CN111681779A - Medical diagnosis system - Google Patents
Medical diagnosis system Download PDFInfo
- Publication number
- CN111681779A CN111681779A CN202010323516.5A CN202010323516A CN111681779A CN 111681779 A CN111681779 A CN 111681779A CN 202010323516 A CN202010323516 A CN 202010323516A CN 111681779 A CN111681779 A CN 111681779A
- Authority
- CN
- China
- Prior art keywords
- characteristic
- module
- voiceprint
- voice data
- symptom
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000003745 diagnosis Methods 0.000 title claims abstract description 93
- 208000024891 symptom Diseases 0.000 claims abstract description 87
- 201000010099 disease Diseases 0.000 claims abstract description 44
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims abstract description 44
- 238000004458 analytical method Methods 0.000 claims abstract description 38
- 238000012549 training Methods 0.000 claims description 57
- 238000012216 screening Methods 0.000 claims description 6
- 238000000034 method Methods 0.000 description 27
- 230000008569 process Effects 0.000 description 11
- 238000005516 engineering process Methods 0.000 description 7
- 201000007100 Pharyngitis Diseases 0.000 description 5
- 206010037660 Pyrexia Diseases 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 230000001755 vocal effect Effects 0.000 description 5
- 206010011224 Cough Diseases 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 3
- 235000006694 eating habits Nutrition 0.000 description 3
- 206010000087 Abdominal pain upper Diseases 0.000 description 2
- 208000003556 Dry Eye Syndromes Diseases 0.000 description 2
- 206010013774 Dry eye Diseases 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 210000005069 ears Anatomy 0.000 description 2
- 208000015181 infectious disease Diseases 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 230000003902 lesion Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 210000000056 organ Anatomy 0.000 description 2
- 230000008447 perception Effects 0.000 description 2
- 230000035479 physiological effects, processes and functions Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 239000013589 supplement Substances 0.000 description 2
- 208000035473 Communicable disease Diseases 0.000 description 1
- 206010013952 Dysphonia Diseases 0.000 description 1
- 208000000059 Dyspnea Diseases 0.000 description 1
- 206010013975 Dyspnoeas Diseases 0.000 description 1
- 206010019233 Headaches Diseases 0.000 description 1
- 208000010473 Hoarseness Diseases 0.000 description 1
- 206010061218 Inflammation Diseases 0.000 description 1
- 208000007117 Oral Ulcer Diseases 0.000 description 1
- 206010035664 Pneumonia Diseases 0.000 description 1
- 206010036790 Productive cough Diseases 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 208000002399 aphthous stomatitis Diseases 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000008451 emotion Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000002650 habitual effect Effects 0.000 description 1
- 210000003128 head Anatomy 0.000 description 1
- 231100000869 headache Toxicity 0.000 description 1
- 230000004054 inflammatory process Effects 0.000 description 1
- 210000004072 lung Anatomy 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 244000000010 microbial pathogen Species 0.000 description 1
- 201000009240 nasopharyngitis Diseases 0.000 description 1
- 230000007935 neutral effect Effects 0.000 description 1
- 230000007170 pathology Effects 0.000 description 1
- 238000012567 pattern recognition method Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- BALXUFOVQVENIU-KXNXZCPBSA-N pseudoephedrine hydrochloride Chemical compound [H+].[Cl-].CN[C@@H](C)[C@@H](O)C1=CC=CC=C1 BALXUFOVQVENIU-KXNXZCPBSA-N 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 208000013220 shortness of breath Diseases 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H80/00—ICT specially adapted for facilitating communication between medical practitioners or patients, e.g. for collaborative diagnosis, therapy or health monitoring
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/06—Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice
- G10L15/063—Training
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L17/00—Speaker identification or verification techniques
- G10L17/04—Training, enrolment or model building
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L17/00—Speaker identification or verification techniques
- G10L17/06—Decision making techniques; Pattern matching strategies
- G10L17/14—Use of phonemic categorisation or speech recognition prior to speaker recognition or verification
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/50—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for simulation or modelling of medical disorders
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Public Health (AREA)
- Pathology (AREA)
- Biomedical Technology (AREA)
- Epidemiology (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Computational Linguistics (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Game Theory and Decision Science (AREA)
- Artificial Intelligence (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
An embodiment of the present application provides a medical diagnosis system, including: the voice data acquisition module is used for acquiring the voice data uploaded by the terminal of the user in an off-line manner; the voice data comprises personal information and symptom description of the user; the voice recognition service module is used for recognizing the voice data into text information through a voice recognition model and sending the text information to the text analysis service module; a text analysis service module for extracting text label features for the personal information and the symptom description from the text information; the voiceprint recognition service module is used for analyzing the voice frequency of the voice data through a voiceprint recognition model to obtain voiceprint symptom characteristics; and the medical diagnosis service module is used for inputting the text label characteristics and the voiceprint symptom characteristics into a medical diagnosis model and outputting a disease name. The embodiment of the application can solve the problem of efficiency of effective allocation of medical resources, and has high diagnosis accuracy.
Description
Technical Field
The present application relates to the field of data processing technologies, and in particular, to a medical diagnosis system.
Background
At present, the efficiency of the traditional offline medical registration/appointment inquiry means is low, and a patient needs to go to a hospital site to see a doctor for a doctor, so that the method is not beneficial to infectious disease epidemic situation prevention and control, and consumes time and energy. In order to solve the problem of offline inquiry efficiency, the form of internet inquiry is adopted, so that the working efficiency can be improved, and the infection risk can be reduced.
