CN119049734A - Psychological disease early warning system and method - Google Patents
Psychological disease early warning system and method Download PDFInfo
- Publication number
- CN119049734A CN119049734A CN202411552023.3A CN202411552023A CN119049734A CN 119049734 A CN119049734 A CN 119049734A CN 202411552023 A CN202411552023 A CN 202411552023A CN 119049734 A CN119049734 A CN 119049734A
- Authority
- CN
- China
- Prior art keywords
- early warning
- psychological
- data
- module
- 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.)
- Granted
Links
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
- 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
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2431—Multiple classes
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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 OR CALCULATING; 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
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
- G06N3/0985—Hyperparameter optimisation; Meta-learning; Learning-to-learn
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computing Systems (AREA)
- Evolutionary Biology (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- Databases & Information Systems (AREA)
- Pathology (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention discloses a psychological disease early warning system and method, and belongs to the technical field of psychological disease prevention. The system comprises a data acquisition module, a model training module, a disease prediction module, an early warning interaction module, a feedback optimization module and a feedback optimization module, wherein the data acquisition module processes raw data to generate a target data set, the model training module analyzes and learns the target data set through a machine learning algorithm to generate a large model for predicting psychological diseases, the disease prediction module predicts psychological diseases in the target data set through the large model and generates corresponding predicted values, the early warning interaction module receives inquiry requests of users, after the users are judged to need to be early warned through combining the predicted values, psychological condition early warning is carried out, and the feedback optimization module adjusts model parameters corresponding to the large model according to early warning accuracy reports and reporting records. Through integrating a machine learning model and big data analysis, the system can effectively perform psychological disease early warning and provide accurate and deep judgment results.
Description
Technical Field
The invention relates to the technical field of psychological disease prevention, in particular to a psychological disease early warning system and method.
Background
In the current technical field of mental disease prevention, the traditional mental disease identification mainly has the function of screening the mental disease only through the psychology scale, and a mental disease early warning system often carries out rough prediction according to related data. These systems have significant limitations in dealing with college scenarios, including lack of psychological disease labels, prediction accuracy, and the like. Along with increasing psychological problems of universities, internet and internet of things technologies are popularized in the universities, and higher requirements are put on psychological disease early warning technologies.
In addition, conventional psychological disease early warning systems often lack the ability to capture and analyze behavioral data, employ deep learning model predictions, and lack control over the degree of accuracy of predictions. Greatly increases the cost of solving the psychological diseases in the next step. This results in a shortage of user experience and a low early warning efficiency.
With the development of artificial intelligence and machine learning technologies, particularly in advanced learning and reinforcement learning methods, a new opportunity has emerged to solve the limitations of the traditional psychological disease early warning system by using these advanced technologies. An effective modern psychological disease early warning system can perform self-learning through user feedback data, continuously optimize system parameters and perform more accurate early warning, so that the problem of difficult psychological disease early warning in universities is better solved.
The foregoing is provided merely for the purpose of facilitating understanding of the technical solutions of the present invention and is not intended to represent an admission that the foregoing is prior art.
Disclosure of Invention
The invention mainly aims to provide a psychological disease early warning system and a psychological disease early warning method, and aims to solve the technical problem of how to accurately perform psychological disease early warning.
In order to achieve the aim, the invention provides a psychological disease early warning system, which comprises a data acquisition module, a model training module, a disease prediction module, an early warning interaction module and a feedback optimization module;
the data acquisition module is used for processing the original data to generate a target data set, and sending the target data set to the model training module, wherein the original data comprises psychological investigation data, daily behavior data and social relationship data;
the model training module is used for analyzing and learning the target data set through a machine learning algorithm to generate a large model for psychological disease prediction;
the disease prediction module is used for predicting psychological disease conditions in the target data set by utilizing the large model and generating corresponding predicted values;
the early warning interaction module is used for receiving a query request of a user, and carrying out psychological condition early warning after the user needs to be early warned by combining the predicted value;
The early warning interaction module further comprises an early warning display sub-module, an early warning confirmation sub-module and an early warning reporting sub-module;
the early warning display sub-module is used for carrying out data visualization display according to the query request of the user when carrying out psychological condition early warning;
The early warning confirmation sub-module is used for generating an early warning accuracy report according to the feedback information of the user received after the psychological condition early warning is carried out;
the early warning and reporting sub-module is used for reporting the target object which is not subjected to psychological condition early warning but has psychological disease diagnosis to generate reporting records;
and the feedback optimization module is used for adjusting model parameters corresponding to the large model according to the early warning accuracy report and the reporting record.
