CN114372201A - Physical examination information intelligent recommendation method and system, storage medium and computing equipment - Google Patents
Physical examination information intelligent recommendation method and system, storage medium and computing equipment Download PDFInfo
- Publication number
- CN114372201A CN114372201A CN202210026966.7A CN202210026966A CN114372201A CN 114372201 A CN114372201 A CN 114372201A CN 202210026966 A CN202210026966 A CN 202210026966A CN 114372201 A CN114372201 A CN 114372201A
- Authority
- CN
- China
- Prior art keywords
- physical examination
- information
- package
- target
- disease
- 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
- 238000000034 method Methods 0.000 title claims abstract description 53
- 201000010099 disease Diseases 0.000 claims abstract description 116
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims abstract description 116
- 238000012216 screening Methods 0.000 claims abstract description 60
- 230000036541 health Effects 0.000 claims abstract description 57
- 230000004044 response Effects 0.000 claims abstract description 19
- 238000004458 analytical method Methods 0.000 claims description 14
- 238000007477 logistic regression Methods 0.000 claims description 13
- 238000012360 testing method Methods 0.000 claims description 12
- 238000012549 training Methods 0.000 claims description 11
- 230000007246 mechanism Effects 0.000 claims description 10
- 238000001228 spectrum Methods 0.000 claims description 10
- 206010012601 diabetes mellitus Diseases 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 6
- 238000003062 neural network model Methods 0.000 claims description 6
- 230000000306 recurrent effect Effects 0.000 claims description 6
- 230000008520 organization Effects 0.000 claims description 5
- 208000024172 Cardiovascular disease Diseases 0.000 claims description 4
- 208000017667 Chronic Disease Diseases 0.000 claims description 4
- 208000004930 Fatty Liver Diseases 0.000 claims description 4
- 206010019708 Hepatic steatosis Diseases 0.000 claims description 4
- 206010020772 Hypertension Diseases 0.000 claims description 4
- 201000001431 Hyperuricemia Diseases 0.000 claims description 4
- 208000026106 cerebrovascular disease Diseases 0.000 claims description 4
- 208000029078 coronary artery disease Diseases 0.000 claims description 4
- 230000002526 effect on cardiovascular system Effects 0.000 claims description 4
- 208000010706 fatty liver disease Diseases 0.000 claims description 4
- 238000009432 framing Methods 0.000 claims description 4
- 231100000240 steatosis hepatitis Toxicity 0.000 claims description 4
- 208000023516 stroke disease Diseases 0.000 claims description 4
- 238000005516 engineering process Methods 0.000 abstract description 14
- 238000013473 artificial intelligence Methods 0.000 abstract description 7
- 238000010801 machine learning Methods 0.000 abstract description 4
- 238000011156 evaluation Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 230000008901 benefit Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000036772 blood pressure Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000007704 transition Effects 0.000 description 3
- 239000008280 blood Substances 0.000 description 2
- 210000004369 blood Anatomy 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 208000024891 symptom Diseases 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 238000007476 Maximum Likelihood Methods 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000004931 aggregating effect Effects 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000003862 health status Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005309 stochastic process Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/14—Fourier, Walsh or analogous domain transformations, e.g. Laplace, Hilbert, Karhunen-Loeve, transforms
- G06F17/141—Discrete Fourier transforms
- G06F17/142—Fast Fourier transforms, e.g. using a Cooley-Tukey type algorithm
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- G10L17/02—Preprocessing operations, e.g. segment selection; Pattern representation or modelling, e.g. based on linear discriminant analysis [LDA] or principal components; Feature selection or extraction
-
- 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
- G10L17/16—Hidden Markov models [HMM]
-
- 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
- G10L17/22—Interactive procedures; Man-machine interfaces
-
- 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
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
- G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
- G10L25/66—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination for extracting parameters related to health condition
-
- 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
- G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Physics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Multimedia (AREA)
- General Health & Medical Sciences (AREA)
- Acoustics & Sound (AREA)
- Human Computer Interaction (AREA)
- General Engineering & Computer Science (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Software Systems (AREA)
- Public Health (AREA)
- Evolutionary Computation (AREA)
- Mathematical Optimization (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Artificial Intelligence (AREA)
- Medical Informatics (AREA)
- Pure & Applied Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Epidemiology (AREA)
- Computing Systems (AREA)
- Mathematical Analysis (AREA)
- Computational Mathematics (AREA)
- Discrete Mathematics (AREA)
- Primary Health Care (AREA)
- Pathology (AREA)
- Algebra (AREA)
- Signal Processing (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention provides an intelligent physical examination information recommendation method and system, a storage medium and computing equipment, and relates to the fields of artificial intelligence and intelligent medical treatment, wherein the method comprises the steps of collecting preset question response information of a target object; identifying the object gender of the target object by using preset question reply information, and screening a plurality of physical examination packages matched with the object gender; extracting at least one health index keyword in preset question reply information, and predicting a disease risk type and a disease risk probability corresponding to a target object based on the health index keyword by using a pre-trained disease screening model; and selecting at least one target physical examination package matched with the disease risk type from the plurality of physical examination packages according to the disease risk probability to serve as the physical examination package intelligently recommended to the target object. The scheme provided by the invention combines the voice recognition technology and the machine learning technology, so that the recommendation result of the physical examination package is more accurate, and the user can use the package more conveniently.
Description
Technical Field
The invention relates to the technical field of artificial intelligence and intelligent medical treatment, in particular to an intelligent physical examination information recommendation method and system, a storage medium and computing equipment.
