but also reduces the investigation cost and improves the investigation efficiency.

[0062] Furthermore, a result display module includes a time series diagram, a regional infection level diagram, a route network diagram, an antenna network diagram, and a pre-infection warning diagram.

[0063] The time series diagram is used to show the cumulative number of infected cases on the time axis to statisticians and surveyed objects.

[0064] The regional infection level diagram is used to show the cumulative number of infected cases appearing in the map area to statisticians and surveyed objects.

[0065] Route network diagram, used to show the route network of the behavior tracks of the surveyed objects to the statisticians;

[0066] Antenna network diagram, used to show the antenna network of the contact information of the pre-infected objects to the statisticians;

[0067] Pre-infection warning diagram, used to show the infection risk and possible infection locations to the pre-infected people.

[0068] The beneficial effects of the basic plan are as follows: 1. The time series diagram can show the cumulative number of infection cases on the time axis to the statisticians and the surveyed objects, clearly reflecting the trend of the epidemic situation changing over time. This function not only helps the statisticians quickly grasp the development dynamics of the epidemic situation, but also provides intuitive epidemic information for the surveyed objects, enhancing their awareness and cooperation in epidemic prevention and control.

[0069] 2. The regional infection level diagram can visually show the cumulative number of infection cases in each region on the map, helping the statisticians and the surveyed objects quickly identify high-risk regions. This function not only helps the statisticians formulate regional prevention and control strategies, but also provides guidance for safe travel for the surveyed objects, reducing the infection risk.

[0070] 3. The route network diagram can show the route network of the behavior tracks of the surveyed objects to the statisticians, helping the statisticians trace the transmission path of the infection cases and analyze the correlation between cases. This function not only helps the statisticians quickly lock in high-risk populations and regions, but also provides accurate trajectory data support for epidemic prevention and control. Through the route network diagram, the statisticians can more deeply understand the transmission mechanism of the epidemic situation and provide a strong basis for formulating scientific prevention and control measures.

[0071] 4. The antenna network diagram can show the antenna network of the contact information of the pre-infected objects to the statisticians, helping the statisticians quickly establish an early warning mechanism, timely contact potential infected people, and reduce the risk of epidemic transmission. This function not only helps the statisticians expand the coverage of epidemic prevention and control, but also improves the efficiency and accuracy of the early warning mechanism. Through the antenna network diagram, the statisticians can more efficiently manage the pre-infected objects and win precious time for epidemic prevention and control.

[0072] 5. The pre-infection warning diagram can show the infection risk and possible infection locations to the pre-infected people, helping the pre-infected people timely understand the risk environment they are in and take necessary self-protection measures. This function not only helps enhance the self-protection awareness of the pre-infected people, but also reduces the risk of epidemic transmission.

[0073] 6. The mobile terminal of the present invention realizes a comprehensive and intuitive display of the results of epidemiological investigations. This improved design not only enhances the readability and usability of the results of epidemiological investigations, but also provides a more convenient and efficient information acquisition path for statisticians and survey respondents. Through diverse chart displays, statisticians and survey respondents can more intuitively understand the epidemic situation. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 FIG. is a schematic diagram of an efficient mobile terminal for epidemiological investigations in an embodiment of the present invention.

[0075] Figure 2 FIG. is a schematic diagram of the acquisition module of an efficient mobile terminal for epidemiological investigations in an embodiment of the present invention.

[0076] Figure 3 FIG. is a schematic diagram of the identification and analysis module of an efficient mobile terminal for epidemiological investigations in an embodiment of the present invention.

[0077] Figure 4 FIG. is a schematic diagram of the trajectory judgment module of an efficient mobile terminal for epidemiological investigations in an embodiment of the present invention.

[0078] Figure 5 FIG. is a schematic diagram of the epidemic network module of an efficient mobile terminal for epidemiological investigations in an embodiment of the present invention.

[0079] Figure 6 FIG. is a schematic diagram of the result display module of an efficient mobile terminal for epidemiological investigations in an embodiment of the present invention.