Optionally, the data acquisition module is further configured to acquire standard mental scale surveys and mental interview data to generate the mental survey data, and extract psychological symptom information from the mental interview data using text mining techniques and natural language processing algorithms, the psychological symptom information including physical discomfort symptoms, compulsive symptoms, symptoms of interpersonal disorders, depression symptoms, anxiety symptoms, hostile emotion symptoms, fear symptoms, paranoid symptoms, and psychotic symptoms;
Acquiring daily behavior data through a preset first address set, wherein the daily behavior data comprise cultural lessons, class attendance, sports test achievements, physiological function indexes, campus entrance and exit conditions, dining room dining conditions, dormitory entrance and exit time distribution and dormitory time;
Acquiring social relationship data through combining a preset second address set with a survey report, wherein the social relationship data comprises family conditions, family income conditions, love conditions, marital conditions and interpersonal interaction conditions;
data preprocessing the psychological survey data, the daily behavioral data and the social relationship data to generate a target data set, wherein the preprocessing comprises data cleaning, data standardization and data classification.
Optionally, the model training module is further configured to classify the target data set according to the psychological survey data result, classify the target data set from 6 angles including depression, anxiety disorder, stress disorder, eating disorder, sleep disorder, addiction disorder by using a five-item scoring method, and mark asymptomatic, mild symptom, middle symptom, heavier symptom, and severe symptom as 0, 0.25, 0.5, 0.75, and 1 in each classification result;
Constructing a recurrent neural network RNN model to extract data characteristics F:
Wherein R is the output result of the recurrent neural network RNN model, AndRespectively representing the weight and the bias of the full connection layer;
Defining a loss function, and performing model training by using an Adam optimizer;
And performing cross validation on a preset data set to adjust model parameters corresponding to the recurrent neural network RNN model.
Optionally, the disease prediction module is further configured to generate a psychological disease prediction value through the trained large model parameters, where P is a 6-dimensional vector, each componentRepresenting depression, anxiety disorders, stress disorders, eating disorders, sleep disorders, and addiction disorders, respectively;
Early warning value The calculation formula of (2) is as follows: Wherein A parameter matrix representing a large model is presented,Representing data within the target data set.
Optionally, the early warning display sub-module is further configured to receive query information input by a user, and determine a query type according to keywords in the query information, where the query model includes a personal psychological early warning query and a group psychological early warning query;
Matching corresponding disease prediction data according to the query type;
Generating early warning information according to the disease prediction data and carrying out data visualization display on the early warning information.
Optionally, the human psychology early warning query includes retrieval of personal keywords, personal retrieval results are displayed in a six-dimensional radar chart mode, and similarity between a user query vector q and feature vectors of each item of disease information in the disease prediction module is calculated;
The calculation formula of cosine similarity is as follows:
Where d represents a feature vector of certain disease information in the disease prediction module.
Optionally, the group psychological early warning query includes searching by taking a group name as a keyword, displaying a group searching result in a six-dimensional radar chart mode, and calculating an overall data mean value:
where x represents the data of all individuals of a certain index, and N represents the dimension of the vector x.
In addition, in order to achieve the above objective, the present invention further provides a psychological disease early warning method, the method is applied to the psychological disease early warning system, the system includes a data acquisition module, a model training module, a disease prediction module, an early warning interaction module and a feedback optimization module, the early warning interaction module includes an early warning display sub-module, an early warning confirmation sub-module and an early warning reporting sub-module, the method includes:
the data acquisition module processes the original data to generate a target data set, and sends the target data set to the model training module, wherein the original data comprises psychological investigation data, daily behavior data and social relationship data;
the model training module analyzes and learns the target data set through a machine learning algorithm to generate a large model for psychological disease prediction;
The disease prediction module predicts psychological disease conditions in the target data set by using the large model and generates a corresponding predicted value;
the early warning interaction module receives a query request of a user, and carries out psychological condition early warning after the early warning interaction module judges that the user needs to be subjected to early warning in combination with the predicted value;
the early warning display sub-module performs data visualization display according to the query request of the user when performing psychological condition early warning;
The early warning confirmation sub-module generates an early warning accuracy report according to the feedback information of the user received after the psychological condition early warning is carried out;
The early warning and reporting sub-module reports the target object which is not subjected to psychological condition early warning but has psychological disease diagnosis to generate reporting records;
and the feedback optimization module adjusts model parameters corresponding to the large model according to the early warning accuracy report and the reporting record.