The existing Internet inquiry implementation scheme is as follows: the patient inputs personal information to log in a patient terminal, an information acquisition module acquires disease condition characteristic information of the patient and sends the disease condition characteristic information and the personal information to a cloud inquiry platform, the cloud inquiry platform establishes a first mapping relation between the personal information of the patient and the disease condition characteristic information and generates information containing the patient, the cloud inquiry platform sends the patient information to a doctor terminal, and the doctor acquires the patient information through the doctor terminal and diagnoses the patient information. However, the disadvantage of this inquiry method is that the relationship between the correctness of the diagnosis result and the medical level or attitude of the doctor is large, and the on-line inquiry is virtual, if no third party supervises, the result may be not objective and accurate, and if more than 90% of the on-line conversations are counted, the doctor does not respond in time, so that the patient leaves the conversation, and the manual inquiry is meaningless. While excellent medical resources are limited, this inevitably causes problems of online queuing or increases labor costs.
Disclosure of Invention
In view of the above, embodiments of the present application are proposed to provide a medical diagnostic system that overcomes or at least partially solves the above-mentioned problems.
The embodiment of the application discloses a medical diagnosis system, including: the system comprises a user terminal, a voice data acquisition module, a voice recognition service module, a voiceprint recognition service module, a text analysis service module and a medical diagnosis module;
the voice data acquisition module is used for acquiring the voice data uploaded by the terminal of the user in an off-line manner and respectively sending the voice data to the voice recognition service module and the voiceprint recognition service module; the voice data comprises personal information and symptom description of the user;
the voice recognition service module is used for recognizing the voice data into text information through a voice recognition model and sending the text information to the text analysis service module;
the text analysis service module is used for extracting text label features aiming at the personal information and the symptom description from the text information;
the voiceprint recognition service module is used for analyzing the voice frequency of the voice data through a voiceprint recognition model to obtain voiceprint symptom characteristics;
and the medical diagnosis service module is used for inputting the text label characteristics and the voiceprint symptom characteristics into a medical diagnosis model and outputting a disease name.
Optionally, the voice data obtaining module includes:
an audio storage service submodule and a streaming media service submodule, wherein:
the audio storage service sub-module is used for receiving and storing the voice data uploaded by the user terminal before disconnection after the user terminal is disconnected;
the streaming media service sub-module is used for extracting and obtaining the voice data from the audio storage service sub-module and sending the voice data to the voice recognition service module and the voiceprint recognition service module respectively.
Optionally, the method further includes:
the personal characteristic judgment service module is respectively connected with the voiceprint recognition service module and the text analysis service module; wherein:
the personal characteristic judgment service module is used for screening the text label characteristic and the voiceprint symptom characteristic to obtain personal characteristic information and writing the personal characteristic information into the database;
the database is used for storing the personal characteristic information;
the medical diagnosis service module is also used for extracting and obtaining the personal characteristic information from the database, inputting the personal characteristic information into a medical diagnosis model and outputting a disease name.
Optionally, a plurality of characteristic slot positions corresponding to the personal characteristic information are set in the database, and the characteristic slot positions have slot names;
the personal characteristic judgment service module comprises:
a slot value determination submodule and a characteristic slot position filling submodule, wherein:
the slot value determining submodule is used for judging whether the text label characteristic and the voiceprint symptom characteristic accord with the slot value corresponding to the slot name of the characteristic slot position or not;
the characteristic slot position filling sub-module is used for filling the text label characteristic into the characteristic slot position when the text label characteristic and the voiceprint symptom characteristic both accord with the slot value; and filling the voiceprint symptom feature into the feature slot when only the voiceprint symptom feature matches the slot value.
Optionally, the medical diagnosis service module includes:
a diagnostic model determination submodule and a medical diagnosis execution submodule, wherein:
the diagnosis model determining submodule is used for obtaining the region characteristics and determining the medical diagnosis model corresponding to the region characteristics; the regional characteristic belongs to at least one of the text label characteristic and the voiceprint symptom characteristic;
and the medical diagnosis execution submodule is used for inputting the text label characteristic and the voiceprint symptom characteristic into a medical diagnosis model corresponding to the region characteristic and outputting a disease name.
Optionally, the medical diagnosis performing sub-module includes:
a confidence determination submodule and a condition name output submodule, wherein:
the confidence coefficient determining submodule is used for determining the confidence coefficient of the disease name;
and the disease name output submodule is used for outputting the disease names with the confidence degrees larger than a preset threshold value.
Optionally, the method further includes:
speech recognition model training module and voiceprint recognition model training module, wherein:
the voice recognition model training module is used for acquiring the voice data uploaded by all the user terminals from the voice data acquisition module, and training a first preset model based on the voice data uploaded by all the user terminals to obtain the voice recognition model;
the voiceprint recognition model training module is used for acquiring the voice data uploaded by all the user terminals from the voice data acquisition module, and training a second preset model based on the voice data uploaded by all the user terminals to obtain the voiceprint recognition model.
Optionally, the method further includes:
a training sample set obtaining module and a diagnostic model training module, wherein:
the training sample set obtaining module is used for obtaining training sample sets of different regions, each training sample set comprises a plurality of training samples, and each training sample comprises a symptom characteristic and a symptom name with a corresponding relationship;
and the diagnosis model training module is used for training a third preset model based on the training sample set to obtain a medical diagnosis model of a corresponding region.