[0080] Figure 7 FIG. is a schematic diagram of the implementation method of an efficient mobile terminal for epidemiological investigations in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0081] The following is a further detailed description through specific embodiments:

[0082] Embodiment 1

[0083] Basically as shown in FIGS. Figure 1 , Figure 2 , Figure 3 , Figure 4 , Figure 5 , Figure 6 and Figure 7 shown: An efficient mobile terminal for epidemiological investigations includes an acquisition module, an identification and analysis module, a medical record calling module, a trajectory judgment module, an epidemic network module, and a result display module.

[0084] Collection module, which is used to collect the basic information of the surveyed object, guide and collect the voice information of the behavior track of the surveyed object. The collection module includes a touch screen and a non-touch collection unit. The touch screen is used to input the basic information of the surveyed object, including name, gender, age, address and contact information. The non-touch collection unit is used to verify the identity information of the surveyed object, guide and collect the voice information of the surveyed object. The non-touch collection unit includes an RFID reader, a camera, a voice player and a real-time voice collector. The RFID reader is used to read the identity certificate of the surveyed object to verify the true identity of the surveyed object. The camera is used to verify the true identity of the surveyed object through face recognition. The voice player is used to play voice to guide the surveyed object to describe the voice information of the behavior track route within a preset time, including place names, road names, road directions, road conditions and means of transportation. The real-time voice collector is used to collect the voice information of the behavior track described by the surveyed object.

[0085] Recognition and analysis module, which is used to identify the living location of the surveyed object according to the collected information, identify the dialect voice text information of the surveyed object, and mark the location of the address of the surveyed object on the map. The recognition and analysis module includes an acoustic feature extraction unit, an acoustic model unit and a map location marking unit. The acoustic feature extraction unit is used to filter and extract the acoustic features in the voice information of the behavior track described by the surveyed object. The acoustic model unit is used to convert the extracted acoustic features into phonemes and identify the voice text information. The acoustic model unit includes several dialect acoustic pronunciation models. The map location marking unit is used to mark the address of the surveyed object obtained according to the collection on the map to obtain the address location information.

[0086] Medical record calling module, which is used to call the relevant medical record information of the surveyed object through the medical database.

[0087] Trajectory judgment module, which is used to use the trained neural network to judge the behavior track route information of the surveyed object on the map within a preset time according to the living location and voice text information of the surveyed object, and make a colloquial note for the map place names, and call the traffic information according to the map route. The trajectory judgment module includes a judgment unit, a map note unit and a traffic information calling unit. The judgment unit includes a long short-term memory network trained by the description text, which is used to judge the true arrival location and route of the surveyed object according to the recognized voice text information to obtain the behavior track route information and display it on the map. The map note unit is used to make a note of the colloquial place names in the corresponding voice text information after judgment corresponding to the official map name. The traffic information calling unit is used to call the traffic order information after the surveyed object takes the corresponding means of transportation on the smart phone according to the voice text information and add it to the behavior track route information.

[0088] The description text for training the long short-term memory network includes text information on locations, routes, road conditions, and directions. The long short-term memory network includes an input gate, a hidden layer, a forget gate, and an output gate. The input gate is used to select the current input and the hidden layer units at the previous time step to update the hidden layer. The hidden layer is used to encode the input feature vectors to capture long-range dependencies in the encoded sequence. The forget gate is used to select the hidden layer state at the previous time step and discard invalid information. The output gate is used to select the hidden layer to generate an output encoded sequence. The long short-term memory network is optimized by minimizing the loss function through the backpropagation algorithm.

[0089] The epidemic network module is used to construct an epidemic network based on the behavioral trajectory route of the surveyed object, call relevant traffic information and medical record information, and obtain the contact information of other preparatory objects or the infection risk of the surveyed object. The epidemic network module includes an epidemic classification unit, a route network unit, and an antenna network unit. The epidemic classification unit is used to classify the epidemic network according to the types of epidemics based on the relevant medical record information of the surveyed object obtained by the call. The route network unit is used to overlap the routes of several surveyed objects of the same epidemic type on the map according to the behavioral trajectory route information to construct a behavioral trajectory route network. The antenna network unit is used to obtain pre-infected objects with infection risks based on the behavioral trajectory route network and traffic order information, call the order information to obtain the contact information of the pre-infected objects, and construct an antenna network.

[0090] The result display module is used to display the personal survey results and epidemic survey results based on the constructed epidemic network.