Optionally, the step of processing the raw data to generate a target data set and sending the target data set to the model training module includes:
Collecting standard mental scale surveys and mental interview data to generate the mental survey data, and extracting psychological symptom information from the mental interview data using text mining techniques and natural language processing algorithms, wherein the psychological symptom information comprises physical discomfort symptoms, obsessive-compulsive symptoms, interpersonal disorder symptoms, depression symptoms, anxiety symptoms, hostile emotion symptoms, fear symptoms, paranoid symptoms and psychogenic disease symptoms;
Acquiring daily behavior data through a preset first address set, wherein the daily behavior data comprise cultural lessons, class attendance, sports test achievements, physiological function indexes, campus entrance and exit conditions, dining room dining conditions, dormitory entrance and exit time distribution and dormitory time;
Acquiring social relationship data through combining a preset second address set with a survey report, wherein the social relationship data comprises family conditions, family income conditions, love conditions, marital conditions and interpersonal interaction conditions;
data preprocessing the psychological survey data, the daily behavioral data and the social relationship data to generate a target data set, wherein the preprocessing comprises data cleaning, data standardization and data classification.
Optionally, the step of analyzing and learning the target data set by a machine learning algorithm to generate a large model for psychological disease prediction comprises classifying the target data set according to psychological investigation data results, classifying the target data set from 6 angles of depression, anxiety disorder, stress disorder, eating disorder, sleep disorder and addiction disorder by adopting a five-item scoring method, and marking asymptomatic, mild symptom, middle symptom, heavier symptom and serious symptom as 0, 0.25, 0.5, 0.75 and 1 respectively in each classification result;
Constructing a recurrent neural network RNN model to extract data characteristics F:
Wherein R is the output result of the recurrent neural network RNN model, AndRespectively representing the weight and the bias of the full connection layer;
Defining a loss function, and performing model training by using an Adam optimizer;
And performing cross validation on a preset data set to adjust model parameters corresponding to the recurrent neural network RNN model.
The system comprises a data acquisition module, a model training module, a disease prediction module, an early warning interaction module, a feedback optimization module and a feedback optimization module, wherein the data acquisition module processes raw data to generate a target data set, the model training module analyzes and learns the target data set through a machine learning algorithm to generate a large model for predicting psychological diseases, the disease prediction module predicts psychological diseases in the target data set through the large model and generates corresponding predicted values, the early warning interaction module receives inquiry requests of users, after the users are judged to need early warning through combining the predicted values, psychological condition early warning is carried out, and the feedback optimization module adjusts model parameters corresponding to the large model according to early warning accuracy reports and reporting records. Through integrating a machine learning model and big data analysis, the system can effectively perform psychological disease early warning and provide accurate and deep judgment results.
Drawings
FIG. 1 is a block diagram of a psychological disease pre-warning system according to a first embodiment of the present invention;
fig. 2 is a flowchart of a psychological disease pre-warning method according to a first embodiment of the present invention.
The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Referring to fig. 1, an embodiment of the present invention provides a psychological disease early warning system, and fig. 1 is a block diagram of a first embodiment of the psychological disease early warning system according to the present invention.
The psychological disease early warning system in the embodiment is composed of a data acquisition module 10, a model training module 20, a disease prediction module 30, an early warning interaction module 40 and a feedback optimization module 50;
The data acquisition module 10 is used for processing the original data to generate a target data set, and sending the target data set to the model training module 20, wherein the original data comprises psychological investigation data, daily behavior data and social relationship data;
A model training module 20 for analyzing and learning the target data set by a machine learning algorithm to generate a large model for psychological disease prediction;
a disease prediction module 30, configured to predict a psychological disease condition in the target data set by using the large model and generate a corresponding predicted value;
the early warning interaction module 40 is configured to receive a query request of a user, and perform psychological condition early warning after combining with the predicted value to determine that early warning is required to be performed on the user;
The early warning interaction module 40 also comprises an early warning display sub-module 401, an early warning confirmation sub-module 402 and an early warning reporting sub-module 403;
The early warning display sub-module 401 is used for performing data visualization display according to a query request of a user when performing psychological condition early warning;
An early warning confirmation sub-module 402, configured to generate an early warning accuracy report according to feedback information of a receiving user after performing psychological condition early warning;
The early warning and reporting sub-module 403 is configured to report a target object that is not subject to psychological condition early warning but has a psychological disease diagnosis to generate a reporting record;
The feedback optimization module 50 is configured to adjust model parameters corresponding to the large model according to the early warning accuracy report and the reporting record.