Background
The physical examination is a regular health examination performed to enable a user to know the health condition of the user, and particularly, people with large working pressure and irregular work and rest need regular physical examination and pay attention to the physical health of the user. Typically, different institutions often push out fixed packages of physical examination for selection by the user. However, the conventional physical examination package recommendation requires that a user check various physical examination package applicable rules by himself, most of the physical examination package applicable rules are presented in a text form, and poor experience is easily caused to the user. Meanwhile, due to the difference of medical knowledge reserves of different users, a package really suitable for the user can not be selected according to the written description of the package, so that the physical examination effect cannot be effectively achieved.
Disclosure of Invention
In view of the above problems, the present invention has been made to provide a physical examination information intelligent recommendation method and system, a storage medium, and a computing device that overcome or at least partially solve the above problems.
According to a first aspect of the present invention, there is provided a physical examination information intelligent recommendation method, the method comprising:
acquiring preset question reply information input by a target object in a voice mode;
extracting the object gender of the target object through the preset question reply information, and screening a plurality of physical examination packages matched with the object gender;
extracting at least one health index keyword in the preset question reply information, and predicting a disease risk type and a disease risk probability corresponding to the target object based on the health index keyword by using a pre-trained disease screening model;
and selecting at least one target physical examination package matched with the disease risk type from the plurality of physical examination packages according to the disease risk probability to serve as a physical examination package intelligently recommended to the target object.
Optionally, the extracting, by the preset question reply information, the subject gender of the target subject includes:
extracting a Mel frequency cepstrum coefficient in the preset question reply information;
and inputting the Mel frequency cepstrum coefficient into a trained hidden Markov model, and identifying the object gender of the target object by using the hidden Markov model.
Optionally, the extracting mel-frequency cepstrum coefficients from the preset question reply information includes:
performing pre-emphasis, framing and windowing on the preset question reply information to generate a plurality of short-time analysis windows;
for each short time analysis window, utilizing fast Fourier transform to obtain a corresponding frequency spectrum;
and inputting the frequency spectrum corresponding to each short-time analysis window into a Mel filter bank to obtain a Mel frequency cepstrum coefficient corresponding to the preset question response information.
Optionally, the extracting at least one health indicator keyword from the preset question reply information includes:
and converting the preset question reply information into text introduction information by using a pre-trained encoder-decoder recurrent neural network model, and extracting at least one health index keyword in the text introduction information by using a pre-trained keyword extraction model.
Optionally, before extracting at least one health indicator keyword from the preset question reply information and predicting a disease risk type and a disease risk probability corresponding to the target object based on the health indicator keyword by using a pre-trained disease screening model, the method further includes:
establishing a logistic regression model;
collecting case sample data in batch, and establishing a training set and a test set based on the case sample data;
learning the logistic regression model by using the training set to obtain model parameters, and testing the performance of the model by using the test set to obtain at least one disease screening model capable of identifying a corresponding health risk type based on health indexes;
the disease screening model at least comprises a chronic disease screening model corresponding to hypertension, fatty liver, diabetes and hyperuricemia and a cardiovascular and cerebrovascular disease screening model corresponding to heart age, stroke and coronary heart disease.
Optionally, before the extracting the subject gender of the target subject through the preset question answering information and screening out a plurality of physical examination packages matching the subject gender, the method further includes:
collecting physical examination package information provided by each third-party physical examination organization; the physical examination package information comprises a plurality of physical examination packages, physical examination items contained in each of the physical examination packages and package characteristic information of the adapted population;
and adding corresponding characteristic labels for the alternative physical examination packages according to the applicable gender, the physical examination items and the package characteristic information.
Optionally, after the selecting at least one target physical examination package matching the disease risk type from the plurality of physical examination packages according to the disease risk probability as a physical examination package intelligently recommended to the target object, the method further includes:
acquiring feedback information provided by at least one physical examination object for the target physical examination package;
and transmitting the feedback information to a target third-party physical examination mechanism corresponding to the target physical examination package, and adjusting physical examination items contained in the target physical examination package by the target third-party physical examination mechanism.
According to a second aspect of the present invention, there is provided a physical examination information intelligent recommendation system, the system comprising:
the information acquisition module is used for acquiring preset question reply information input by a target object in a voice mode;
the package screening module is used for extracting the object gender of the target object through the preset question reply information and screening a plurality of physical examination packages matched with the object gender;
the risk prediction module is used for extracting at least one health index keyword in the preset question reply information and predicting a disease risk type and a disease risk probability corresponding to the target object based on the health index keyword by utilizing a pre-trained disease screening model;
and the package recommending module is used for selecting at least one target physical examination package matched with the disease risk type from the plurality of physical examination packages according to the disease risk probability to serve as the physical examination package intelligently recommended to the target object.
According to a third aspect of the present invention, there is provided a computer-readable storage medium for storing program code for executing the intelligent physical examination information recommendation method of any one of the first aspect.
According to a fourth aspect of the invention, there is provided a computing device comprising a processor and a memory:
the memory is used for storing program codes and transmitting the program codes to the processor;
the processor is configured to execute the intelligent physical examination information recommendation method according to any one of the first aspect according to instructions in the program code.