[0091] The specific implementation process is as follows: The main purpose of epidemiological investigation is to collect case information, find the source of infection, transmission routes, and transmission times through statistical analysis, and promptly control and eliminate the source of infection to inhibit the spread and development of epidemics. However, in some remote areas, due to the limited promotion of smartphones, the collection of behavioral trajectory information of surveyed objects is not accurate and comprehensive enough, making it difficult to construct a complete path network, find the source of infection, and remind other pre-infected individuals with infection risks, thus wasting valuable time to prevent the spread of epidemics.

[0092] Especially the widespread existence of dialects, language text information must be accurately identified through a dialect acoustic model. And there are common local names for locations in many places, which cannot be simply translated into official place names through dialects and can only be clearly understood by local acquaintances. This undoubtedly affects the timeliness and accuracy of epidemiological investigations.

[0093] Such as Figure 2As shown, for the purpose of solving the above problems, the present invention takes the form of a mobile terminal, which can be either an APP software in a smart phone or a physical device. Through the touch screen of the acquisition module, the respondent or the staff can register basic information. Then, through a non-touch RFID reader and a camera, the identity of the respondent is verified to ensure the authenticity of the epidemiological survey content. The program-controlled language player is used to ask and guide the respondent to describe their behavior trajectory route, which includes the places the respondent went to, the names of the roads passed through, the road directions, the road conditions (such as whether the road is uphill or downhill, near a river or winding around a mountain, etc.), and the means of transportation taken to reach the destination, etc. These voice messages are recorded by the real-time voice collector and used as the basis for subsequent identification and judgment of the real behavior trajectory. The voice recording method also minimizes the difficulty for people with lower educational levels or older people to operate the acquisition device, expanding the scope of the objects of the epidemiological survey.

[0094] The collected voice information is filtered by the identification and analysis module and the acoustic features are extracted. Through the acoustic model unit conversion of the pre-set dialect acoustic pronunciation model for the corresponding region, accurate voice text information is obtained through identification, avoiding the influence of dialect corresponding voice recognition and improving the accuracy of the identification of the behavior trajectory description information. At the same time, the map location marking unit marks the address filled in the basic information in the map information to obtain azimuthal information data for subsequent judgment of the behavior trajectory route.

[0095] Such as Figure 7As shown in the figure, the behavioral trajectory description text information after speech recognition is judged by the long short-term memory network. The hidden layer of the long short-term memory network can capture the information text of the location and road status, which serves as an auxiliary reference for clarifying the location and road. For example, in a certain county near a river, there are two bridges, named Jian'an Bridge and Lin'an Bridge respectively, but the local people commonly call them South Bridge and North Bridge. If directly recognized by the map information program, it may be difficult to accurately determine which bridge the unnoted South Bridge and North Bridge are respectively. Similarly, the same situation may occur with road names. When there are multiple unnoted local names of locations in the behavioral trajectory description text, the behavioral trajectory route of the survey object will be difficult to identify and represented on the map. However, the long short-term memory network trained with the description text of location, route, road conditions, and direction can accurately determine the connection between the official road name and the local road name based on the auxiliary reference combined with the map information, so as to correspond the local name and the official name one by one, reduce the interference of local names of locations on the collection of epidemiological information, and obtain a complete and coherent behavioral trajectory route on the map. After being judged by the judgment unit, the local name can be noted after the official name on the map as the judgment basis for the next occurrence. The traffic information call unit takes advantage of the convenient feature of extracting ecological information from the mobile phone APP and can directly extract the orders of the means of transportation taken by the survey object during the corresponding time period through the mobile phone, such as online car-hailing, scanned code-paying buses, and taxis. Obviously, the drivers and co-riders of these means of transportation are pre-infected individuals with a relatively high risk of infection.

[0096] After obtaining the real behavioral trajectory route of the survey object, the epidemic classification unit in the epidemic network module can classify the epidemic according to the medical records called by the medical record call module and obtain the transmission route of the epidemic, such as droplet transmission, aerosol transmission, blood transmission, etc. The route network unit overlaps and combines the behavioral trajectory routes of all survey objects collected according to the behavioral trajectory route information and the map to obtain a trajectory route network, which will serve as an important basis for finding the source of infection and controlling the infection path. The antenna network unit is to count all possible contact information of pre-infected individuals and construct an antenna network in the form of a relationship diagram to facilitate disease control personnel to contact pre-infected individuals, understand the epidemic situation, and remind pre-infected individuals to avoid the occurrence of a larger number of pre-infected individuals in a wider range under unknown circumstances.

