CN112100286A - Computer-aided decision-making method, device, system and server based on multi-dimensional data - Google Patents

Abstract

The invention discloses a computer-aided decision-making method, device, system and server based on multi-dimensional data. The method includes data acquisition, data processing, data analysis and decision-making; the device includes a collection and input unit, a processing and analysis unit , storage unit, decision-making and display unit; the system includes acquisition and input equipment, distributed server and client equipment; the distributed server is used for distributed queue, database and container arrangement; the information obtained by the system The data is sent to the distributed queue, and the containerized service performs information formatting, preprocessing, and feature extraction, and then stores it in the distributed database. The data analysis service container conducts model training and decision-making, and sends the auxiliary decision-making results to the client. The invention is cross-platform, decentralized, privacy-safe, uses big data and artificial intelligence technology, and is suitable for assisting doctors to use objective biological criteria to assist in decision-making scenarios.

Description

Computer-aided decision-making method, device, system and server based on multi-dimensional data

technical field

The invention relates to a computer-aided decision-making method, device, system and server based on multi-dimensional data, and belongs to the field of computer-aided decision-making.

Background technique

With the development of computer-aided decision-making technology, more and more disease early warning models are used to assist doctors in disease diagnosis and prognosis. However, because the diagnosis of neuropsychiatric diseases temporarily lacks objective biological criteria and relies more on the subjective judgment of doctors, it is very urgent to study the underlying factors of neuropsychiatric diseases and establish disease early warning models. Clinically, the methods for diagnosing neuropsychiatric diseases and their severity are as follows: collection of the subject's medical history, psychiatric examination, and physical and laboratory examinations. The physical and laboratory examinations mainly include: routine examinations, neurophysiological and imaging examinations.

At present, the popular neuropsychiatric disease-assisted decision-making research in academia mainly adopts methods such as statistics or machine learning to analyze the multi-dimensional medical data of the subjects, find potential disease-inducing factors, and establish an effective early warning model. However, due to the large amount of private information contained in patient medical history, scale data and medical examination data, data maintenance and management are very difficult, and an effective unified database is lacking. At the same time, the existing models are all designed for small data sets, which cannot meet the requirements of big data analysis and model modeling of the information collected in the unified database, resulting in research difficulties, and sometimes even erroneous results, which cannot provide doctors with correct and appropriate assistance. .

SUMMARY OF THE INVENTION

The purpose of the present invention is to solve the above-mentioned defects of the prior art, and provide a computer-aided decision-making method, device, system and server based on multi-dimensional data. The computer-aided decision-making scenario of dimensional big data analysis can provide reliable auxiliary decision-making results while ensuring data security. The system users are doctors and subjects, and it has the advantages of complete functions, simple operation, privacy protection, stability and reliability.

The first object of the present invention is to provide a computer-aided decision-making method based on multi-dimensional data.

The second object of the present invention is to provide a multi-dimensional data processing device for computer-aided decision-making.

The third object of the present invention is to provide a data management and analysis system for computer-aided decision-making.

The fourth object of the present invention is to provide a multi-center distributed server for computer-aided decision-making.

The first purpose of the present invention can be achieved by adopting the following technical solutions:

A computer-aided decision-making method based on multi-dimensional data, the method comprising:

Multi-dimensional data collection, the information of the subjects is input, and sent to the distributed queue by the front-end equipment;

Perform information formatting, preprocessing and feature extraction on the data in the queue, and store the processed data in a distributed database;

Perform targeted analysis on the data of each dimension extracted from the distributed database, obtain the corresponding auxiliary decision-making model after training, and store the model in the database;

According to the input data of each dimension, use a specific auxiliary decision-making model for inference and decision-making, output prediction data and similar related type data, and send it to the user-end equipment for doctors and subjects to view.