The invention provides an intelligent physical examination information recommendation method and system, a storage medium and computing equipment, wherein based on the intelligent physical examination information recommendation method provided by the invention, preset question reply information of a target object is acquired, basic information and health condition information of a user are automatically identified by adopting a voice recognition technology, and a disease risk type and a disease risk probability corresponding to the target object are predicted by utilizing a pre-trained disease screening model; the physical examination package matched with the self condition of the target object is matched according to the disease risk probability, the recommendation result of the scheme provided by the invention is more accurate, the use of the scheme is more convenient and faster, and the method can help a third-party physical examination structure to improve the service quality and improve the customer satisfaction.
The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.
The above and other objects, advantages and features of the present invention will become more apparent to those skilled in the art from the following detailed description of specific embodiments thereof, taken in conjunction with the accompanying drawings.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
fig. 1 is a schematic flow chart illustrating an intelligent physical examination information recommendation method according to an embodiment of the invention;
FIG. 2 illustrates a hidden Markov model inference principle diagram according to an embodiment of the invention;
FIG. 3 is a schematic structural diagram of an intelligent physical examination information recommendation device according to an embodiment of the invention;
fig. 4 is a schematic structural diagram of an intelligent physical examination information recommendation device according to another embodiment of the invention;
FIG. 5 shows a schematic diagram of a computing device architecture according to an embodiment of the invention.
Detailed Description
Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.
The solution provided in this embodiment mainly combines with an Artificial Intelligence (AI) technology to acquire and process related data, wherein the AI is a theory, method, technique, and application system that simulates, extends, and expands human Intelligence using a digital computer or a machine of a digital computer controller, senses the environment, acquires knowledge, and uses the knowledge to obtain the best result.
The artificial intelligence technology base generally comprises technologies such as sensors, special artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation/interaction systems, electromechanical integration and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a robot technology, a biological recognition technology, a voice processing technology, a natural voice processing technology, machine learning/deep learning and the like.
The physical examination information intelligent recommendation method provided by the embodiment can be applied to a physical examination information intelligent recommendation system, such as a self-service machine provided by a third-party physical examination organization, or other service organizations providing physical examination package recommendation. Fig. 1 is a schematic flow chart of an intelligent physical examination information recommendation method according to an embodiment of the present invention, and as can be seen from fig. 1, the intelligent physical examination information recommendation method provided in this embodiment at least includes the following steps S101 to S104.
S101, preset question reply information input by a target object in a voice mode is obtained.
The target object can be any user needing physical examination package recommendation, and preset question reply information of the user needs to be collected before the physical examination package recommendation. The preset question reply information refers to reply information to a preset question that the target object inputs in a voice manner, wherein the preset question may be related information on the basic condition and health condition of the target object, which is asked for the target object. The preset question answering information comprises basic information and health condition information of the target object, wherein the basic information comprises information of birth year (or age), marital state, occupation and the like, and the health condition information comprises information related to physical health such as living habits, disease history, sleep condition, blood pressure condition, body type, physical discomfort symptoms and the like.
As described above, the physical examination information intelligent recommendation system (e.g., a host) and other devices can be used to collect the target preset question response information, and before the preset question response information corresponding to the target object is obtained, the question can be preset, so that the preset question response information corresponding to the target object can be obtained by combining the preset question. For example, the target object is asked by voice or text, including basic information and health condition information, such as "whether there is physical discomfort recently", "please introduce basic information", "whether there is disease history", and so on. Further, preset question reply information is generated by combining the query information during the query and the answer information fed back by the target object for the query information. Based on the method provided by the embodiment of the invention, the preset question reply information corresponding to the target object is acquired by utilizing the preset inquiry telephone, so that more comprehensive basic information and health condition information of the target object can be acquired, the physical condition of the target object can be more accurately known, and the physical examination package can be more accurately recommended.
S102, extracting the object gender of the target object through preset question reply information, and screening a plurality of physical examination packages matched with the object gender.
In the scheme provided by this embodiment, a database for storing the physical examination packages provided by different institutions is provided, and after preset question response information corresponding to the target object is collected, the physical examination package matched with the object gender of the target object can be searched from the database.
Optionally, physical examination package information provided by each third-party physical examination institution may be collected in advance; the physical examination package information comprises a plurality of physical examination packages, physical examination items contained in each physical examination package and package characteristic information of the adapted group; and further adding corresponding characteristic labels for the alternative physical examination packages according to the applicable gender, the physical examination items and the package characteristic information, and correspondingly storing the alternative physical examination packages and the corresponding characteristic labels. The package characteristic information may include characteristic information such as screening disease types, and the characteristic labels may be labels such as group categories, sexes, ages, and the like, which are applicable to the physical examination package.
After the preset question reply information of the target object is acquired, gender identification can be performed on the target object, and then a plurality of physical examination packages matched with the gender of the target object are screened out by using the feature tags corresponding to the various alternative physical examination packages in the database, so that the recommendation range is narrowed to a package suitable for males or a package suitable for females.
In an optional embodiment of the present invention, the extracting, in the step S102, the subject gender of the target subject through the preset question answering information may include:
s102-1, extracting Mel Frequency Cepstrum Coefficients (MFCC) in the preset question reply information. In the field of sound processing, mel-frequency cepstrum is a linear transformation of the log energy spectrum based on the non-linear mel scale (mel scale) of sound frequencies. Optionally, the extracting the mel-frequency cepstrum coefficient in the preset question response information may specifically include:
s102-1-1, performing pre-emphasis, framing and windowing on the preset question reply information to generate a plurality of short-time analysis windows;
s102-1-2, for each short-time analysis window, obtaining a corresponding frequency spectrum by utilizing fast Fourier transform;
and S102-1-3, inputting the frequency spectrum corresponding to each short-time analysis window into a Mel filter bank to obtain Mel frequency cepstrum coefficients corresponding to preset question response information. Cepstral analysis (taking the logarithm, inverse transformation) is performed on top of the Mel-frequency spectrum to obtain Mel-frequency cepstral coefficients MFCC.