The data acquisition module is also used for acquiring standard psychological scale investigation and psychological interview data to generate psychological investigation data, extracting psychological symptom information from the psychological interview data by adopting a text mining technology and a natural language processing algorithm, wherein the psychological symptom information comprises physical discomfort symptoms, compulsive symptoms, interpersonal relation disorder symptoms, depression symptoms, anxiety symptoms, hostile emotion symptoms, fear symptoms, paranoid symptoms and mental disease symptoms, acquiring daily behavior data through a preset first address set, wherein the daily behavior data comprises cultural class achievements and changes, classroom attendance rates, sports test achievements and changes, physiological function indexes and changes, campus in-out conditions, dining conditions, in-out-of-dormitory time distribution and sleeping time, acquiring social relation data through a preset second address set in combination with investigation reports, wherein the social relation data comprises family conditions, marital conditions and interpersonal interaction conditions, and performing data preprocessing on the daily behavior data and the social relation data to generate a target data set, wherein the preprocessing comprises cleaning, income data standards and data classification.
It should be noted that, in this embodiment, the first address set refers to acquiring daily behavior data through the campus internet of things system and the educational administration system, where the daily behavior data mainly corresponds to behavior data of students.
It may be understood that, in this embodiment, the second address set refers to a student status system and a student file system, and the step of acquiring social relationship data by combining a survey report with the preset second address set refers to acquiring social relationship data about students including a family situation, a family income situation, a love situation, a marital situation and an interpersonal interaction situation by the student status system and the student file system and simultaneously adopting a survey interview, and combining the data and a text processing system.
It can be understood that the model training module is further configured to classify the target data set according to the psychological survey data result, classify the target data set from 6 angles of depression, anxiety disorder, stress disorder, eating disorder, sleep disorder, addiction disorder by adopting a five-item scoring method, and mark asymptomatic, mild symptom, middle symptom, heavier symptom and severe symptom as 0, 0.25, 0.5, 0.75 and 1 in each classification result respectively, and construct a recurrent neural network RNN model to extract data characteristics F:
Wherein R is the output result of the recurrent neural network RNN model, AndRespectively representing the weight and the bias of the full connection layer;
Defining a loss function, and performing model training by using an Adam optimizer;
And performing cross validation on the preset data set to adjust model parameters corresponding to the recurrent neural network RNN model.
In an implementation, the model training module is further configured to structure data processing (RNN part):
structured data such as psychological survey data, daily behavioral data, social relationship data, and the like are processed using Recurrent Neural Networks (RNNs). The RNN can process the sequence data through its cyclic structure, preserve the time information of the data, and for each time step t, the calculation formula of the RNN is:
Wherein the method comprises the steps of Is the input of a time step t,Is the hidden state of the time step t,、And b is a network parameter. Final stateInformation representing the entire input sequence.
Loss function and optimization algorithm:
the loss function combines the cross-over loss (classification task) and the mean square error loss (regression task), and for the classification task, the cross-over loss calculation formula is:
Wherein the method comprises the steps of Is a real tag that is not a real tag,Is the prediction probability, and for the regression task, the mean square error loss calculation formula is:
Wherein the method comprises the steps of Is a true value of the code,Is a predicted value.
The optimization algorithm is that an Adam optimizer is used, the advantages of momentum and self-adaptive learning rate are combined, and an update formula of the Adam optimizer is as follows:
Where θ is a parameter, η is a learning rate, AndIs a first order and second order moment estimate, corresponding to the mean and variance of the gradient, respectively, epsilon being a small constant to prevent division by zero.
The early warning display sub-module is also used for receiving query information input by a user, determining a query type according to keywords in the query information, and the query model comprises personal psychological early warning query and group psychological early warning query;
matching corresponding disease prediction data according to the query type;
generating early warning information according to the disease prediction data and carrying out data visualization display on the early warning information.
The early warning confirmation submodule specifically comprises the steps that a user feeds back in a system through investigation of real conditions, marks a predicted result, confirms a determinable psychological disease problem, refutes an incorrect psychological disease problem and records the incorrect psychological disease problem.
The early warning and reporting submodule specifically comprises the steps of reporting the university student who has no early warning but has confirmed psychological diseases or suspected psychological diseases to the user, and recording the reporting condition by the system.
It can be understood that the human psychology early warning query comprises retrieval of personal keywords, personal retrieval results are displayed in a six-dimensional radar chart mode, and the similarity between a user query vector q and a feature vector of each item of disease information in the disease prediction module is calculated;
The calculation formula of cosine similarity is as follows:
Where d represents a feature vector of certain disease information in the disease prediction module.