Further, the multi-dimensional data includes neuroimaging data, scale evaluation data, and biochemical information data, and the subject information includes demographic information, clinical information, and disease history. further,

The auxiliary decision-making model is for specific data, including machine learning models, deep neural network models and graph neural network models, wherein,

The machine learning model takes structured data as input, including calculated neuroimaging and biochemical information characteristic data, as well as quantified scale evaluation, demographic information, clinical information and disease history, and takes the loss function as the evaluation criterion , using the optimization algorithm to solve the machine learning model parameters when the loss is the smallest, and output the feature weights and decision results. The parameter formula is as follows:

θ=argmin(L(y,f(x)))

where θ is the machine learning model parameter, L is the loss function, f is the machine learning model function, y is the label, and x is the input;

The deep neural network model takes image data or the same structured data as the machine learning model as input, takes the loss function as the evaluation criterion, takes the back propagation as the optimization algorithm, and uses the multi-layer neuron to fit the input-output mapping function, Output feature contribution and decision results. The neural network layer operators in the deep neural network model are as follows:

Z=a(W·A)

where Z is the layer output, a is the activation function, W is the weight matrix, and A is the layer input;

The graph neural network model takes the brain connection network node feature vector and the adjacency matrix calculated and constructed from the neuroimaging data as input, and according to its topological structure characteristics, by aggregating network node and connecting edge features, supervised learning and fusion node features and High-level features of network information, discover saliency maps for decision-making targets, and output decision-making results. The graph neural network layer operators in the graph neural network model are as follows:

Z=a(L·H·W)

where Z is the layer output, a is the activation function, L is the Laplacian matrix, H is the layer input, and W is the weight matrix.

The second object of the present invention can be achieved by adopting the following technical solutions:

A multi-dimensional data processing device for computer-aided decision-making, the device comprising:

The collection and input unit is used for the equipment to collect multi-dimensional data and to manually input the information of the test subjects;

The processing and analysis unit is used for formatting information, data preprocessing and feature extraction, and analyzing the data in the database to obtain an auxiliary decision-making model;

A storage unit for storing the queue of data to be processed, the data to be analyzed, the analysis model and the analysis result data;

The decision-making and display unit is used to use the model to make decisions on the input data and output the predicted data and similar related type data, integrate and display the result information, and assist the doctor in the diagnosis of the disease.

The third object of the present invention can be achieved by adopting the following technical solutions:

A data management and analysis system for computer-aided decision-making, the system includes:

Collection and input equipment, used to collect multi-dimensional data, including neuroimaging data, scale evaluation data, biochemical information data, and input subject information, including demographic information, clinical information, and medical history;

Distributed server, used to store pending message queues, manage and schedule data processing and analysis containers, instantiate container services for data processing and analysis, save processed and analyzed data, and send auxiliary decision-making results to the client;

The user-end device is used to receive, integrate and display the auxiliary decision-making results, including model performance indicators, the contribution map of the model's input features, the model's decision-making results and confidence in the input data, clinical information and diagnosis results of similar cases.

Further, the distributed server includes distributed queues, distributed databases and distributed container orchestration;

The distributed queue is used to receive and temporarily store multi-site pending data, and send it to an available server for processing and analysis;

The distributed database, distributed in each data source, is used for storing multi-site data, ensuring data privacy and security, and satisfying distributed data analysis;

The distributed container orchestration is used to deploy containerized services for data processing and analysis, and the master node server performs unified container orchestration to ensure high system availability;

Further, the distributed database is used to store structured and unstructured data in units of subjects, including neuroimaging data, scale evaluation data, biochemical information data, demographic information, clinical information, and medical history. , Data analysis intermediate results, data models and decision aids results.

Further, in the distributed container orchestration, each node server saves all container images, including data processing service containers, data analysis service containers and client service containers, and the master node uniformly performs container creation and service scheduling. .

The fourth object of the present invention can be achieved by adopting the following technical solutions:

A multi-center distributed server for computer-aided decision-making. A single server includes a central processing unit, a graphics processing unit, and a storage unit. When the central processing unit executes a program stored in the storage unit to run a containerized service, the above-mentioned multidimensional Computer-aided decision-making methods for data.

The present invention has the following beneficial effects with respect to the prior art:

1. The computer-aided decision-making system of the present invention adopts a browser/server architecture, which can be used by multiple users, across platforms, and at the same time, which is convenient and fast; the background adopts a distributed server, which can be deployed in multiple sites respectively, and supports decentralized data management. , safe and effective; the user-side learning and use cost is low, user-friendly, and supports customization, especially for doctors who need objective criteria to assist decision-making, it can quickly generate model decision-making results based on data information entered by doctors and data collected by equipment , and combined with disease prediction and similar cases, to provide doctors with detailed and effective result reports to assist doctors in decision-making.