S102-2, inputting the Mel frequency cepstrum coefficient into the trained hidden Markov model, and identifying the object gender of the target object by using the hidden Markov model.
Hidden markov is a double stochastic process with a hidden markov chain of a certain number of states and a set of display random functions. It plays an important role in speech recognition, behavior recognition, character recognition and fault diagnosis. Hidden Markov model HMM principle:
for the hidden-cause markov model, consider a set of discrete states Q and a set of observed states V:
Q＝{q1，q2，...，qN}，V＝{v1，v2，...，vm}
wherein q is1，q2，...，qNSequences respectively representing discrete states corresponding to different moments are determined by initial state probability and state transition probability (pi, A); v. of1，v2，...，vmRespectively representing discrete bright characters, and determined by the state sequence and the bright character generation probability (Q, B) of each state;
s is a state sequence of length T, O is the corresponding observation sequence, i.e.:
S＝(s1，s2，...，sT)，O＝(o1，o2，...，oT)
wherein s is1，s2，...，sTRespectively representing hidden state variables corresponding to different moments; o1，o2，...，oTRespectively representing observation values corresponding to different moments;
state variable S (t) e S, observation state O (t) e O, defining probability:
aijindicating state to state siS ofjTransition probabilities, i and j respectively correspond to two different states, and t represents time; such aijThe state transition matrix a of the markov chain is formed. And the output probability for each state is:
such a bi(k) Constituting a probability matrix B of observation state generation. Probabilistic inference graph of itAs shown in fig. 2.
The square row nodes in fig. 2 represent the observations o (t). The circular nodes represent hidden state variables s (t), the horizontal arrows represent transition matrices a, and the vertical arrows represent observed state probability matrices B. To simplify the model, the HMM model has two important assumptions:
1) a homogeneous markov chain hypothesis. I.e. the observed state value at any moment in time depends only on its previous hidden state.
2) Observe the independence assumption. I.e. the observed state at any moment only depends on the hidden state at the current moment.
In addition, a set of hidden state probability distributions pi at time t ═ 1 is needed, and the HMM model can be represented by a triplet as follows:
λ＝(A，B，Π)
in operation, the HMM model has three basic steps:
a) and (3) probability calculation: given a model λ ═ (a, B, Π) and an observation sequence O ═ O (O)1，o2，...，oT) The probability P (O | λ) of the occurrence of the observation sequence O under the model λ is calculated.
b) Learning parameters: the known observation sequence O ═ O (O)1，o2，...，oT) The model λ is estimated as the (a, B, Π) parameter so that the probability of observing the sequence P (O | λ) under the model is maximized. I.e. the parameters are estimated using maximum likelihood estimation.
c) And (3) prediction: the known model λ ═ (a, B, Π) and the observation sequence O ═ O (O)1，o2，...，oT) The state sequence S which has the maximum conditional probability P (S | O) for a given observation sequence is determined as (S)1，s2，...，sT)。
Before the hidden-cause markov model is used for gender identification, a feature set which comprises MFCC feature information corresponding to male voice and MFCC feature information corresponding to female voice and is collected in advance can be used, the hidden-cause markov model is trained by the feature set, and then the trained hidden-cause markov model is used for identifying the object gender of the target object.
S103, extracting at least one health index keyword in the preset question reply information, and predicting the disease risk type and the disease risk probability corresponding to the target object based on the health index keyword by using a pre-trained disease screening model.
As introduced above, the preset question response information of the target object may include basic information and health condition information, and after the preset question response information of the target object is collected, voice recognition may be performed to obtain a corresponding health index keyword. In this embodiment, a pre-trained encoder-decoder recurrent neural network model may be used to convert the preset question reply information into text introduction information, and a trained keyword extraction model is used to extract at least one health index keyword in the text introduction information; the recurrent neural network model and the keyword extraction model are obtained by training through batch training samples collected in advance.
Alternatively, the recurrent neural network model may be a seq2seq model. seq2seq belongs to one of the Encoder-Decoder structures, which uses two RNNs, one RNN as Encoder and the other RNN as Decoder. The Encoder is responsible for compressing an input sequence into a vector with a specified length, and the Decode is responsible for generating a specified sequence according to a semantic vector. In this embodiment, the seq2seq model is used to perform text translation on the preset question reply information provided by the target object, so that text introduction information corresponding to the preset question reply information can be quickly and accurately obtained. In addition, the keyword extraction model may be a KeyBERT model, which is a method for creating keywords and keyword phrases that are most similar to a document by using BERT embedding, and is mainly a quick and simple method for creating keywords and keywords.
That is to say, for the obtained preset question response information, firstly, the seq2seq model is used to translate the preset question response information, and then, the KeyBERT model is used to extract keywords, so as to identify basic information, living habits, medical histories and index-related characters in the preset question response information, such as health index keywords such as "year of birth", "marital status", "occupation", "blood pressure" and "blood sugar". Further, the disease screening model can be used to predict the disease risk type and the disease risk probability corresponding to the target object based on the health index keyword.