The embodiment of the application has the following advantages:
according to the method and the system for uploading the voice data offline, the system does not need to receive the voice data in real time or process the voice data in real time, and the patient user can accept that the returned result is not real-time subconsciously, so that online queuing of the patient user can be avoided, the efficiency problem of effective distribution of medical resources is solved, and the problem that short conversation interaction audio time is not beneficial to voiceprint feature analysis is also avoided;
the voiceprint symptom characteristics of the user are extracted and used as auxiliary characteristics to supplement information missing from the text label characteristics, the correctness of a diagnosis result can be further improved, and compared with an existing manual on-line inquiry mode, the diagnosis data are more objective;
according to the embodiment of the application, the medical diagnosis models in different regions are distinguished through the region characteristics in the personal characteristic information of the user, all the characteristic information of the patient is subjected to medical diagnosis based on the medical diagnosis models corresponding to the region characteristics, the disease of the user can be combined with the factors such as the living environment, the eating habits or the spread range of epidemic diseases of the region where the disease of the user is located, and the diagnosis accuracy is further improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the description of the embodiments of the present application will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without inventive exercise.
Fig. 1 is a schematic structural diagram of a medical diagnostic system according to an embodiment of the present application;
FIG. 2 is an off-line medical diagnostic flow diagram of the system of an embodiment of the present application;
fig. 3 is a block diagram of an off-line medical interrogation of the system of an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some, but not all, embodiments of the present application. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The terms "comprises," "comprising," or any other variation thereof, in the description and claims of this application, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
In the embodiments of the present invention, words such as "exemplary" or "for example" are used to mean serving as examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "e.g.," an embodiment of the present invention is not necessarily to be construed as preferred or advantageous over other embodiments or designs. Rather, use of the word "exemplary" or "such as" is intended to present concepts related in a concrete fashion.
Aiming at the technical problem of the medical diagnosis system, under the condition that a patient is inconvenient to input characters or on-line medical resources are in shortage and the like, the system analyzes the voice data uploaded by the patient based on the technology of combining voiceprint voice feature analysis, voice recognition and text analysis, the patient can obtain a medical diagnosis result through a section of voice data comprising personal information and symptom description uploaded to the system, and the problem of efficiency of effective allocation of the medical resources is solved.
Referring to fig. 1, a schematic structural diagram of a medical diagnosis system according to an embodiment of the present application is shown, which may include: the system comprises a user terminal, a voice data acquisition module, a voice recognition service module, a voiceprint recognition service module, a text analysis service module and a medical diagnosis module;
specifically, the terminal of the user is connected with a voice data acquisition module, the voice data acquisition module is respectively connected with a voice recognition service module and a voiceprint recognition service module, the voice recognition service module is respectively connected with a text analysis service module, and the medical diagnosis module can be directly connected with the text analysis service module and the voiceprint recognition service module or indirectly connected with the text analysis service module and the voiceprint recognition service module;
the voice data acquisition module is used for acquiring the voice data uploaded by the terminal of the user in an off-line manner and respectively sending the voice data to the voice recognition service module and the voiceprint recognition service module; the voice data comprises personal information and symptom description of the user;
the voice data obtaining module comprises an audio storage service sub-module and a streaming media service sub-module, and the execution function of the voice data obtaining module can be realized through the audio storage service sub-module and the streaming media service sub-module. In one embodiment of the present application, the respective roles of the audio storage service submodule and the streaming media service submodule in the medical diagnosis system are provided:
the audio storage service sub-module is used for receiving and storing the voice data uploaded by the user terminal before disconnection after the user terminal is disconnected;
the streaming media service sub-module is used for extracting and obtaining the voice data from the audio storage service sub-module and sending the voice data to the voice recognition service module and the voiceprint recognition service module respectively.
The terminal of the user can be a client or a WEB end provided by the system for the user to use, and the user can access the system through application software APP installed on a mobile phone, a tablet computer or a smart watch, or access the system through a browser of a computer. The audio storage service sub-module may preferably be a disk array storage server with large capacity and good disaster recovery effect, and the streaming media service sub-module may specifically be a streaming media server.
Offline upload may refer to: after receiving the voice which is recorded by the user and related to personal information and symptom description, the terminal of the user automatically stores the voice as voice data and uploads the voice data to the system, the user can close the terminal (namely quit the client or the WEB terminal) after recording the voice, at the moment, the connection between the terminal of the user and the system is disconnected, and then the system receives and stores the voice data.
The personal information may refer to information such as age, sex, region, etc., and the symptom description may refer to a description of the user's current physical state, including duration, degree of state, etc., of the physical state, such as "XX times, some coughs, expectoration, chills and fever, etc. are found". Of course, the personal information of the present application can also be obtained by identifying the terminal of the user, but in practice, the user may not fill in the actual personal information of the user, such as age and gender, when registering the system, and the same symptom may have different diagnosis results for people of different ages or different genders. Therefore, in practice, the user generally needs to introduce his/her personal information in a voice manner.
The voice recognition service module is used for recognizing the voice data into text information through a voice recognition model and sending the text information to the text analysis service module;
the voice recognition model recognizes voice data based on a voice recognition (ASR) technology, and converts the voice data into text information. It should be noted that, the speech recognition model in the embodiment of the present application does not adopt a conventional ASR model, and the inventor separately sets a speech recognition model training module to obtain the speech recognition model of the present application, specifically:
the voice recognition model training module is used for acquiring the voice data uploaded by all the user terminals from the voice data acquisition module, and training a first preset model based on the voice data uploaded by all the user terminals to obtain the voice recognition model.
In practice, a plurality of patients exist, a plurality of terminals of users connected with the system exist, the voice data uploaded by all the terminals of the users are used as a training set to train a first preset model, a voice recognition model special for medical inquiry is formed, and accuracy of voice recognition on symptom description and personal information of the patients can be improved. The first preset model may specifically refer to an acoustic model and a language model generated by performing a large amount of training on psychology, physiology, acoustics, linguistics, information theory, signal processing, computer science, and a pattern recognition method based on deep learning models such as HMMs and DNNs on the basis of a speech database and a language database, and may convert human speech into text.