In a specific implementation, the feedback optimization module specifically includes early warning confirmation or declared data feedback, monitors system performance indexes including query response time and error rate, identifies model parameters affecting query accuracy and response speed, and adjusts the model parameters based on the feedback, specifically as follows:
if the feedback of the user shows that the query result is not relevant enough, adjusting a model weight updating rule or increasing the training data volume;
And (3) adjusting the learning rate eta, namely if the accuracy of the query result is not obviously improved, reducing the learning rate to avoid oscillation caused by the overlarge step length in the optimization process, otherwise, if the model is too slow to converge, and properly increasing the learning rate.
Optimizing the number of layers and the number of neurons by increasing or decreasing the number of network layers and/or the number of neurons, adjusting the complexity of the model, too many layers or neurons may result in an overfitting, while too few may result in a underfilling.
And (3) adjusting the activation function, namely selecting an appropriate activation function (such as ReLU, sigmoid, tanh and the like) according to the model performance and the training effect so as to improve the nonlinear fitting capacity of the model.
Response speed optimization, namely adjusting a model according to system performance data, particularly response time, so as to improve efficiency. For example, for slower response times, the speed of reasoning can be increased by reducing model complexity or applying model compression techniques.
Batch and parallel processing techniques are implemented to optimize the processing power of the system, reducing the average response time of a single query.
The system also comprises a user interface module, which specifically comprises:
Visual interface layout:
And the clear interface layout is designed, so that the user can quickly understand and use the interface layout. The main functions such as search box, inquiry button, result display area, etc. should be visually displayed on the main interface. For complex query results, such as images, tables, or text, appropriate layout and visualization techniques are used for ease of interpretation by the user. The integrated advanced search options allow users to customize the query accurately, including screening by class, college or relevance, and allow users to evaluate or feed back the query results.
The data acquisition module processes the original data to generate a target data set, the model training module analyzes and learns the target data set through a machine learning algorithm to generate a large model for predicting psychological diseases, the disease prediction module predicts psychological diseases in the target data set through the large model and generates corresponding predicted values, the early warning interaction module receives query requests of users, after the users are judged to need to be early warned by combining the predicted values, psychological condition early warning is conducted, and the feedback optimization module adjusts model parameters corresponding to the large model according to early warning accuracy reports and reporting records. Through integrating a machine learning model and big data analysis, the system can effectively perform psychological disease early warning and provide accurate and deep judgment results.
Further, an embodiment of the present invention provides a psychological disease early warning method, referring to fig. 2, and fig. 2 is a schematic flow chart of a first embodiment of a psychological disease early warning method of the present invention.
In this embodiment, the psychological disease early warning method includes the following steps:
and S10, the data acquisition module processes the original data to generate a target data set, and the target data set is sent to the model training module, wherein the original data comprises psychological investigation data, daily behavior data and social relationship data.
The model training module analyzes and learns the target data set through a machine learning algorithm to generate a large model for psychological disease prediction S20.
And S30, predicting the psychological disease condition in the target data set by using the large model by using the disease prediction module and generating a corresponding predicted value.
And S40, the early warning interaction module receives the inquiry request of the user, and carries out psychological condition early warning after the early warning interaction module judges that the user needs to be early warned by combining the predicted value.
And S50, when the early warning display sub-module performs psychological condition early warning, data visualization display is performed according to the query request of the user.
And step 60, the early warning confirmation sub-module generates an early warning accuracy report according to the feedback information of the user after the psychological condition early warning is carried out.
And step S70, the early warning and reporting sub-module reports the target object which is not subjected to psychological condition early warning but has psychological disease diagnosis to generate reporting records.
And S80, the feedback optimization module adjusts model parameters corresponding to the large model according to the early warning accuracy report and the reporting record.
The data acquisition module processes the original data to generate a target data set, the model training module analyzes and learns the target data set through a machine learning algorithm to generate a large model for predicting psychological diseases, the disease prediction module predicts psychological diseases in the target data set through the large model and generates corresponding predicted values, the early warning interaction module receives query requests of users, after the users are judged to need to be early warned by combining the predicted values, psychological condition early warning is conducted, and the feedback optimization module adjusts model parameters corresponding to the large model according to early warning accuracy reports and reporting records. Through integrating a machine learning model and big data analysis, the system can effectively perform psychological disease early warning and provide accurate and deep judgment results.
It should be understood that the foregoing is illustrative only and is not limiting, and that in specific applications, those skilled in the art may set the invention as desired, and the invention is not limited thereto.
It should be noted that the above-described working procedure is merely illustrative, and does not limit the scope of the present invention, and in practical application, a person skilled in the art may select part or all of them according to actual needs to achieve the purpose of the embodiment, which is not limited herein.
In addition, technical details not described in detail in this embodiment may refer to the psychological disease early warning system and method provided by any embodiment of the present invention, and are not described herein.