2. The present invention adopts distributed big data management and analysis technology to carry out safe and reliable management and analysis of data distributed in different hospitals and other sites, and can combine big data and artificial intelligence while protecting the privacy and security of data at each site. Intelligent technology analyzes a large amount of multi-dimensional data and uses it to build a decision-making model to obtain reliable and effective auxiliary decision-making results.

3. The visual content displayed by the user terminal of the present invention through the front-end web page includes disease prediction and similar case reports, wherein the disease prediction report includes model performance indicators, model input features contribution graph, model decision results and confidence levels for input data , to help doctors obtain direct disease prediction results while judging the reliability of the model; similar case reports include case information with a high degree of similarity obtained after the background model uses the input data to perform similarity retrieval in the distributed database, including clinical information , diagnosis results, help doctors to further evaluate the patient's condition.

Description of drawings

In order to explain the embodiments of the present invention or the technical solutions in the prior art more clearly, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only These are some embodiments of the present invention, and for those of ordinary skill in the art, other drawings can also be obtained according to the structures shown in these drawings without creative efforts.

1 is a flowchart of a computer-aided decision-making method according to Embodiment 1 of the present invention;

2 is a structural block diagram of a computer-aided decision-making apparatus according to Embodiment 2 of the present invention;

FIG. 3 is an overall structural diagram of a computer-aided decision-making system according to Embodiment 3 of the present invention.

Detailed ways

In order to make the purposes, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments It is a part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present invention. .

Example 1:

As shown in FIG. 1 , the present embodiment provides a computer-aided decision-making method based on multi-dimensional data. The method is implemented by a distributed server and includes the following steps:

S101, multi-dimensional data collection, the information of the subject is input, and sent to the distributed queue by the front-end device.

The multi-dimensional data and testee information collected and entered by the device will be directly sent to the distributed queue, and further data processing will be performed by the containerized service deployed on the distributed server.

In this implementation, multi-dimensional data includes neuroimaging data, scale evaluation data, and biochemical information data. The specific descriptions of neuroimaging data, scale evaluation data, and biochemical information data are as follows:

1. Neuroimaging data: including Structural/Functional Magnetic Resonance Imaging (s/fMRI), Diffusion Tensor Imaging (DTI), Electroencephalogram (EEG);

2. Scale evaluation data: including self-assessment and other scales commonly used in the diagnosis and research of neuropsychiatric diseases, Positive And Negative Syndrome Scale (PANSS), Hamilton Depression, Hamilton Anxiety, Young Mania Crazy rating, UKU side effect scale, Insight and Treatment Attitude Questionnaire (ITAQ), gross functional assessment, Clinical Global Impression (CGI), PHQ-9 depression screening scale , Health Survey Short Form (the MOS itemshort from health survey, referred to as SF-12), symptom self-rating scale (The self-report symptominventory, referred to as SCL-90), general well-being scale (General Well-Being Schedule, referred to as GWB).

3. Biochemical information data: including intestinal flora data and blood index data, the intestinal flora data is specifically the α diversity, β diversity, flora composition analysis, flora difference analysis, KEGG (Kyoto Encyclopedia of Genes and Genomes, Kyoto Encyclopedia of Genes and Genomes) functional analysis results; blood index data are specifically the number of blood cells, immune factors, and oxidative stress.

In this implementation, the subject information includes demographic information, clinical information, and medical history. The specific descriptions of demographic information, clinical information, and medical history are as follows:

1. Demographic information: including basic information such as the age and gender of the subjects;

2. Clinical information: including the subject's physical disease, clinical symptoms, and doctor's clinical assessment;

3. Medical history: including personal medical history and family medical history.

S102. Perform information formatting, preprocessing and feature extraction on the data in the queue, and store the processed data in a distributed database.

In this implementation, the information formatting includes quantifying and structuring the test subject information entered by the doctor to ensure data consistency.