The disease screening model in this embodiment is an intelligent machine learning model established based on an artificial intelligence technique. The input data of the disease screening model are a plurality of health index keywords related to the health condition of the target object, such as keyword information of user occupation, marital state, blood pressure height, blood sugar height and the like. The output data of the disease screening model is the disease risk type matched with the health index keyword of the target object and the probability corresponding to the disease risk type. The disease screening model can be used for intelligently evaluating the possible disease risk of a target object according to the health condition of the target object.
In an alternative embodiment of the present invention, the establishment and training of the disease screening model may be performed through the following steps A1-A3.
And A1, establishing a logistic regression model. Logistic Regression (Logistic Regression) is a classification model in machine learning, belongs to a generalized linear model, and in the embodiment, by establishing a Logistic Regression model and using the Logistic Regression model as an initial model of a disease screening model, the disease screening model can be simple and efficient when executing a specific screening algorithm.
A2, collecting the sample data of cases in batch, and establishing a training set and a testing set based on the sample data of cases. The case sample data may be actual case data acquired by an institution such as a hospital, and may include related information such as health indexes, affected diseases, pre-disease symptoms, and health indexes corresponding to each case.
A3, learning the logistic regression model by using the training set to obtain model parameters, and testing the model performance by using the test set to obtain at least one disease screening model capable of identifying the corresponding health risk type based on the health index. The disease screening model is an intelligent model which is established in advance and trained for evaluating disease risk. The disease screening model at least includes a chronic disease screening model corresponding to hypertension, fatty liver, diabetes, and hyperuricemia, and a cardiovascular and cerebrovascular disease screening model corresponding to heart age, stroke, and coronary heart disease, and specifically, the disease screening model may be a model capable of performing multiple disease risk evaluations, or may be multiple sub-models corresponding to different types of disease risks, which is not limited in this embodiment. The disease screening model of this embodiment may extract a feature value of each health index keyword in the preset question response information, and compare the extracted feature value with a normal value range and an abnormal value range of the index obtained through learning in advance, thereby obtaining a comparison result. For different types of diseases, different weights can be correspondingly set for each health index, and the probabilities corresponding to the various diseases can be obtained by aggregating according to the comparison results and the respective corresponding weight values.
In the method provided by this embodiment, for at least one health index keyword in the extracted preset question reply information, the health index keyword may be input into a disease screening model as input data, and the disease screening model performs comprehensive intelligent evaluation and analysis to output probabilities that a target object suffers from various diseases. For example, the disease screening model outputs a result with a probability of a% of the risk of disease a, a probability of B% of the risk of disease B, and a probability of C% of the risk of disease C. According to the method provided by the embodiment of the invention, the disease screening model for carrying out multiple disease risk evaluations is established by adopting a logistic regression method, so that not only can the evaluation of different disease risks corresponding to the target object be simply and efficiently realized, but also the comprehensive disease risk evaluation can be completed aiming at different health indexes of the target object.
S104, selecting at least one target physical examination package matched with the disease risk type from the plurality of physical examination packages according to the disease risk probability, and using the selected target physical examination package as a physical examination package intelligently recommended to a target object.
Continuing to take the disease risk corresponding to the disease a in the above embodiment as a probability a%, the disease risk corresponding to the disease B as a probability B%, and the disease risk corresponding to the disease C as a probability C%, after obtaining the disease risk probabilities of the above diseases, comparing the values of a%, B%, and C%, assuming that a% is the largest, it indicates that the risk of the target object suffering from the disease a is higher, at this time, screening out the high-risk disease a, and selecting a physical examination package recommendation corresponding to the matched row from the multiple individual physical examination packages for the disease a. For example, if the target subject is at high risk for diabetes, a diabetes screening package is recommended. According to the method provided by the embodiment of the invention, after the disease risk type and the disease risk probability corresponding to the target object are predicted, the disease risk type with the higher disease risk probability possibly existing in the target object is selected according to the disease risk probability to determine the high-risk disease, and then the target physical examination package matched with the high-risk disease is selected from a plurality of physical examination packages.
In the above embodiment, each physical examination package stored in the database has a corresponding feature tag, so that when a target physical examination package is selected, a keyword matching technology can be used to select a physical examination package having an examination item corresponding to a disease risk type from a plurality of physical examination packages as the target physical examination package and recommend the target physical examination package to a user. In practical applications, two or more target medical examination packages may be matched with the actual health status of the target object in the plurality of medical examination packages selected in step S102, and at this time, two or more target medical examination packages may be simultaneously displayed to the target object and selected by the target object. Or, for the two or more matched target physical examination packages, the difference information between the target physical examination packages can be intelligently analyzed and acquired, and then a physical examination package recommendation table is generated by combining the physical examination package information and the difference information corresponding to the target physical examination packages so as to recommend the physical examination package recommendation table to the user for selection. The difference information corresponding to any target physical examination package is used for indicating the difference information between the target physical examination package and other target physical examination packages, such as the difference physical examination items, the difference price, and the like.
In this embodiment, when the target physical examination package is a plurality of, the difference information through with a plurality of target physical examination suites is directly perceived to show for the user and look over to the supplementary user selects the physical examination package that needs more to match with self, thereby effectively reaches the physical examination effect. As described above, the physical examination information intelligent recommendation method provided in this embodiment can be applied to a physical examination information intelligent recommendation system, and when the physical examination information intelligent recommendation system is set with any third-party physical examination organization, when a target physical examination package is selected, the physical examination package provided by the third-party physical examination organization is preferentially selected from the database as the target physical examination package.