The text analysis service module is used for extracting text label features aiming at the personal information and the symptom description from the text information;
text refers to a structure of information composed of certain symbols or symbols, and the structure can take different forms of expression, such as linguistic, textual, visual, etc. Text is made by a particular person, and the semantics of the text inevitably reflect the person's particular standpoint, perspective, value and benefit. Therefore, the text analysis service module is used for receiving the text information sent by the voice recognition service module, analyzing the text information, extracting characteristic words for the personal information and the symptom description from the text information by using natural semantic understanding and big data analysis technologies, and using the characteristic words as text label characteristics, so that the intention and the purpose of the user can be inferred.
For example, the text label features may be "age," "region," "cough," "shortness of breath," "fever," and the like.
The voiceprint recognition service module is used for analyzing the voice frequency of the voice data through a voiceprint recognition model to obtain voiceprint symptom characteristics;
the voiceprint recognition model recognizes the voice frequency of voice data based on a voiceprint characteristic analysis (VPR) technology, converts a voice signal into an electric signal, and then recognizes the electric signal by using a computer to obtain the voiceprint symptom characteristic of the application.
The generation of human language is a complex physiological and physical process between human language center and pronunciation organ, and the voiceprint atlas of the same person is different under different body states. The voice acoustic characteristics of each person are both relatively stable and variable, not absolute, but invariable, and the variation can come from physiology, pathology and psychology. Therefore, the voice characteristics can be extracted, and the characteristic parameters capable of representing the specific organ structure or the habitual behaviors of the speaker can be extracted from the voice of the speaker by the voiceprint characteristic analysis technology. The characteristic parameter can mainly adopt MFCC (Mel frequency cepstrum), which is a short-time frequency spectrum characteristic, and the characteristic parameter is generated on the basis of the nonlinear perception of human ears on sound frequency, and the sensitivity of human ears on sound perception changes along with the change of frequency.
It should be noted that, the voiceprint recognition model in the embodiment of the present application may also be trained separately, and the system further includes a voiceprint recognition model training module to train and obtain the voiceprint recognition model of the present application, specifically:
and the voiceprint recognition model training module is used for acquiring the voice data uploaded by all the user terminals from the audio storage service submodule, and training a second preset model based on the voice data uploaded by all the user terminals to obtain the voiceprint recognition model.
Referring to the training process of the voice recognition model, the embodiment of the application also specially adopts the voice data of a plurality of patients as a training set to train the second preset model, so that the voiceprint recognition model special for medical inquiry can be obtained, and the accuracy of detecting diseases through the voice of the patients is improved. The second preset model is not limited in the embodiment of the application, and an existing GMM-UBM model, DNN i-vector model, or the like may be used.
For example, the vocal print symptom characteristic may be "age, gender, emotion," territory, "lesion," etc. of the patient.
And the medical diagnosis service module is used for inputting the text label characteristics and the voiceprint symptom characteristics into a medical diagnosis model and outputting a disease name.
The medical diagnosis model is trained in advance, a large number of electronic cases or labeled characteristic disease symptoms-disease names corresponding tables can be adopted for training, and modeling is not limited to a certain deep learning method, such as CNN, DNN and the like. And after the text label characteristic and the voiceprint symptom characteristic are input into the medical diagnosis model, the disease name can be automatically output and sent to a terminal of a user, so that the on-line medical diagnosis is realized.
According to the method and the system for uploading the voice data offline, the system does not need to receive the voice data in real time and process the voice data in real time, the returned result can be accepted in subconscious of the patient user and is not real-time, so that online queuing of the patient user can be avoided, the problem of efficiency of effective distribution of medical resources is solved, and meanwhile, the problem that short conversation interaction audio time is not beneficial to voiceprint feature analysis is also avoided. In addition, the inventor considers that the patient may not be able to provide the audio frequency corresponding to the content completely according to the requirement, and the information is lost, so the voiceprint symptom characteristic of the user is also extracted to be used as the auxiliary characteristic to supplement the lost information, the correctness of the diagnosis result can be further improved, and compared with the existing manual on-line inquiry mode, the diagnosis data is more objective.
In an optional embodiment of the present application, the system may further include a personal characteristic determination service module and a database, where the personal characteristic determination service module is connected to the voiceprint recognition service module and the text analysis service module, respectively; wherein:
the personal characteristic judgment service module is used for screening the text label characteristic and the voiceprint symptom characteristic to obtain personal characteristic information and writing the personal characteristic information into the database;
the database is used for storing the personal characteristic information;
the medical diagnosis service module is also used for extracting and obtaining the personal characteristic information from the database, inputting the personal characteristic information into a medical diagnosis model and outputting a disease name.
Because the text label characteristics and the voiceprint symptom characteristics are the same and similar, the personal characteristic judgment service module can verify the similar or similar characteristics between the text label characteristics and the voiceprint symptom characteristics by screening the text label characteristics and the voiceprint symptom characteristics, so that the problems that the accuracy of information input into a medical diagnosis model is reduced due to voiceprint recognition errors, and finally the output disease name is deviated are solved. Meanwhile, after screening, the same and similar features between the text label features and the voiceprint symptom features can be deduplicated to obtain personal feature information, and the training amount of the medical diagnosis model can be reduced while the accuracy of the personal feature information is ensured.
The screened personal characteristic information is stored in a database, and the medical diagnosis service module can extract the personal characteristic information from the database at any time and process the personal characteristic information of a plurality of users according to a certain sequence or synchronization, so that the problem of the efficiency of effective allocation of medical resources is solved.