In this implementation, preprocessing and feature extraction include preprocessing and feature extraction of neuroimaging data, and the specific description of the extracted features is as follows:

1. Structural MRI: gray matter volume, white matter volume, and structural connectivity networks;

2. Functional magnetic resonance imaging: regional consistency, low frequency oscillation amplitude, degree centrality and functional connectivity network;

3. Diffusion tensor imaging: fractional anisotropy, mean dispersion, radial dispersion, axial dispersion and tensor network;

4. EEG: Event-related potentials, complexity and topological properties of brain functional networks.

S103: Analyze the data extracted from the distributed database, train an auxiliary decision-making model, and store the data in the database.

In this implementation, the auxiliary decision-making model includes a graph neural network model, a machine learning model, and a deep neural network model. The specific descriptions of the graph neural network model, the machine learning model, and the deep neural network model are as follows:

1. Graph neural network model: According to the topology characteristics of network data, by aggregating node and edge features in network graph data, supervised learning of high-level features including node features and network information, to complete node and graph classification and prediction tasks;

2. Machine learning model: According to the vectorized features of structured data, select the appropriate model from the model space, use the optimization algorithm to fit the decision function, and complete the data discrimination and prediction tasks;

3. Deep neural network model: According to the vectorized characteristics of structured data, the optimization algorithm based on forward deduction and back propagation is used to update the neuron variables in the deep neural network to fit the original model function to complete data discrimination and analysis. prediction task.

S104 , use the specified auxiliary decision-making model to make decisions based on the input data, output the predicted data and data of similar association types, and send them to the client device for the doctor and the subject to view.

In this implementation, the content output to the client device includes disease prediction and similar case data. The specific description of disease prediction and similar case data is as follows:

1. Disease prediction: including model performance indicators, the contribution map of model input features, the model's decision-making results and confidence in the input data, helping doctors to obtain direct disease prediction results while judging the reliability of the model;

2. Similar cases: including the case information with a high degree of similarity obtained after the background model uses the input data to perform similarity retrieval in the distributed database, including clinical information and diagnosis results, to help doctors further evaluate the patient's condition.

The data, models and results in the above steps S102 to S104 will be persistently stored in the distributed database.

Those skilled in the art can understand that all or part of the steps in the methods of the above embodiments can be implemented by instructing relevant hardware through a program, and the corresponding program can be stored in a computer-readable storage medium.

It should be noted that although the method operations of the above-described embodiments are depicted in a particular order in the drawings, this does not require or imply that the operations must be performed in that particular order, or that all illustrated operations must be performed to achieve the desired results . Conversely, the depicted steps may change the order of execution. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined to be performed as one step, and/or one step may be decomposed into multiple steps to be performed.

Example 2:

As shown in FIG. 2 , this embodiment provides a multi-dimensional data processing device for computer-aided decision-making. The device is applied to implement the above method, and includes a collection and input unit 501, a processing and analysis unit 502, a storage unit 503, Decision and display unit 504, the specific functions of each unit are as follows:

The collection and input unit 501 is used for equipment to collect multi-dimensional data, and to manually input subject information;

The processing and analysis unit 502 is used for formatting information, data preprocessing and feature extraction, and analyzing data in the database to obtain a model;

The storage unit 503 is used to store the data queue to be processed, the data to be analyzed, the analysis model and the analysis result data;

The decision-making and display unit 504 is configured to use the model to make decisions on the input data, output disease predictions and similar cases, integrate and display the result information, and assist doctors in diagnosing diseases.

It should be noted that, the apparatus provided in the above-mentioned embodiments only takes the division of the above-mentioned functional units as an example. Functional unit to complete all or part of the functions described above.

Example 3:

As shown in FIG. 3 , this embodiment provides a data management and analysis system for computer-aided decision-making. The system includes a collection and input device, a distributed server, and a client device.

The collection and input equipment collects multi-dimensional data, including neuroimaging data, scale evaluation data, and biochemical information data, and records the subject information, including demographic information, clinical information, and medical history. The collected and entered data will be sent to the distributed queue stored on the distributed server through the http protocol. In addition to this, the distributed server manages and schedules data processing and analysis containers, instantiates container services for data processing and analysis, stores processed and analyzed data, and sends auxiliary decision-making results to the client. The client device receives, integrates and displays the auxiliary decision-making results, including model performance indicators, the contribution map of the model input features, the model's decision results and confidence in the input data, clinical information and diagnosis results of similar cases.