In an optional embodiment of the present invention, after the step S104, feedback information provided by at least one physical examination object for the target physical examination package may also be acquired; and transmitting the feedback information to a target third-party physical examination mechanism corresponding to the target physical examination package, and adjusting the physical examination items contained in the target physical examination package by the target third-party physical examination mechanism. That is, after the target physical examination package is recommended to the target object, feedback information of the target object for the recommended target physical examination package may be collected, where the feedback information may include user satisfaction (which may be classified into multiple levels or a scoring system), supplementary advice information (such as advice of which physical examination items to add or advice of which physical examination items to delete), and so on. After the collected feedback information of the target object for the target physical examination package is received, the feedback information may be fed back to a third-party physical examination mechanism corresponding to the target physical examination package, and the third-party physical examination mechanism adjusts information included in the target physical examination package, or creates a new physical examination package. In addition, individual physical examination items can be added automatically according to the feedback information, or any item in the physical examination package is replaced according to the feedback information, so that a new physical examination package is generated. The finally generated physical examination package more meets the personalized requirements of the user.
The embodiment of the invention provides an intelligent physical examination information recommendation method, which adopts a technology combining voice recognition and disease screening, can automatically recognize information such as basic information, living habits, disease histories and physical sign indexes of a user according to preset question response information of a target object, and matches a physical examination package matched with the self condition of the target object according to the probability output by a disease screening model. The physical examination package recommendation method provided by the embodiment is more accurate in corresponding recommendation result, and more convenient for a user to use, and can help a third-party physical examination structure to improve service quality and improve customer satisfaction.
Based on the same inventive concept, an embodiment of the present invention further provides an intelligent physical examination information recommendation system, as shown in fig. 3, the intelligent physical examination information recommendation system may include:
the information acquisition module 310 is configured to acquire preset question reply information input by a target object in a voice manner;
a package screening module 320, configured to extract a subject gender of the target subject through the preset question reply information, and screen out a plurality of physical examination packages matched with the subject gender;
the risk prediction module 330 is configured to extract at least one health indicator keyword from the preset question response information, and predict a disease risk type and a disease risk probability corresponding to the target object based on the health indicator keyword by using a pre-trained disease screening model;
the package recommending module 340 is configured to select at least one target physical examination package matched with the disease risk type from the multiple physical examination packages according to the disease risk probability, and use the selected target physical examination package as a physical examination package intelligently recommended to a target object.
In an optional embodiment of the present invention, package screening module 320 may further be configured to: extracting a Mel frequency cepstrum coefficient in preset question response information; and inputting the Mel frequency cepstrum coefficient into the trained hidden Markov model, and identifying the object gender of the target object by using the hidden Markov model.
In an optional embodiment of the present invention, package screening module 320 may further be configured to:
pre-emphasis, framing and windowing are carried out on the preset question reply information to generate a plurality of short-time analysis windows; for each short-time analysis window, obtaining a corresponding frequency spectrum by utilizing fast Fourier transform; and inputting the frequency spectrum corresponding to each short-time analysis window into a Mel filter bank to obtain a Mel frequency cepstrum coefficient corresponding to the preset question response information.
In an optional embodiment of the present invention, the risk prediction module 330 may further be configured to: and converting the preset question reply information into text introduction information by using a pre-trained encoder-decoder recurrent neural network model, and extracting at least one health index keyword in the text introduction information by using a pre-trained keyword extraction model.
In an optional embodiment of the present invention, as shown in fig. 4, the physical examination information intelligent recommendation system may further include: a model building module 350;
a model building module 350 for building a logistic regression model; collecting case sample data in batch, and establishing a training set and a test set based on the case sample data; learning the logistic regression model by using a training set to obtain model parameters, and testing the performance of the model by using a test set to obtain at least one disease screening model capable of identifying a corresponding health risk type based on health indexes;
the disease screening model at least comprises a chronic disease screening model corresponding to hypertension, fatty liver, diabetes and hyperuricemia and a cardiovascular and cerebrovascular disease screening model corresponding to heart age, stroke and coronary heart disease.
In an optional embodiment of the present invention, as shown in fig. 4, the physical examination information intelligent recommendation system may further include a package collection module 360;
a package collecting module 360, configured to collect physical examination package information provided by each third-party physical examination institution; the physical examination package information comprises a plurality of physical examination packages, physical examination items contained in each physical examination package and package characteristic information of the adapted group; and adding corresponding characteristic labels for each optional physical examination package according to the applicable gender, the physical examination items and the package characteristic information.
In an optional embodiment of the present invention, as shown in fig. 4, the physical examination information intelligent recommendation system may further include:
a feedback module 370, configured to obtain feedback information provided by at least one physical examination subject for the target physical examination package; and transmitting the feedback information to a target third-party physical examination mechanism corresponding to the target physical examination package, and adjusting the physical examination items contained in the target physical examination package by the target third-party physical examination mechanism.
The embodiment of the invention also provides a computer-readable storage medium, wherein the computer-readable storage medium is used for storing a program code, and the program code is used for executing the physical examination information intelligent recommendation method in the embodiment.
An embodiment of the present invention further provides a computing device, as shown in fig. 5, where the computing device includes a processor and a memory: the memory is used for storing program codes and transmitting the program codes to the processor; the processor is used for executing the intelligent physical examination information recommendation method according to the instruction in the program code. Optionally, the computer device may also include a user interface, a network interface, a camera, Radio Frequency (RF) circuitry, sensors, audio circuitry, a WI-FI module, and so forth. The user interface may include a Display screen (Display), an input unit such as a keypad (Keyboard), etc., and the optional user interface may also include a USB interface, a card reader interface, etc. The network interface may optionally include a standard wired interface, a wireless interface (e.g., a bluetooth interface, WI-FI interface), etc.
It is clear to those skilled in the art that the specific working processes of the above-described systems, devices, modules and units may refer to the corresponding processes in the foregoing method embodiments, and for the sake of brevity, further description is omitted here.
In addition, the functional units in the embodiments of the present invention may be physically independent of each other, two or more functional units may be integrated together, or all the functional units may be integrated in one processing unit. The integrated functional units may be implemented in the form of hardware, or in the form of software or firmware.
Those of ordinary skill in the art will understand that: the integrated functional units, if implemented in software and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computing device (e.g., a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention when the instructions are executed. And the aforementioned storage medium includes: u disk, removable hard disk, Read Only Memory (ROM), Random Access Memory (RAM), magnetic or optical disk, and other various media capable of storing program code.
Alternatively, all or part of the steps of implementing the foregoing method embodiments may be implemented by hardware (such as a computing device, e.g., a personal computer, a server, or a network device) associated with program instructions, which may be stored in a computer-readable storage medium, and when the program instructions are executed by a processor of the computing device, the computing device executes all or part of the steps of the method according to the embodiments of the present invention.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments can be modified or some or all of the technical features can be equivalently replaced within the spirit and principle of the present invention; such modifications or substitutions do not depart from the scope of the present invention.
Claims (10)
1. An intelligent physical examination information recommendation method is characterized by comprising the following steps:
acquiring preset question reply information input by a target object in a voice mode;
extracting the object gender of the target object through the preset question reply information, and screening a plurality of physical examination packages matched with the object gender;
extracting at least one health index keyword in the preset question reply information, and predicting a disease risk type and a disease risk probability corresponding to the target object based on the health index keyword by using a pre-trained disease screening model;
and selecting at least one target physical examination package matched with the disease risk type from the plurality of physical examination packages according to the disease risk probability to serve as a physical examination package intelligently recommended to the target object.
2. The method according to claim 1, wherein said extracting the subject gender of the target subject through the preset question reply information comprises:
extracting a Mel frequency cepstrum coefficient in the preset question reply information;
and inputting the Mel frequency cepstrum coefficient into a trained hidden Markov model, and identifying the object gender of the target object by using the hidden Markov model.
3. The method according to claim 2, wherein said extracting mel-frequency cepstrum coefficients from said preset question-answer information comprises:
performing pre-emphasis, framing and windowing on the preset question reply information to generate a plurality of short-time analysis windows;
for each short time analysis window, utilizing fast Fourier transform to obtain a corresponding frequency spectrum;
and inputting the frequency spectrum corresponding to each short-time analysis window into a Mel filter bank to obtain a Mel frequency cepstrum coefficient corresponding to the preset question response information.
4. The method according to claim 1, wherein said extracting at least one health indicator keyword from the preset question reply message comprises:
and converting the preset question reply information into text introduction information by using a pre-trained encoder-decoder recurrent neural network model, and extracting at least one health index keyword in the text introduction information by using a pre-trained keyword extraction model.
5. The method according to claim 1, wherein before extracting at least one health index keyword from the preset question reply information and predicting a disease risk type and a disease risk probability corresponding to the target object based on the health index keyword by using a pre-trained disease screening model, the method further comprises:
establishing a logistic regression model;
collecting case sample data in batch, and establishing a training set and a test set based on the case sample data;
learning the logistic regression model by using the training set to obtain model parameters, and testing the performance of the model by using the test set to obtain at least one disease screening model capable of identifying a corresponding health risk type based on health indexes;
the disease screening model at least comprises a chronic disease screening model corresponding to hypertension, fatty liver, diabetes and hyperuricemia and a cardiovascular and cerebrovascular disease screening model corresponding to heart age, stroke and coronary heart disease.
6. The method according to any one of claims 1 to 5, wherein before the extracting the subject gender of the target subject from the preset question reply information and screening out a plurality of physical examination packages matching the subject gender, the method further comprises:
collecting physical examination package information provided by each third-party physical examination organization; the physical examination package information comprises a plurality of physical examination packages, physical examination items contained in each of the physical examination packages and package characteristic information of the adapted population;
and adding corresponding characteristic labels for the alternative physical examination packages according to the applicable gender, the physical examination items and the package characteristic information.
7. The method of any one of claims 1-5, wherein the selecting at least one target physical examination package matching the disease risk type from the plurality of physical examination packages according to the disease risk probability as the physical examination package intelligently recommended to the target subject, further comprises:
acquiring feedback information provided by at least one physical examination object for the target physical examination package;
and transmitting the feedback information to a target third-party physical examination mechanism corresponding to the target physical examination package, and automatically adjusting the physical examination items contained in the target physical examination package by the target third-party physical examination mechanism.
8. An intelligent physical examination information recommendation system, which is characterized by comprising:
the information acquisition module is used for acquiring preset question reply information input by a target object in a voice mode;
the package screening module is used for extracting the object gender of the target object through the preset question reply information and screening a plurality of physical examination packages matched with the object gender;
the risk prediction module is used for extracting at least one health index keyword in the preset question reply information and predicting a disease risk type and a disease risk probability corresponding to the target object based on the health index keyword by utilizing a pre-trained disease screening model;
and the package recommending module is used for selecting at least one target physical examination package matched with the disease risk type from the plurality of physical examination packages according to the disease risk probability to serve as the physical examination package intelligently recommended to the target object.
9. A computer-readable storage medium for storing program code for executing the intelligent physical examination information recommendation method according to any one of claims 1 to 7.
10. A computing device, the computing device comprising a processor and a memory:
the memory is used for storing program codes and transmitting the program codes to the processor;
the processor is used for executing the intelligent physical examination information recommendation method of any one of claims 1 to 7 according to instructions in the program code.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210026966.7A CN114372201A (en) | 2022-01-11 | 2022-01-11 | Physical examination information intelligent recommendation method and system, storage medium and computing equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210026966.7A CN114372201A (en) | 2022-01-11 | 2022-01-11 | Physical examination information intelligent recommendation method and system, storage medium and computing equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114372201A true CN114372201A (en) | 2022-04-19 |
Family
ID=81143871
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210026966.7A Pending CN114372201A (en) | 2022-01-11 | 2022-01-11 | Physical examination information intelligent recommendation method and system, storage medium and computing equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114372201A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116030984A (en) * | 2023-03-31 | 2023-04-28 | 武汉携康智能健康设备有限公司 | User physical examination system and physical examination method based on intelligent health station |
| CN116759061A (en) * | 2023-08-17 | 2023-09-15 | 简阳市人民医院 | Physical examination project recommendation system based on personal demands |
| CN117951190A (en) * | 2024-03-18 | 2024-04-30 | 深圳市双佳医疗科技有限公司 | Human body index abnormal data processing method and system based on artificial intelligence |
| CN117995412A (en) * | 2024-04-07 | 2024-05-07 | 粤港澳大湾区数字经济研究院(福田) | Future incidence probability prediction method, device, terminal and storage medium |
-
2022
- 2022-01-11 CN CN202210026966.7A patent/CN114372201A/en active Pending
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116030984A (en) * | 2023-03-31 | 2023-04-28 | 武汉携康智能健康设备有限公司 | User physical examination system and physical examination method based on intelligent health station |
| CN116759061A (en) * | 2023-08-17 | 2023-09-15 | 简阳市人民医院 | Physical examination project recommendation system based on personal demands |
| CN116759061B (en) * | 2023-08-17 | 2023-10-27 | 简阳市人民医院 | Physical examination project recommendation system based on personal demands |
| CN117951190A (en) * | 2024-03-18 | 2024-04-30 | 深圳市双佳医疗科技有限公司 | Human body index abnormal data processing method and system based on artificial intelligence |
| CN117951190B (en) * | 2024-03-18 | 2024-07-09 | 深圳市双佳医疗科技有限公司 | Human body index abnormal data processing method and system based on artificial intelligence |
| CN117995412A (en) * | 2024-04-07 | 2024-05-07 | 粤港澳大湾区数字经济研究院(福田) | Future incidence probability prediction method, device, terminal and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10896763B2 (en) | System and method for providing model-based treatment recommendation via individual-specific machine learning models | |
| JP7566939B2 (en) | Selecting Audio Features to Build a Model to Detect Medical Conditions | |
| CN114372201A (en) | Physical examination information intelligent recommendation method and system, storage medium and computing equipment | |
| CA3155809A1 (en) | Acoustic and natural language processing models for speech-based screening and monitoring of behavioral health conditions | |
| CN110570941B (en) | System and device for assessing psychological state based on text semantic vector model | |
| CN112470143A (en) | Dementia prediction device, prediction model generation device, and dementia prediction program | |
| CN111145903B (en) | Method and device for acquiring vertigo inquiry text, electronic equipment and inquiry system | |
| CN111933291A (en) | Medical information recommendation device, method, system, equipment and readable storage medium | |
| WO2022257630A1 (en) | Risk detection method and apparatus based on multi-modal concealed information test | |
| CN115221941A (en) | Cognitive disorder detection method and related device, electronic equipment and storage medium | |
| Feng | Toward knowledge-driven speech-based models of depression: Leveraging spectrotemporal variations in speech vowels | |
| Walker et al. | Beyond percent correct: Measuring change in individual picture naming ability | |
| Krstev et al. | Multimodal data fusion for automatic detection of alzheimer’s disease | |
| Özkanca et al. | Multi-lingual depression-level assessment from conversational speech using acoustic and text features | |
| Sindhu et al. | Automatic Speech and Voice Disorder Detection using Deep Learning-A Systematic Literature Review | |
| CN110648754A (en) | Department recommendation method, device and equipment | |
| CN118230952A (en) | Psychological assessment method and system based on BPRS (Business Process reference System) concise psychosis table | |
| Ketpupong et al. | Applying text mining for classifying disease from symptoms | |
| CN113808709B (en) | Psychological elasticity prediction method and system based on text analysis | |
| CN114863911A (en) | Parkinson prediction method and device based on voice signals | |
| CN113761899A (en) | Medical text generation method, device, equipment and storage medium | |
| CN112508603A (en) | Method and device for mining potential customer information of endowment community | |
| CN118471540B (en) | Cardiovascular case data processing method and system | |
| CN116913527B (en) | Hypertension evaluation method and system based on multi-round dialogue frame | |
| Maheshwar et al. | Development of an SVM-based Depression Detection Model using MFCC Feature Extraction |
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 |