Furthermore, it should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system 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 system. Without further limitation, an element defined by the phrase "comprising one does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
The foregoing embodiment numbers of the present invention are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.
From the above description of embodiments, it will be clear to a person skilled in the art that the above embodiment method may be implemented by means of software plus a necessary general hardware platform, but may of course also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. Read Only Memory (ROM)/RAM, magnetic disk, optical disk) and comprising several instructions for causing a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to perform the method according to the embodiments of the present invention.
The foregoing description is only of the preferred embodiments of the present invention, and is not intended to limit the scope of the invention, and all equivalent structures or equivalent processes using the descriptions and drawings of the present invention or directly or indirectly applied to other related technical fields are included in the scope of the present invention.
Claims (10)
1. The psychological disease early warning system is characterized by comprising a data acquisition module, a model training module, a disease prediction module, an early warning interaction module and a feedback optimization module;
the data acquisition module is used for processing the original data to generate a target data set, and sending the target data set to the model training module, wherein the original data comprises psychological investigation data, daily behavior data and social relationship data;
the model training module is used for analyzing and learning the target data set through a machine learning algorithm to generate a large model for psychological disease prediction;
the disease prediction module is used for predicting psychological disease conditions in the target data set by utilizing the large model and generating corresponding predicted values;
the early warning interaction module is used for receiving a query request of a user, and carrying out psychological condition early warning after the user needs to be early warned by combining the predicted value;
The early warning interaction module further comprises an early warning display sub-module, an early warning confirmation sub-module and an early warning reporting sub-module;
the early warning display sub-module is used for carrying out data visualization display according to the query request of the user when carrying out psychological condition early warning;
The early warning confirmation sub-module is used for generating an early warning accuracy report according to the feedback information of the user received after the psychological condition early warning is carried out;
the early warning and reporting sub-module is used for reporting the target object which is not subjected to psychological condition early warning but has psychological disease diagnosis to generate reporting records;
and the feedback optimization module is used for adjusting model parameters corresponding to the large model according to the early warning accuracy report and the reporting record.
2. The psychological illness early warning system of claim 1 wherein the data collection module is further configured to collect standard psychological scale surveys and psychological interview data to generate the psychological survey data, and to extract psychological symptom information from the psychological interview data using text mining techniques and natural language processing algorithms, the psychological symptom information including physical discomfort symptoms, obsessive-compulsive symptoms, human relationship disorder symptoms, depression symptoms, anxiety symptoms, hostile mood symptoms, fear symptoms, paranoid symptoms, and psychotic symptoms;
Acquiring daily behavior data through a preset first address set, wherein the daily behavior data comprise cultural lessons, class attendance, sports test achievements, physiological function indexes, campus entrance and exit conditions, dining room dining conditions, dormitory entrance and exit time distribution and dormitory time;
Acquiring social relationship data through combining a preset second address set with a survey report, wherein the social relationship data comprises family conditions, family income conditions, love conditions, marital conditions and interpersonal interaction conditions;
data preprocessing the psychological survey data, the daily behavioral data and the social relationship data to generate a target data set, wherein the preprocessing comprises data cleaning, data standardization and data classification.
3. The psychological disease early warning system according to claim 1, wherein the model training module is further configured to classify the target data set according to the psychological survey data result, classify the target data set from 6 angles of depression, anxiety disorder, stress disorder, eating disorder, sleep disorder, addiction disorder by five scoring method, and mark asymptomatic, mild symptom, moderate symptom, heavier symptom, severe symptom as 0, 0.25, 0.5, 0.75, 1 in each classification result;
Constructing a recurrent neural network RNN model to extract data characteristics F:
Wherein R is the output result of the recurrent neural network RNN model, AndRespectively representing the weight and the bias of the full connection layer;
Defining a loss function, and performing model training by using an Adam optimizer;
And performing cross validation on a preset data set to adjust model parameters corresponding to the recurrent neural network RNN model.
4. The psychological disease forewarning system of claim 3, wherein the disease prediction module is further configured to generate psychological disease prediction values from trained large model parameters, wherein P is a 6-dimensional vector, each componentRepresenting depression, anxiety disorders, stress disorders, eating disorders, sleep disorders, and addiction disorders, respectively;
Early warning value The calculation formula of (2) is as follows: Wherein A parameter matrix representing a large model is presented,Representing data within the target data set.
5. The psychological disease early warning system according to claim 1, wherein the early warning display sub-module is further configured to receive query information input by a user, and determine a query type according to keywords in the query information, and the query model includes a personal psychological early warning query and a group psychological early warning query;
Matching corresponding disease prediction data according to the query type;
Generating early warning information according to the disease prediction data and carrying out data visualization display on the early warning information.