[0097] Embodiment 2

[0098] The difference from the above embodiment is that as shown in the appendix Figure 1 、 Figure 6 and Figure 7As shown: the result display module includes a time series graph, a regional infection level graph, a route network graph, a tentacle network graph and a pre-infection warning graph. The time series graph is used to show the statisticians and the survey subjects the cumulative number of infection cases on the timeline; the regional infection level graph is used to show the statisticians and the survey subjects the cumulative number of infection cases that appear in the map area; the route network graph is used to show the statisticians the route network of the behavioral trajectories of the survey subjects; the tentacle network graph is used to show the statisticians the tentacle network of the contact information of the pre-infected subjects; the pre-infection warning graph is used to show the infection risks and possible infection locations to the pre-infected subjects.

[0099] The specific implementation process is as follows: Figure 7 As shown in the figure, the above-mentioned investigation object behavior trajectory route network is statistically analyzed through the time series bar chart, which can warn the infected exposed population and trace the outbreak time and events of the epidemic; the behavior trajectory route network is represented by the regional infection level map, and the map is distinguished by different colors, which can reduce the contact between the population and the susceptible areas and the population with infection risks, reduce the speed of epidemic transmission, and also remind the disease control personnel of the areas where the investigation and control forces should be concentrated, so as to remove the infection source in time and curb the infection situation; the behavior trajectory route network is directly visualized as a route network map, which can investigate the population along the path, help trace the infection source and infection events, and take external disinfection measures to reduce the spread of the epidemic; the antenna network is constructed in the form of a relationship diagram, which can not only remind the disease control personnel of the pre-infected population with a greater risk of infection, contact the investigation of the epidemic in advance, but also conveniently notify the corresponding personnel through the APP to check and seek medical treatment in time to avoid the occurrence of malignant symptoms; the pre-infection warning map is mainly based on the behavior trajectory route network combined with the infection risk locations marked on the map, and the infection risk and infection range of the epidemic are reflected by the color and size of the circle, which is of great significance for some epidemics transmitted through air and aerosols.

[0100] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0101] The above are only embodiments of the present invention. Well-known specific structures and characteristics in the solution are not described in detail here. Those of ordinary skill in the art know all the common general technical knowledge in the technical field to which the invention pertains before the filing date or the priority date, can know all the prior art in this field, and have the ability to apply the conventional experimental means before this date. Those of ordinary skill in the art can, under the inspiration given in this application, combine their own abilities to complete and implement this solution. Some typical well-known structures or well-known methods should not become obstacles for those of ordinary skill in the art to implement this application. It should be noted that for those skilled in the art, without departing from the structure of the present invention, several deformations and improvements can be made, which should also be regarded as the protection scope of the present invention, and these will not affect the implementation effect of the present invention and the practicality of the patent. The protection scope required by this application should be based on the content of its claims, and the specific implementation manners and the like recorded in the specification can be used to interpret the content of the claims.

Claims

1. A mobile terminal for efficient epidemiological investigations, characterized in that: It includes a collection module, an identification and analysis module, a medical record retrieval module, a trajectory judgment module, an epidemic network module, and a result display module. The collection module is used to collect the basic information of the surveyed object and guide and collect the voice information of the behavior trajectory of the surveyed object. The identification and analysis module is used to identify the residential location of the surveyed object based on the collected information, identify the dialect voice text information of the surveyed object, and mark the location of the surveyed object's address on the map. The medical record retrieval module is used to retrieve the relevant medical record information of the surveyed object through the medical database. The trajectory judgment module is used to use the trained neural network to judge the behavior trajectory route information of the surveyed object on the map within the preset time according to the residential location and voice text information of the surveyed object, make a colloquial note for the map place names, and call the traffic information according to the map route. The epidemic network module is used to construct an epidemic network based on the behavior trajectory route of the surveyed object, call the relevant traffic information and medical record information, and obtain the contact information of other potential objects or the infection risk of the surveyed object. The result display module is used to display the personal survey results and the epidemic survey results according to the constructed epidemic network.