Specifically, a plurality of characteristic slot positions corresponding to the personal characteristic information are arranged in the database, and the characteristic slot positions have slot names;
the personal characteristic judgment service module can comprise a slot value determining submodule and a characteristic slot position filling submodule, and the mode of screening the text label characteristic and the voiceprint symptom characteristic is as follows:
the slot value determining submodule is used for judging whether the text label characteristic and the voiceprint symptom characteristic accord with the slot value corresponding to the slot name of the characteristic slot position or not;
the characteristic slot position filling sub-module is used for filling the text label characteristic into the characteristic slot position when the text label characteristic and the voiceprint symptom characteristic both accord with the slot value; and filling the voiceprint symptom feature into the feature slot when only the voiceprint symptom feature matches the slot value.
When storing the personal characteristic information, the database stores the personal characteristic information in the form of a table, wherein the characteristic slot refers to a plurality of attribute lists of the table of the personal characteristic information, and can be set as gender, age, region, symptom, focus and the like. The slot name of the characteristic slot can be understood as the header of a plurality of attributes of the table of the "personal characteristic information", such as "sex", "age", "region", "symptom", "focus", and the like. The slot value refers to information filled in the characteristic slot position, for example, the slot value corresponding to the gender can be male; the trough value corresponding to "symptoms" may be "head fever" and the trough value corresponding to "lesions" may be "throat inflammation". It should be noted that a plurality of identical slot names may be provided for filling different slot values, and the identical slot names are numbered for distinction. For example, the number of the slot names of "symptom" may be plural, and may be respectively expressed as "symptom 1", "symptom 2", "symptom 3", etc., and the slot values filled in the different slot names may be "fever", "headache", "cough", etc.; different slot names may have an association relationship, for example, the slot name of "symptom" and the slot name of "focus" may be associated with each other, which indicates that the slot values filled in the characteristic slot are also associated with each other, as shown in the following example: the symptom of "hoarseness" is related to the focus of "inflammation of the throat".
Whether the text label feature and the voiceprint symptom feature conform to the slot value corresponding to the slot name of the feature slot or not is judged, and judgment can be performed by adopting a related existing technology, for example, judgment can be performed according to attribute confidence degrees between the text label feature and the voiceprint symptom feature and the slot name, and the embodiment of the application is not limited herein.
When the text label feature and the voiceprint symptom feature both conform to the slot value, since the confidence of the text label feature is greater than the confidence of the voiceprint symptom feature, the text label feature is preferably used to fill the feature slot. For example, the patient says that the sex is male in voice, namely the text label feature comprises 'male', and the voiceprint analysis also analyzes that the provider of the voice data is male, and at the moment, the selection of storage is possible; however, if the patient's voice is off-neutral, the results of the voiceprint analysis may be inaccurate, and therefore, text-tagged features are preferred for population.
In practice, the text label features have many pieces of key information which cannot be reflected, such as a focus, which is a part of an organism which is particularly diseased, and is a limited diseased tissue with pathogenic microorganisms, and a user is generally unclear and cannot accurately express the information through voice. The embodiment of the present application can be obtained by voiceprint analysis, and is embodied in the characteristics of voiceprint symptoms. The following are exemplified: the vocal print symptom characteristic of the focus on sound is obtained through the vocal print analysis, and the vocal print symptom characteristic of the throat inflammation only accords with the groove value corresponding to the focus, so that the vocal print symptom characteristic of the throat inflammation is filled into the characteristic groove position corresponding to the focus. By filling the voiceprint symptom features into the feature slot positions under the slot names, the personal feature information of the patient can be further improved.
In addition, the inventor of the application finds that similar symptoms may correspond to different diseases due to different living environments, eating habits or epidemic spread ranges and other factors in each region; in an alternative embodiment of the present application, the inventor further improves the medical diagnosis model, and the system of the present application further includes a training sample set obtaining module and a diagnosis model training module, wherein:
the training sample set obtaining module is used for obtaining training sample sets of different regions, each training sample set comprises a plurality of training samples, and each training sample comprises a symptom characteristic and a symptom name with a corresponding relationship;
and the diagnosis model training module is used for training a third preset model based on the training sample set to obtain a medical diagnosis model of a corresponding region.
The embodiment of the application provides a scheme for distinguishing regions for training, a training sample set with corresponding disease symptoms and disease names in the same region is collected, a third preset model is trained by using the training sample set, and a medical diagnosis model suitable for the region can be obtained. The modeling of the third preset model is not limited to a deep learning method, such as CNN, DNN, etc.
Based on medical diagnosis models of different regions, the medical diagnosis service module of the embodiment of the present application may include a diagnosis model determination sub-module and a medical diagnosis execution sub-module, wherein:
the diagnosis model determining submodule is used for obtaining the region characteristics and determining the medical diagnosis model corresponding to the region characteristics; the regional characteristic belongs to at least one of the text label characteristic and the voiceprint symptom characteristic;
and the medical diagnosis execution submodule is used for inputting the text label characteristic and the voiceprint symptom characteristic into a medical diagnosis model corresponding to the region characteristic and outputting a disease name.
In the embodiment of the present application, the regional characteristic may be a province, a city, a county, or the like, or may be a region, such as a northeast region, a southern Fujian region, a southwest region, or the like, and the present application is not limited herein. In practice, the regional features may be self-explained by the user through voice, or extracted from the voiceprint of the user by the system.
If the extracted regional characteristic is 'region A', the system can determine and call out a 'medical diagnosis model of region A', then input the personal characteristic information of the user into the 'medical diagnosis model of region A', and finally output a more accurate disease name. According to the embodiment of the application, the disease of the user can be combined with factors such as the living environment of the region where the disease is located, the eating habits or the spread range of epidemic diseases, and the diagnosis accuracy can be further improved.