In this implementation, the acquisition and input device integrates a data transmission module, including a processor, a memory, an input/output unit and other components. The processor runs the program stored in the memory, and the input/output unit transmits the acquired data according to the network transmission protocol through the input/output unit. Send to the specified network address.

In this implementation, distributed queues, distributed databases, and distributed container orchestration are deployed on the distributed server. The specific descriptions of distributed queues, distributed databases, and distributed container orchestration are as follows:

1. A distributed queue for receiving and temporarily storing multi-site pending data, and sending it to an available server for processing and analysis;

2. Distributed database, distributed in various data sources, storing structured and unstructured data in units of subjects, including neuroimaging data, scale evaluation data, biochemical information data, demographic information, clinical information, disease information History, data analysis intermediate results, data models and auxiliary decision-making results; its decentralized storage ensures data privacy and security, and satisfies distributed data analysis;

3. Distributed container orchestration is used to deploy containerized services for data processing and analysis. The master node server performs unified container orchestration to ensure high system availability. Each slave node server saves all container images, including data processing service containers. , data analysis service container and client service container, and the master node uniformly performs container creation and service scheduling.

In this implementation, the container image includes a data processing service container, a data analysis service container, and a client service container. The specific descriptions of the data processing service container, the data analysis service container, and the client service container are as follows:

1. Data processing service container: contains the system environment, software dependencies and service interface programs required for data processing services. Its service interface program integrates the preprocessing and feature extraction programs in the above-mentioned embodiment 1, can process corresponding data in a targeted manner, and connect to a distributed database, and save the processing results therein;

2. Data analysis service container: It contains the system environment, software dependencies and service interface programs required by the data analysis service. Its service interface program integrates the auxiliary decision-making model training and deduction program in the above-mentioned embodiment 1, and can use a specific model interface to train and deduce specific data, and connect to a distributed database, and save the obtained auxiliary decision-making model and result data in it. ;

3. Client service container: Contains the system environment, software dependencies, static resources and service interface programs required for client services. Its service interface program can return a specific response and a static interface according to the user's request, connect to the distributed database, integrate the data requested by the user and return it to the user, and display it in the static interface.

For the description of the multi-dimensional data and the content displayed by the user terminal in this embodiment, reference may be made to the foregoing Embodiment 1, and details are not repeated here.

Example 4:

This embodiment provides a multi-center distributed server for computer-aided decision-making, the server includes a central processing unit, a graphics processing unit, and a storage unit, characterized in that the distributed server implements distributed queues, distributed Database and distributed container orchestration. When the server runs the containerized service, the computer-aided decision-making method based on multi-dimensional data of the above-mentioned embodiment 1 is implemented as follows:

Store the multi-dimensional data and testee information sent by the front-end equipment in the distributed queue;

Perform information formatting, preprocessing and feature extraction on the data in the queue, and store the processed data in a distributed database;

Analyze the data extracted from the distributed database, train the auxiliary decision-making model, and store it in the database;

Use the specified auxiliary decision-making model to make decisions based on input data, output prediction data and data of similar association types, and send them to the user-end equipment for doctors and subjects to view.

To sum up, the system of the present invention can realize computer-oriented decision-making, and provide a simple operation and perfect function decision-making system for doctors or subjects. The dimensional data analysis model, as well as the arrangement of distributed queues, databases and containers, protects the privacy of the subjects' medical information, provides a way for big data analysis, improves the reliability of the auxiliary decision-making model, and provides auxiliary decision-making results. .

The above is only a preferred embodiment of the patent of the present invention, but the protection scope of the patent of the present invention is not limited to this. The technical solution and the inventive concept of the invention are equivalently replaced or changed, all belong to the protection scope of the patent of the present invention.