2. The mobile terminal for efficient epidemiological investigation according to claim 1, characterized in that: The collection module includes a touch screen and a non-touch collection unit. The touch screen is used to input the basic information of the surveyed object, including name, gender, age, address, and contact information. The non-touch collection unit is used to verify the identity information of the surveyed object and guide and collect the voice information of the surveyed object.

3. The mobile terminal for efficient epidemiological investigation according to claim 2, characterized in that: The non-touch collection unit includes an RFID reader, a camera, a voice player, and a real-time voice collector. The RFID reader is used to read the identity certificate of the surveyed object to verify the true identity of the surveyed object. The camera is used to verify the true identity of the surveyed object through face recognition. The voice player is used to play the voice to guide the surveyed object to describe the voice information of the behavior trajectory route within the preset time, including place names, road names, road directions, road conditions, and means of transportation. The real-time voice collector is used to collect the voice information of the behavior trajectory described by the surveyed object.

4. The mobile terminal for efficient epidemiological investigation according to claim 1, characterized in that: The identification and analysis module includes an acoustic feature extraction unit, an acoustic model unit, and a map location marking unit. The acoustic feature extraction unit is used to filter and extract the acoustic features in the voice information of the behavior trajectory described by the surveyed object. The acoustic model unit is used to convert the extracted acoustic features into phonemes and identify the voice text information. The acoustic model unit includes several dialect acoustic pronunciation models. The map location marking unit is used to mark the collected residential address of the surveyed object on the map to obtain the address location information.

5. The mobile terminal for efficient epidemiological investigation according to claim 1, characterized in that: The trajectory judgment module includes a judgment unit, a map note unit, and a traffic information call unit. The judgment unit includes a long short-term memory network trained by descriptive text, and is used to judge the true arrival location and route of the surveyed object according to the recognized voice text information to obtain the behavior trajectory route information, and display it on the map. The map note unit is used to make a note of the colloquial place names in the corresponding voice text information after judgment corresponding to the official map names. A traffic information calling unit is used to call the traffic order information after the corresponding means of transportation of the surveyed object's smartphone is taken according to the voice text information and add it to the behavior trajectory route information.

6. The mobile terminal for efficient epidemiological investigation according to claim 5, characterized in that: The long short-term memory network includes an input gate, a hidden layer, a forget gate, and an output gate. The input gate is used to select the current input and the hidden layer unit at the previous time step to update the hidden layer. The hidden layer is used to encode the input feature vector to capture the long-distance dependencies in the encoded sequence. The forget gate is used to select the hidden layer state at the previous time step and discard invalid information. The output gate is used to select the hidden layer to generate an output encoded sequence.

7. The mobile terminal for efficient epidemiological investigation according to claim 6, wherein: The long short-term memory network is optimized by the backpropagation algorithm to minimize the loss function of the network.

8. The mobile terminal for efficient epidemiological investigation according to claim 7, characterized in that: The description text for training the long short-term memory network includes text information on locations, routes, road conditions, and directions.

9. The mobile terminal for efficient epidemiological investigation according to claim 1, characterized in that: The epidemic network module includes an epidemic classification unit, a route network unit, and an antenna network unit. The epidemic classification unit is used to classify the epidemic network according to the types of epidemics based on the relevant medical record information of the surveyed object obtained by the call. The route network unit is used to overlap the routes of several surveyed objects of the same epidemic type on the map according to the behavior trajectory route information to construct a behavior trajectory route network. The antenna network unit is used to obtain pre-infected objects with infection risks based on the behavior trajectory route network and traffic order information, call the order information to obtain the contact information of the pre-infected objects, and construct an antenna network.

10. The mobile terminal for efficient epidemiological investigation according to claim 1, characterized in that: The result display module includes a time series graph, a regional infection level graph, a route network graph, an antenna network graph, and a pre-infection warning graph. The time series graph is used to show the cumulative number of infected cases on the time axis to statisticians and surveyed objects. The regional infection level graph is used to show the cumulative number of infected cases appearing in the map area to statisticians and surveyed objects. The route network graph is used to show the behavior trajectory route network of the surveyed objects to statisticians. The antenna network graph is used to show the antenna network of the contact information of the pre-infected objects to statisticians. The pre-infection warning graph is used to show the infection risk and possible infection locations to pre-infected persons.

Images