In an optional embodiment of the present application, the medical diagnosis performing sub-module may further include a confidence level determining sub-module and a disease name outputting sub-module, wherein:
the confidence coefficient determining submodule is used for determining the confidence coefficient of the disease name;
and the disease name output submodule is used for outputting the disease names with the confidence degrees larger than a preset threshold value.
In practice, there may be some similar symptoms for some names of conditions, exemplified by the personal characteristics of "cough", "fever", etc., but one is the common cold and the other is pneumonia. Therefore, in the embodiment of the present application, a threshold of the confidence level is set for the output disease name, for example, the preset threshold may be set to 0.5, and only the disease name with the confidence level greater than the preset threshold is output to the terminal of the user, so that the diagnosis accuracy can be further improved. Particularly, the disease names can be output to the user according to the sequence of the confidence level, so that the patient can know the real disease of the patient conveniently.
In summary, referring to fig. 2, an off-line medical diagnosis flowchart of the system of the embodiment of the present application is shown, and fig. 3 is a diagram of an off-line medical inquiry structure of the system of the embodiment of the present application. Next, with reference to fig. 2 and fig. 3, an offline diagnosis process of the medical diagnosis system according to the embodiment of the present application will be further described.
In fig. 3, the user terminal is a client, the audio storage service sub-module is implemented by an audio storage service, the streaming media service sub-module is implemented by an audio streaming media service, the speech recognition service module is implemented by an ASR speech recognition service, the text analysis service module is implemented by a TA text analysis service, the voiceprint recognition service module is implemented by a VPR voiceprint feature analysis service, the personal feature judgment service module is implemented by a personal feature judgment service, the database is specifically a personal feature information database, and the medical diagnosis service module has medical diagnosis models of different regions, such as a region a medical diagnosis model, a region B medical diagnosis model, a region C medical diagnosis model, a region D medical diagnosis model and the like.
The method comprises the steps that a patient uploads a section of audio (voice data) with personal information and current personal symptom description at a client, the audio is stored in a disk array storage server of the system, ASR voice recognition service and VPR voiceprint feature analysis service respectively take the audio from the audio storage server through audio streaming media service to perform voice-to-text and voiceprint feature analysis, TA text analysis service takes text results transcribed by the ASR voice recognition service to perform text label analysis, and text label features with the personal information and the symptoms are extracted. The personal characteristic judgment service obtains the text label characteristic and the voiceprint symptom characteristic, sets all required characteristic slot positions, wherein the slot names can be age, gender, region, focus, symptom and the like, preferentially selects the text label characteristic as personal characteristic information to be filled and stored if the text label characteristic and the voiceprint symptom characteristic have all the characteristics, and the personal characteristic information is stored in a personal characteristic information database. And distinguishing medical diagnosis models in different regions according to region features in the personal feature information of the patient, performing medical diagnosis on all feature information of the patient based on the medical diagnosis models corresponding to the region features, outputting a suspected disease name list with confidence coefficient greater than 0.5, and completing the whole inquiry flow.
For example, the audio content may be "i am 26 years old, gender maid. At 8 am today, the mouth is slightly painful, eyes are dry, the voice is accompanied by Sichuan accent, and the voice is somewhat hoarse. The ASR speech recognition service converts the audio into words, and the text label features extracted after being analyzed by the TA text analysis service comprise: "26 years old", "female", "8 am on 1/2/2020", "stomachache", "dry eyes"; the voiceprint symptom features extracted by the VPR voiceprint feature analysis service after audio analysis comprise: "Sichuan", "throat inflammation" and "neutral sound". After the personal characteristic judgment service analyzes the text label characteristic and the voiceprint symptom characteristic, the personal characteristic information stored in the personal characteristic information database comprises the following steps: "26 years old", "female", "8 am on 1/2/2020", "stomachache", "dry eyes", "Sichuan" and "inflammation of throat". After personal characteristic information is input into a medical diagnosis model of Sichuan, the name list of suspected diseases with confidence coefficient higher than 0.5, which may include 'lung heat flaming up' and 'oral ulcer', is obtained by considering that Sichuan people are spicy and heavy in moisture.
It should be noted that the system embodiments are described as a series of acts or combinations for simplicity in description, but those skilled in the art will recognize that the embodiments are not limited by the described order of acts, as some steps may be performed in other orders or simultaneously according to the embodiments. Further, those skilled in the art will also appreciate that the embodiments described in the specification are presently preferred and that no particular act is required of the embodiments of the application.
The embodiments in the present specification are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other.
While preferred embodiments of the present application have been described, additional variations and modifications of these embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including the preferred embodiment and all such alterations and modifications as fall within the true scope of the embodiments of the application.
Finally, it should also be noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or terminal that comprises the element.
The above detailed description is provided for a medical diagnosis system provided by the present application, and the principles and embodiments of the present application are explained in detail by using specific examples, and the descriptions of the above examples are only used to help understanding the method and the core ideas of the present application; meanwhile, for a person skilled in the art, according to the idea of the present application, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present application.
Claims (8)
1. A medical diagnostic system, comprising: the system comprises a user terminal, a voice data acquisition module, a voice recognition service module, a voiceprint recognition service module, a text analysis service module and a medical diagnosis module;
the voice data acquisition module is used for acquiring the voice data uploaded by the terminal of the user in an off-line manner and respectively sending the voice data to the voice recognition service module and the voiceprint recognition service module; the voice data comprises personal information and symptom description of the user;
the voice recognition service module is used for recognizing the voice data into text information through a voice recognition model and sending the text information to the text analysis service module;
the text analysis service module is used for extracting text label features aiming at the personal information and the symptom description from the text information;
the voiceprint recognition service module is used for analyzing the voice frequency of the voice data through a voiceprint recognition model to obtain voiceprint symptom characteristics;
and the medical diagnosis service module is used for inputting the text label characteristics and the voiceprint symptom characteristics into a medical diagnosis model and outputting a disease name.
2. The system of claim 1, wherein the voice data obtaining module comprises:
an audio storage service submodule and a streaming media service submodule, wherein:
the audio storage service sub-module is used for receiving and storing the voice data uploaded by the user terminal before disconnection after the user terminal is disconnected;
the streaming media service sub-module is used for extracting and obtaining the voice data from the audio storage service sub-module and sending the voice data to the voice recognition service module and the voiceprint recognition service module respectively.
3. The system of claim 1, further comprising:
the personal characteristic judgment service module is respectively connected with the voiceprint recognition service module and the text analysis service module; wherein:
the personal characteristic judgment service module is used for screening the text label characteristic and the voiceprint symptom characteristic to obtain personal characteristic information and writing the personal characteristic information into the database;
the database is used for storing the personal characteristic information;
the medical diagnosis service module is also used for extracting and obtaining the personal characteristic information from the database, inputting the personal characteristic information into a medical diagnosis model and outputting a disease name.
4. The system of claim 3, wherein a plurality of characteristic slots corresponding to the personal characteristic information are provided in the database, the characteristic slots having slot names;
the personal characteristic judgment service module comprises:
a slot value determination submodule and a characteristic slot position filling submodule, wherein:
the slot value determining submodule is used for judging whether the text label characteristic and the voiceprint symptom characteristic accord with the slot value corresponding to the slot name of the characteristic slot position or not;
the characteristic slot position filling sub-module is used for filling the text label characteristic into the characteristic slot position when the text label characteristic and the voiceprint symptom characteristic both accord with the slot value; and filling the voiceprint symptom feature into the feature slot when only the voiceprint symptom feature matches the slot value.
5. The system of claim 1 or 3, wherein the medical diagnostic service module comprises:
a diagnostic model determination submodule and a medical diagnosis execution submodule, wherein:
the diagnosis model determining submodule is used for obtaining the region characteristics and determining the medical diagnosis model corresponding to the region characteristics; the regional characteristic belongs to at least one of the text label characteristic and the voiceprint symptom characteristic;
and the medical diagnosis execution submodule is used for inputting the text label characteristic and the voiceprint symptom characteristic into a medical diagnosis model corresponding to the region characteristic and outputting a disease name.
6. The system of claim 5, wherein the medical diagnosis execution submodule comprises:
a confidence determination submodule and a condition name output submodule, wherein:
the confidence coefficient determining submodule is used for determining the confidence coefficient of the disease name;
and the disease name output submodule is used for outputting the disease names with the confidence degrees larger than a preset threshold value.
7. The system of claim 1, further comprising:
speech recognition model training module and voiceprint recognition model training module, wherein:
the voice recognition model training module is used for acquiring the voice data uploaded by all the user terminals from the voice data acquisition module, and training a first preset model based on the voice data uploaded by all the user terminals to obtain the voice recognition model;
the voiceprint recognition model training module is used for acquiring the voice data uploaded by all the user terminals from the voice data acquisition module, and training a second preset model based on the voice data uploaded by all the user terminals to obtain the voiceprint recognition model.
8. The system of claim 5, further comprising:
a training sample set obtaining module and a diagnostic model training module, wherein:
the training sample set obtaining module is used for obtaining training sample sets of different regions, each training sample set comprises a plurality of training samples, and each training sample comprises a symptom characteristic and a symptom name with a corresponding relationship;
and the diagnosis model training module is used for training a third preset model based on the training sample set to obtain a medical diagnosis model of a corresponding region.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010323516.5A CN111681779A (en) | 2020-04-22 | 2020-04-22 | Medical diagnosis system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010323516.5A CN111681779A (en) | 2020-04-22 | 2020-04-22 | Medical diagnosis system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111681779A true CN111681779A (en) | 2020-09-18 |
Family
ID=72433602
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010323516.5A Pending CN111681779A (en) | 2020-04-22 | 2020-04-22 | Medical diagnosis system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111681779A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112315703A (en) * | 2020-09-21 | 2021-02-05 | 河南大学淮河医院 | New coronary pneumonia patient nursing system based on cloud server tracing system |
| CN112885334A (en) * | 2021-01-18 | 2021-06-01 | 吾征智能技术(北京)有限公司 | Disease recognition system, device, storage medium based on multi-modal features |
| CN112908339A (en) * | 2021-03-18 | 2021-06-04 | 龙马智芯（珠海横琴）科技有限公司 | Conference link positioning method and device, positioning equipment and readable storage medium |