Claims (9)

1. a computer-aided decision-making method based on multi-dimensional data, is characterized in that, described decision-making method comprises: Multi-dimensional data collection, the information of the subjects is input, and sent to the distributed queue by the front-end equipment; Perform information formatting, preprocessing and feature extraction on the data in the queue, and store the processed data in a distributed database; Perform targeted analysis on the data of each dimension extracted from the distributed database, obtain the corresponding auxiliary decision-making model after training, and store the auxiliary decision-making model in the database; According to the input data of each dimension, use the specified auxiliary decision-making model to infer and make decisions, and output the predicted data and similar related type data. 2 . The multi-dimensional data processing and analysis method according to claim 1 , wherein the multi-dimensional data includes neuroimaging data, scale evaluation data, and biochemical information data, and the subject information includes demographic information. 3 . , clinical information, medical history. 3. The multi-dimensional data processing and analysis method according to claim 1, wherein the auxiliary decision-making model comprises a machine learning model, a deep neural network model and a graph neural network model, wherein, The machine learning model takes structured data as input, including calculated neuroimaging and biochemical information characteristic data, as well as quantified scale evaluation, demographic information, clinical information and disease history, and takes the loss function as the evaluation criterion , using the optimization algorithm to solve the machine learning model parameters when the loss is the smallest, and output the feature weights and decision results. The parameter formula is as follows: θ=argmin(L(y,f(x))) where θ is the machine learning model parameter, L is the loss function, f is the machine learning model function, y is the label, and x is the input; The deep neural network model takes image data or the same structured data as the machine learning model as input, takes the loss function as the evaluation criterion, takes the back propagation as the optimization algorithm, and uses the multi-layer neuron to fit the input-output mapping function, Output feature contribution and decision results. The neural network layer operators in the deep neural network model are as follows: Z=a(W·A) where Z is the layer output, a is the activation function, W is the weight matrix, and A is the layer input; The graph neural network model takes the brain connection network node feature vector and the adjacency matrix calculated and constructed from the neuroimaging data as the input, and outputs the significant feature map and the decision result. The graph neural network layer operator in the graph neural network model is as follows: Z=a(L·H·W) where Z is the layer output, a is the activation function, L is the Laplacian matrix, H is the layer input, and W is the weight matrix. 4. A multi-dimensional data processing device for computer-aided decision-making, wherein the device comprises: The collection and input unit is used for the equipment to collect multi-dimensional data and to manually input the information of the test subject; The processing and analysis unit is used for formatting information, data preprocessing and feature extraction, and analyzing the data in the database to obtain an auxiliary decision-making model; A storage unit for storing the queue of data to be processed, the data to be analyzed, the analysis model and the analysis result data; The decision and display unit is used to make decisions on input data using an assisted decision model and output predicted data and similar association type data. 5. A data management and analysis system for computer-aided decision-making, wherein the system comprises: Collection and input equipment, used to collect multi-dimensional data, including neuroimaging data, scale evaluation data, biochemical information data, and input subject information, including demographic information, clinical information, and medical history; Distributed server, used to store pending message queues, manage and schedule data processing and analysis containers, instantiate container services for data processing and analysis, save processed and analyzed data, and send auxiliary decision-making results to the client; The user-end device is used to receive, integrate and display the auxiliary decision-making results, including model performance indicators, the contribution map of the model's input features, the model's decision-making results and confidence on the input data, and clinical information of similar cases. 6. The data management and analysis system according to claim 5, wherein the distributed server comprises a distributed queue, a distributed database and a distributed container orchestration; The distributed queue is used to receive and temporarily store multi-site pending data, and send it to an available server for processing and analysis; The distributed database, distributed in each data source, is used for storing multi-site data, ensuring data privacy and security, and satisfying distributed data analysis; The distributed container orchestration is used to deploy containerized services for data processing and analysis, and the master node server performs unified container orchestration to ensure high system availability. 7. The data management and analysis system according to claim 6, wherein the distributed database is used to store structured and unstructured data in units of subjects, including neuroimaging data, scale evaluation Data, biochemical information data, demographic information, clinical information, medical history, data analysis intermediate results, data models, and decision aids results. 8. The data management and analysis system according to claim 6, wherein in the distributed container arrangement, each node server saves all container images, including data processing service containers, data analysis service containers and The client service container is unified by the master node for container creation and service scheduling. 9. A multi-center distributed server for computer-aided decision-making, a single server includes a central processing unit, a graphics processing unit, and a storage unit, wherein the central processing unit executes the program stored in the storage unit when running a containerized service. , to realize the computer-aided decision-making method based on multi-dimensional data according to claims 1-3.

Images