6. The psychological illness early warning system according to claim 5, wherein the personal psychological early warning query includes retrieval of personal keywords, personal retrieval results are displayed in a six-dimensional radar chart mode, and similarity between a user query vector q and feature vectors of each item of illness information in the illness prediction module is calculated;
The calculation formula of cosine similarity is as follows:
Where d represents a feature vector of certain disease information in the disease prediction module.
7. The psychological illness early-warning system of claim 5 wherein the group psychological early-warning query includes a search with group names as keywords, the group search results are presented in a six-dimensional radar chart, and the overall data mean is calculated:
where x represents the data of all individuals of a certain index, and N represents the dimension of the vector x.
8. The utility model provides a psychological disease early warning method which is characterized in that the method is applied to psychological disease early warning system, the system includes data acquisition module, model training module, disease prediction module, early warning interaction module and feedback optimization module, early warning interaction module includes early warning display submodule, early warning confirm submodule and early warning declare submodule, the method includes:
the data acquisition module processes the original data to generate a target data set, and sends the target data set to the model training module, wherein the original data comprises psychological investigation data, daily behavior data and social relationship data;
the model training module analyzes and learns the target data set through a machine learning algorithm to generate a large model for psychological disease prediction;
The disease prediction module predicts psychological disease conditions in the target data set by using the large model and generates a corresponding predicted value;
the early warning interaction module receives a query request of a user, and carries out psychological condition early warning after the early warning interaction module judges that the user needs to be subjected to early warning in combination with the predicted value;
the early warning display sub-module performs data visualization display according to the query request of the user when performing psychological condition early warning;
The early warning confirmation sub-module generates an early warning accuracy report according to the feedback information of the user received after the psychological condition early warning is carried out;
The early warning and reporting sub-module reports the target object which is not subjected to psychological condition early warning but has psychological disease diagnosis to generate reporting records;
and the feedback optimization module adjusts model parameters corresponding to the large model according to the early warning accuracy report and the reporting record.
9. The psychological disease pre-warning method of claim 8, wherein the step of processing the raw data to generate a target data set, and transmitting the target data set to the model training module, comprises:
Collecting standard mental scale surveys and mental interview data to generate the mental survey data, and extracting psychological symptom information from the mental interview data using text mining techniques and natural language processing algorithms, wherein the psychological symptom information comprises physical discomfort symptoms, obsessive-compulsive symptoms, interpersonal disorder symptoms, depression symptoms, anxiety symptoms, hostile emotion symptoms, fear symptoms, paranoid symptoms and psychogenic disease symptoms;
Acquiring daily behavior data through a preset first address set, wherein the daily behavior data comprise cultural lessons, class attendance, sports test achievements, physiological function indexes, campus entrance and exit conditions, dining room dining conditions, dormitory entrance and exit time distribution and dormitory time;
Acquiring social relationship data through combining a preset second address set with a survey report, wherein the social relationship data comprises family conditions, family income conditions, love conditions, marital conditions and interpersonal interaction conditions;
data preprocessing the psychological survey data, the daily behavioral data and the social relationship data to generate a target data set, wherein the preprocessing comprises data cleaning, data standardization and data classification.