| CN113724898A (en) * | 2021-08-31 | 2021-11-30 | 平安科技（深圳）有限公司 | Intelligent inquiry method, device, equipment and storage medium |
| CN113744897A (en) * | 2021-08-30 | 2021-12-03 | 康键信息技术（深圳）有限公司 | Network inquiry method, computer device and storage medium |
| CN114188041A (en) * | 2021-12-02 | 2022-03-15 | 深圳市华美泰科技发展有限公司 | Medical system for completing doctor-patient service in remote dialogue mode |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050102145A1 (en) * | 2002-11-11 | 2005-05-12 | Kakuichi Shiomi | Psychosomatic diagnosis system |
| US20070118399A1 (en) * | 2005-11-22 | 2007-05-24 | Avinash Gopal B | System and method for integrated learning and understanding of healthcare informatics |
| CN106709254A (en) * | 2016-12-29 | 2017-05-24 | 天津中科智能识别产业技术研究院有限公司 | Medical diagnostic robot system |
| CN107247868A (en) * | 2017-05-18 | 2017-10-13 | 深思考人工智能机器人科技（北京）有限公司 | A kind of artificial intelligence aids in interrogation system |
| CN108198620A (en) * | 2018-01-12 | 2018-06-22 | 洛阳飞来石软件开发有限公司 | A kind of skin disease intelligent auxiliary diagnosis system based on deep learning |
| CN109285605A (en) * | 2018-09-11 | 2019-01-29 | 青岛农业大学 | A kind of medical diagnosis system and diagnostic method based on artificial intelligence |
-
2020
- 2020-04-22 CN CN202010323516.5A patent/CN111681779A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050102145A1 (en) * | 2002-11-11 | 2005-05-12 | Kakuichi Shiomi | Psychosomatic diagnosis system |
| US20070118399A1 (en) * | 2005-11-22 | 2007-05-24 | Avinash Gopal B | System and method for integrated learning and understanding of healthcare informatics |
| CN106709254A (en) * | 2016-12-29 | 2017-05-24 | 天津中科智能识别产业技术研究院有限公司 | Medical diagnostic robot system |
| CN107247868A (en) * | 2017-05-18 | 2017-10-13 | 深思考人工智能机器人科技（北京）有限公司 | A kind of artificial intelligence aids in interrogation system |
| CN108198620A (en) * | 2018-01-12 | 2018-06-22 | 洛阳飞来石软件开发有限公司 | A kind of skin disease intelligent auxiliary diagnosis system based on deep learning |
| CN109285605A (en) * | 2018-09-11 | 2019-01-29 | 青岛农业大学 | A kind of medical diagnosis system and diagnostic method based on artificial intelligence |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112315703A (en) * | 2020-09-21 | 2021-02-05 | 河南大学淮河医院 | New coronary pneumonia patient nursing system based on cloud server tracing system |
| CN112885334A (en) * | 2021-01-18 | 2021-06-01 | 吾征智能技术(北京)有限公司 | Disease recognition system, device, storage medium based on multi-modal features |
| CN112908339A (en) * | 2021-03-18 | 2021-06-04 | 龙马智芯（珠海横琴）科技有限公司 | Conference link positioning method and device, positioning equipment and readable storage medium |
| CN113744897A (en) * | 2021-08-30 | 2021-12-03 | 康键信息技术（深圳）有限公司 | Network inquiry method, computer device and storage medium |
| CN113724898A (en) * | 2021-08-31 | 2021-11-30 | 平安科技（深圳）有限公司 | Intelligent inquiry method, device, equipment and storage medium |
| CN113724898B (en) * | 2021-08-31 | 2024-05-07 | 平安科技（深圳）有限公司 | Intelligent inquiry method, device, equipment and storage medium |
| CN114188041A (en) * | 2021-12-02 | 2022-03-15 | 深圳市华美泰科技发展有限公司 | Medical system for completing doctor-patient service in remote dialogue mode |
| CN114188041B (en) * | 2021-12-02 | 2022-11-15 | 深圳市华美泰科技发展有限公司 | Medical system for completing doctor-patient service in remote dialogue mode |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111681779A (en) | Medical diagnosis system | |
| US11322155B2 (en) | Method and apparatus for establishing voiceprint model, computer device, and storage medium | |
| JP6263308B1 (en) | Dementia diagnosis apparatus, dementia diagnosis method, and dementia diagnosis program | |
| CN112750465B (en) | Cloud language ability evaluation system and wearable recording terminal | |
| CN111329494B (en) | Depression reference data acquisition method and device | |
| JP2021508797A (en) | Smart toilet and appliance system | |
| CN113035232B (en) | Psychological state prediction system, method and device based on voice recognition | |
| Fletcher et al. | Assessing vowel centralization in dysarthria: A comparison of methods | |
| CN116343824B (en) | Comprehensive evaluation and solution method, system, device and medium for talent expression capability | |
| JP5335668B2 (en) | Computer-aided pronunciation learning support method using computers applicable to various languages | |
| Wambaugh et al. | Interrater reliability and concurrent validity for the Apraxia of Speech Rating Scale 3.0: Application with persons with acquired apraxia of speech and aphasia | |
| CN109841231B (en) | Early AD (AD) speech auxiliary screening system for Chinese mandarin | |
| CN114898861A (en) | Multi-modal depression detection method and system based on full attention mechanism | |
| WO2022257630A1 (en) | Risk detection method and apparatus based on multi-modal concealed information test | |
| CN104347071B (en) | Method and system for generating reference answers of spoken language test | |
| KR20220048381A (en) | Device, method and program for speech impairment evaluation | |
| CN116487003A (en) | Rehabilitation training system for patient with language disorder | |
| Li et al. | Improvement on speech depression recognition based on deep networks | |
| CN114916921A (en) | Rapid speech cognition assessment method and device | |
| Liu et al. | A novel decision tree for depression recognition in speech | |
| Yu et al. | Enhancing speech recognition for Parkinson’s disease patient using transfer learning technique | |
| JP6699825B2 (en) | Diagnostic device, diagnostic device control method, and diagnostic program | |
| Liu et al. | AI recognition method of pronunciation errors in oral English speech with the help of big data for personalized learning | |
| Mirheidari | Detecting early signs of dementia in conversation | |
| CN112308379A (en) | Service order evaluation method, device, equipment and storage medium for home care |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200918 |
|
| RJ01 | Rejection of invention patent application after publication |