10. The method for pre-warning psychological diseases according to claim 8, wherein the step of analyzing and learning the target data set by the machine learning algorithm to generate a large model for psychological disease prediction comprises classifying the target data set according to psychological survey data results, classifying the target data set from 6 angles of depression, anxiety disorder, stress and stress disorder, eating disorder, sleep disorder, addiction disorder by five scoring method, and marking asymptomatic, slight symptom, middle symptom, heavier symptom, serious symptom as 0, 0.25, 0.5, 0.75, 1 in each classification result;
Constructing a recurrent neural network RNN model to extract data characteristics F:
Wherein R is the output result of the recurrent neural network RNN model, AndRespectively representing the weight and the bias of the full connection layer;
Defining a loss function, and performing model training by using an Adam optimizer;
And performing cross validation on a preset data set to adjust model parameters corresponding to the recurrent neural network RNN model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411552023.3A CN119049734B (en) | 2024-11-01 | 2024-11-01 | Psychological disease early warning system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411552023.3A CN119049734B (en) | 2024-11-01 | 2024-11-01 | Psychological disease early warning system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN119049734A true CN119049734A (en) | 2024-11-29 |
| CN119049734B CN119049734B (en) | 2025-01-17 |
Family
ID=93582436
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411552023.3A Active CN119049734B (en) | 2024-11-01 | 2024-11-01 | Psychological disease early warning system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN119049734B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119672618A (en) * | 2025-02-21 | 2025-03-21 | 绵阳市第三人民医院 | A deep learning-based obsessive-compulsive disorder identification and intervention method and intervention system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017015393A1 (en) * | 2015-07-21 | 2017-01-26 | The Arizona Board Of Regents On Behalf Of The University Of Arizona | Health information (data) medical collection, processing and feedback continuum systems and methods |
| WO2024026283A1 (en) * | 2022-07-26 | 2024-02-01 | Compass Pathfinder Limited | Computing n-dimensional sentiment using a large language model |
| WO2024099098A1 (en) * | 2022-11-07 | 2024-05-16 | 中电科大数据研究院有限公司 | Early warning method and device based on group emotion prediction model, and storage medium |
-
2024
- 2024-11-01 CN CN202411552023.3A patent/CN119049734B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017015393A1 (en) * | 2015-07-21 | 2017-01-26 | The Arizona Board Of Regents On Behalf Of The University Of Arizona | Health information (data) medical collection, processing and feedback continuum systems and methods |
| WO2024026283A1 (en) * | 2022-07-26 | 2024-02-01 | Compass Pathfinder Limited | Computing n-dimensional sentiment using a large language model |
| WO2024099098A1 (en) * | 2022-11-07 | 2024-05-16 | 中电科大数据研究院有限公司 | Early warning method and device based on group emotion prediction model, and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| 袁钦湄;洪志令;王星;帅建伟;曹玉萍;: "人工智能在精神疾病中的应用", 国际精神病学杂志, no. 01, 25 February 2020 (2020-02-25) * |
| 陈晓红: "医疗大模型技术及应用发展研究", 《中国工程科学》, 25 September 2024 (2024-09-25) * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119672618A (en) * | 2025-02-21 | 2025-03-21 | 绵阳市第三人民医院 | A deep learning-based obsessive-compulsive disorder identification and intervention method and intervention system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN119049734B (en) | 2025-01-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Arcinas et al. | Role of data mining in education for improving students performance for social change | |
| Karthikeyan et al. | RETRACTED ARTICLE: Towards developing hybrid educational data mining model (HEDM) for efficient and accurate student performance evaluation | |
| El Morr et al. | Descriptive, predictive, and prescriptive analytics | |
| CN110060772B (en) | Occupational psychological character analysis method based on social network | |
| Aggarwal et al. | Representing uncertainty with information sets | |
| CN119049734B (en) | Psychological disease early warning system and method | |
| WO2022067004A1 (en) | Personal, professional, cultural (ppc) insight system | |
| CN117216419A (en) | Data analysis method based on AI technology | |
| Pryss et al. | Machine learning findings on geospatial data of users from the trackyourstress mhealth crowdsensing platform | |
| Arif et al. | An Improved Prediction System of Students' Performance Using Classification model and Feature Selection Algorithm | |
| CN118035286A (en) | Information query system based on medical large model | |
| Jayawickrama et al. | Ensemble Learning Approach to Human Stress Detection Based on Behaviours During the Sleep | |
| Revathy et al. | Automatic diagnosis of mental illness using optimized dynamically stabilized recurrent neural network | |
| Saranya et al. | ENHANCED PREDICTION OF STUDENT DROPOUTS USING FUZZY INFERENCE SYSTEM AND LOGISTIC REGRESSION. | |
| US20210390390A1 (en) | Multi-task learning framework for multi-context machine learning | |
| Kasthuri et al. | Precaution chatbot for epilepsy patients using natural language processing and deep learning sequential model | |
| KR102671618B1 (en) | Method and system for providing user-customized interview feedback for educational purposes based on deep learning | |
| Dan | Intelligent English resource recommendation and teaching effect based on symmetric SDAE collaborative filtering algorithm | |
| Sahu et al. | Mental health prediction among students using machine learning techniques | |
| Sajja et al. | Using classification data mining for predicting student performance | |
| KR20210060778A (en) | Public system of studying abroad information | |
| Banerjee et al. | A survey on mental health monitoring system via social media data using deep learning framework | |
| Singh et al. | Automatic Criminal Recidivism Risk Estimation in Recidivist Using Classification and Ensemble Techniques | |
| Syakur et al. | Multiclass classification of toddler nutritional status using support vector machine: A case study of community health centers in Bangkalan, Indonesia | |
| Ganesh et al. | Multimodal Data Fusion and Machine Learning for Comprehensive Management of Parkinson's Disease in Healthcare |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |