CN117649943B - Shaping data intelligent analysis system and method based on machine learning - Google Patents
Shaping data intelligent analysis system and method based on machine learning Download PDFInfo
- Publication number
- CN117649943B CN117649943B CN202410122213.5A CN202410122213A CN117649943B CN 117649943 B CN117649943 B CN 117649943B CN 202410122213 A CN202410122213 A CN 202410122213A CN 117649943 B CN117649943 B CN 117649943B
- Authority
- CN
- China
- Prior art keywords
- semantic
- personal information
- medical record
- plastic surgery
- feature vector
- 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.)
- Active
Links
- 238000004458 analytical method Methods 0.000 title claims abstract description 59
- 238000010801 machine learning Methods 0.000 title claims abstract description 52
- 238000000034 method Methods 0.000 title claims abstract description 17
- 238000007493 shaping process Methods 0.000 title claims description 40
- 239000013598 vector Substances 0.000 claims abstract description 249
- 230000004927 fusion Effects 0.000 claims abstract description 58
- 238000001356 surgical procedure Methods 0.000 claims abstract description 31
- 238000002316 cosmetic surgery Methods 0.000 claims description 243
- 238000005457 optimization Methods 0.000 claims description 28
- 230000011218 segmentation Effects 0.000 claims description 20
- 238000001514 detection method Methods 0.000 claims description 16
- 230000006870 function Effects 0.000 claims description 14
- 238000012545 processing Methods 0.000 claims description 11
- 239000011159 matrix material Substances 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000013507 mapping Methods 0.000 claims description 3
- 101100001674 Emericella variicolor andI gene Proteins 0.000 claims 2
- 238000012502 risk assessment Methods 0.000 description 15
- 230000000694 effects Effects 0.000 description 7
- 230000036541 health Effects 0.000 description 7
- 230000000399 orthopedic effect Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 238000013136 deep learning model Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 3
- 238000011084 recovery Methods 0.000 description 3
- 206010020751 Hypersensitivity Diseases 0.000 description 2
- 208000028389 Nerve injury Diseases 0.000 description 2
- 230000003187 abdominal effect Effects 0.000 description 2
- 208000026935 allergic disease Diseases 0.000 description 2
- 230000000172 allergic effect Effects 0.000 description 2
- 230000007815 allergy Effects 0.000 description 2
- 208000010668 atopic eczema Diseases 0.000 description 2
- 230000000740 bleeding effect Effects 0.000 description 2
- 239000002537 cosmetic Substances 0.000 description 2
- 230000001815 facial effect Effects 0.000 description 2
- 208000015181 infectious disease Diseases 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 230000008764 nerve damage Effects 0.000 description 2
- 230000002980 postoperative effect Effects 0.000 description 2
- 230000008439 repair process Effects 0.000 description 2
- 230000037390 scarring Effects 0.000 description 2
- 230000008961 swelling Effects 0.000 description 2
- 206010067484 Adverse reaction Diseases 0.000 description 1
- 238000012274 Preoperative evaluation Methods 0.000 description 1
- 230000006838 adverse reaction Effects 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000004159 blood analysis Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000005465 channeling Effects 0.000 description 1
- 210000000038 chest Anatomy 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000003193 general anesthetic agent Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 238000013058 risk prediction model Methods 0.000 description 1
- 210000000115 thoracic cavity Anatomy 0.000 description 1
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; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/284—Lexical analysis, e.g. tokenisation or collocates
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- 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
- G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
- G16H10/60—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Public Health (AREA)
- Epidemiology (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Primary Health Care (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Databases & Information Systems (AREA)
- Pathology (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The application discloses a system and a method for intelligently analyzing plastic data based on machine learning, which relate to the technical field of intelligent analysis and are used for acquiring personal information and medical record data of a plastic operation patient, carrying out semantic understanding based on word granularity on the personal information of the patient to obtain personal information word granularity context semantic feature vectors, dividing the medical record data according to plastic operation time to obtain a medical record data set, carrying out semantic encoding on the medical record data through a plastic operation medical record semantic encoder to obtain a feature vector sequence, carrying out constraint fusion on the feature sequence based on the personal information semantic features and the medical record data to obtain a plastic operation medical record global semantic encoding feature vector under the personal information semantic constraint, and finally determining a risk grade label of plastic operation. In this way, the risk of the overall surgery can be automatically classified and predicted, providing a more scientific and efficient reference for the patient and doctor.
Description
Technical Field
The application relates to the technical field of intelligent analysis, in particular to an intelligent analysis system and an intelligent analysis method for shaping data based on machine learning.
Background
Plastic surgery is a procedure for cosmetic or repair purposes by changing the shape or function of the human body. Plastic surgery is a wide variety of including ocular plastic, nasal plastic, facial plastic, thoracic plastic, abdominal plastic, and the like. Although plastic surgery can improve patient confidence and quality of life, there is also a certain risk and complications such as infection, bleeding, swelling, scarring, nerve damage, and the like. Thus, for patients desiring plastic surgery, it is important to know the risk of the surgery in order to make reasonable decisions and precautions.
Currently, risk assessment for plastic surgery relies primarily on the experience and judgment of the physician, lacking objective and quantitative criteria and methods. In addition, since plastic surgery involves a number of factors, such as age, sex, physical condition, allergy history, past medical history, type of surgery, site, mode, doctor, hospital, etc., there are complex interactions and influences between these factors, which makes it difficult to accurately predict and judge the risk of surgery.
Accordingly, an optimized intelligent analysis system and method for shaping data is desired.
Disclosure of Invention
In order to overcome the defects, the application provides an intelligent analysis system and an intelligent analysis method for shaping data based on machine learning.
The application provides a shaping data intelligent analysis system based on machine learning, which comprises:
The patient data acquisition module is used for acquiring personal information and medical record data of the plastic surgery patient;
The personal information semantic coding module is used for carrying out semantic understanding based on word granularity on the personal information of the plastic surgery patient so as to obtain a personal information word granularity context semantic feature vector;
the medical record data time sequence dividing module is used for dividing the medical record data according to the time of the plastic surgery to obtain a set of medical record data of the plastic surgery;
the medical record data semantic coding module is used for enabling each piece of plastic operation medical record data in the plastic operation medical record data set to pass through the plastic operation medical record semantic coder to obtain a sequence of plastic operation medical record data semantic coding feature vectors;
The semantic constraint expression module is used for carrying out constraint fusion on the sequence of the semantic coding feature vectors of the plastic surgery medical record data based on the personal information word granularity context semantic feature vectors so as to obtain the global semantic coding feature vectors of the plastic surgery medical record under the personal information semantic constraint as the global semantic coding features of the plastic surgery medical record under the personal information semantic constraint;
The surgery risk level detection module is used for determining a risk level label of plastic surgery based on global semantic coding features of the plastic surgery medical record under the semantic constraint of the personal information;
Wherein, the semantic constraint expression module is used for:
based on the personal information word granularity context semantic feature vector, performing constraint fusion on the sequence of the plastic surgery medical record data semantic coding feature vector by using the following fusion formula to obtain a plastic surgery medical record global semantic coding feature vector under the personal information semantic constraint;
wherein, the fusion formula is:
；
；
Wherein, Representing the personal information word granularity context semantic feature vector,/>Represents 1×/>Matrix of/>Equal to the dimension of the personal information word granularity context semantic feature vector, B is 1×/>Matrix of/>The number of the semantic coding feature vectors of the plastic surgery medical record data in the sequence equal to the semantic coding feature vectors of the plastic surgery medical record data,/>Is a Sigmoid function,/>Is a weight coefficient,/>And/>Convolution operation representing a1 x 1 convolution kernel,/>Representing each of the sequence of plastic surgery medical record data semantic coding feature vectors, N representing a scale of each of the sequence of plastic surgery medical record data semantic coding feature vectors,/>And expressing the global semantic coding feature vector of the plastic surgery medical record under the semantic constraint of the personal information.
In the shaping data intelligent analysis system based on machine learning, the personal information semantic coding module is used for: and after the personal information of the plastic surgery patient is subjected to word segmentation, the personal information word granularity context semantic feature vector is obtained through a personal information context semantic encoder comprising a word embedding layer.
In the shaping data intelligent analysis system based on machine learning, the personal information semantic coding module comprises: the word segmentation unit is used for carrying out word segmentation processing on the personal information of the plastic surgery patient so as to convert the personal information of the plastic surgery patient into a word sequence composed of a plurality of words; an embedded encoding unit, configured to map each word in the word sequence to a word vector by using the word embedding layer of the personal information context semantic encoder including the word embedding layer to obtain a sequence of word vectors; and a context coding unit, configured to perform global context semantic coding on the sequence of word vectors using the personal information context semantic encoder including the word embedding layer to obtain the personal information word granularity context semantic feature vector.
In the above-mentioned plastic data intelligent analysis system based on machine learning, the surgery risk level detection module includes: the medical record global semantic feature optimization expression unit is used for performing feature distribution optimization on the plastic operation medical record global semantic coding feature vector under the personal information semantic constraint to obtain the plastic operation medical record global semantic coding feature vector under the optimized personal information semantic constraint; the plastic surgery risk detection unit is used for enabling the plastic surgery medical record global semantic coding feature vector under the optimized personal information semantic constraint to pass through the classifier to obtain a classification result, and the classification result is used for representing a risk grade label of plastic surgery.
In the shaping data intelligent analysis system based on machine learning, the medical record global semantic feature optimization expression unit comprises: the feature optimization fusion subunit is used for carrying out optimization fusion on the sequence of the personal information word granularity context semantic feature vector and the plastic surgery medical record data semantic coding feature vector by taking a feature value as a granularity to obtain an optimization fusion feature vector; and the cascade fusion subunit is used for cascading the optimized fusion feature vector with the global semantic coding feature vector of the plastic surgery medical record under the semantic constraint of the personal information to obtain the global semantic coding feature vector of the plastic surgery medical record under the semantic constraint of the optimized personal information.
In the above-mentioned intelligent analysis system for plastic data based on machine learning, the plastic surgery risk detection unit includes: the full-connection coding subunit is used for carrying out full-connection coding on the global semantic coding feature vectors of the plastic surgery medical record under the semantic constraint of the optimized personal information by using a plurality of full-connection layers of the classifier so as to obtain coding classification feature vectors; and the classification subunit is used for passing the coding classification feature vector through a Softmax classification function of the classifier to obtain the classification result.
The application also provides a shaping data intelligent analysis method based on machine learning, which comprises the following steps:
acquiring personal information and medical record data of a plastic surgery patient;
semantic understanding based on word granularity is carried out on personal information of the plastic surgery patient so as to obtain a personal information word granularity context semantic feature vector;
dividing the medical record data according to the time of plastic surgery to obtain a set of medical record data of plastic surgery;
Passing each plastic surgery medical record data in the set of plastic surgery medical record data through a plastic surgery medical record semantic encoder to obtain a sequence of plastic surgery medical record data semantic encoding feature vectors;
Based on the personal information word granularity context semantic feature vector, performing constraint fusion on the sequence of the plastic surgery medical record data semantic coding feature vector to obtain a plastic surgery medical record global semantic coding feature vector under personal information semantic constraint as a plastic surgery medical record global semantic coding feature under personal information semantic constraint;
determining a risk level label of plastic surgery based on global semantic coding features of the plastic surgery medical record under the semantic constraint of the personal information;
Based on the personal information word granularity context semantic feature vector, performing constraint fusion on the sequence of the plastic surgery medical record data semantic coding feature vector to obtain a plastic surgery medical record global semantic coding feature vector under personal information semantic constraint, wherein the plastic surgery medical record global semantic coding feature vector is used as a plastic surgery medical record global semantic coding feature under personal information semantic constraint and is used for:
based on the personal information word granularity context semantic feature vector, performing constraint fusion on the sequence of the plastic surgery medical record data semantic coding feature vector by using the following fusion formula to obtain a plastic surgery medical record global semantic coding feature vector under the personal information semantic constraint;
wherein, the fusion formula is:
；
；
Wherein, Representing the personal information word granularity context semantic feature vector,/>Represents 1×/>Matrix of/>Equal to the dimension of the personal information word granularity context semantic feature vector, B is 1×/>Matrix of/>The number of the semantic coding feature vectors of the plastic surgery medical record data in the sequence equal to the semantic coding feature vectors of the plastic surgery medical record data,/>Is a Sigmoid function,/>Is a weight coefficient,/>And/>Convolution operation representing a1 x 1 convolution kernel,/>Representing each of the sequence of plastic surgery medical record data semantic coding feature vectors, N representing a scale of each of the sequence of plastic surgery medical record data semantic coding feature vectors,/>And expressing the global semantic coding feature vector of the plastic surgery medical record under the semantic constraint of the personal information.
In the above-mentioned intelligent analysis method for shaping data based on machine learning, the semantic understanding based on word granularity is performed on the personal information of the plastic surgery patient to obtain the context semantic feature vector of the personal information word granularity, which comprises the following steps: and after the personal information of the plastic surgery patient is subjected to word segmentation, the personal information word granularity context semantic feature vector is obtained through a personal information context semantic encoder comprising a word embedding layer.
In the above-mentioned intelligent analysis method for shaping data based on machine learning, the step of obtaining the personal information word granularity context semantic feature vector by a personal information context semantic encoder comprising a word embedding layer after performing word segmentation processing on the personal information of the shaping surgery patient comprises the following steps: word segmentation processing is carried out on personal information of the plastic surgery patient so as to convert the personal information of the plastic surgery patient into a word sequence composed of a plurality of words; mapping each word in the word sequence to a word vector by using a word embedding layer of the personal information context semantic encoder comprising the word embedding layer to obtain a sequence of word vectors; and performing global-based context semantic coding on the sequence of word vectors using the personal information context semantic coder comprising a word embedding layer to obtain the personal information word granularity context semantic feature vector.
Compared with the prior art, the intelligent analysis system and the intelligent analysis method for the plastic data based on the machine learning are used for acquiring personal information and medical record data of a plastic operation patient, performing semantic understanding based on word granularity on the personal information of the patient to obtain personal information word granularity context semantic feature vectors, dividing the medical record data according to plastic operation time to obtain a medical record data set, performing semantic encoding on the medical record data through a plastic operation medical record semantic encoder to obtain a feature vector sequence, performing constraint fusion on the feature sequences of the personal information semantic features and the medical record data to obtain a plastic operation medical record global semantic encoding feature vector under the constraint of the personal information semantic, and finally determining a risk grade label of plastic operation. In this way, the risk of the overall surgery can be automatically classified and predicted, providing a more scientific and efficient reference for the patient and doctor.
Drawings
In order to more clearly illustrate the embodiments of the application or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. In the drawings:
fig. 1 is a block diagram of an intelligent analysis system for shaping data based on machine learning according to an embodiment of the present application.
Fig. 2 is a flowchart of a shaping data intelligent analysis method based on machine learning according to an embodiment of the present application.
Fig. 3 is a schematic diagram of a system architecture of a shaping data intelligent analysis method based on machine learning according to an embodiment of the present application.
Fig. 4 is an application scenario diagram of an intelligent analysis system for shaping data based on machine learning according to an embodiment of the present application.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present application more apparent, the embodiments of the present application will be described in further detail with reference to the accompanying drawings. The exemplary embodiments of the present application and their descriptions herein are for the purpose of explaining the present application, but are not to be construed as limiting the application.
Unless defined otherwise, all technical and scientific terms used in the embodiments of the application have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in the present application is for the purpose of describing particular embodiments only and is not intended to limit the scope of the present application.
In describing embodiments of the present application, unless otherwise indicated and limited thereto, the term "connected" should be construed broadly, for example, it may be an electrical connection, or may be a communication between two elements, or may be a direct connection, or may be an indirect connection via an intermediate medium, and it will be understood by those skilled in the art that the specific meaning of the term may be interpreted according to circumstances.
It should be noted that, the term "first\second\third" related to the embodiment of the present application merely distinguishes similar objects, and does not represent a specific order for the objects, and it is understood that "first\second\third" may interchange a specific order or sequence where allowed. It is to be understood that the "first\second\third" distinguishing objects may be interchanged where appropriate such that embodiments of the application described herein may be practiced in sequences other than those illustrated or described herein.
Plastic surgery is a medical procedure for cosmetic or repair purposes by surgically changing the shape or function of the human body, and is a very wide variety of such procedures, including ocular reshaping, nasal reshaping, facial reshaping, chest reshaping, abdominal reshaping, etc., and is generally aimed at improving the appearance, repairing injured tissue, or improving body function. Although plastic surgery can significantly improve patient confidence and quality of life, it is also accompanied by certain risks and complications. These risks include infection, bleeding, swelling, scarring, nerve damage, and the like. In addition, the problems of undesirable appearance effect after operation, adverse reaction to anesthetic drugs and the like can also occur.
It is therefore important for patients considering plastic surgery to fully understand the risk of surgery so that they can assist in making rational decisions and take the necessary precautions to minimize the risk of surgery. Before considering an orthopedic operation, the patient should be fully communicated and consulted with a professional orthopedic surgeon to understand the procedure, possible risks and recovery after the operation. In addition, the patient should choose a well-experienced, well-qualified orthopedic surgeon and medical institution to ensure the safety and expertise of the surgical procedure. After surgery, the patient needs to follow the physician's recommendations strictly for effective post-operative care and recovery to reduce the occurrence of complications and obtain optimal plastic results.
The risk assessment of plastic surgery does depend on the experience and judgment of the physician, but can also be assessed by means of objective and quantitative criteria and methods. When facing the condition of the patient, the doctor can carry out preliminary assessment on the operation risk according to clinical experience and professional knowledge, and the doctor can consider factors such as the whole health condition of the patient, the operation type, the preoperative assessment result and the like, so that the risk is primarily judged. Some fields of orthopedics have begun to use objective assessment tools, such as ASA (american society of anesthesiologists) grading systems, for assessing the overall health of a patient. In addition, there are some risk assessment indicators specific to plastic surgery, such as Body Mass Index (BMI), blood examination results, electrocardiogram, etc., estimated preoperatively. Doctors generally comprehensively consider the personal condition and objective evaluation result of patients, and comprehensively evaluate the surgical risk by combining clinical experience and professional knowledge.
However, risk assessment for plastic surgery is affected by a number of factors, which have complex interactions between them, making accurate prediction and judgment of risk difficult. Patients vary widely in individuals, including age, sex, physical condition, allergic history, past medical history, etc., and these individual differences complicate risk assessment. Although pre-operative evaluation may initially determine the risk of surgery, the post-operative outcome is affected by a number of factors, sometimes difficult to predict completely. Current plastic surgery risk assessment lacks uniform criteria and methods, which challenges the objectivity and consistency of risk assessment.
Therefore, there is a need to formulate criteria and guidelines for plastic surgery risk assessment to improve the objectivity and consistency of the assessment. By means of big data and artificial intelligence technology, a plastic surgery risk prediction model is developed to improve the accuracy of risk assessment. And, by more comprehensive pre-operative assessment, including physical examination, laboratory examination, psychological assessment, etc., the comprehensiveness and accuracy of risk assessment is improved.
The application discloses a shaping data intelligent analysis system based on machine learning, which can predict and evaluate the effect, risk and satisfaction of a shaping operation. The intelligent analysis system for plastic data based on machine learning utilizes a deep learning model to extract relevant characteristics of plastic surgery from multidimensional data sources, including surgery types, parts, modes, doctors, hospitals and the like. The intelligent analysis system for plastic data based on machine learning then predicts the possible results of the plastic surgery, such as appearance after surgery, complications, recovery time, etc., according to the characteristics. Finally, the intelligent analysis system of the plastic data based on the machine learning evaluates the risk and satisfaction degree of the plastic operation according to the prediction result and gives corresponding advice and feedback. The intelligent analysis system for the plastic data based on the machine learning aims to provide a scientific, objective and effective decision support tool for plastic patients and doctors, help the patients to select the most suitable plastic scheme, and improve the safety and effect of plastic surgery.
In one embodiment of the present application, fig. 1 is a block diagram of an intelligent analysis system for shaping data based on machine learning according to an embodiment of the present application. As shown in fig. 1, the intelligent analysis system 100 for shaping data based on machine learning according to an embodiment of the present application includes: a patient data acquisition module 110 for acquiring personal information and medical record data of the plastic surgery patient; the personal information semantic coding module 120 is configured to perform semantic understanding based on word granularity on the personal information of the plastic surgery patient to obtain a context semantic feature vector with the word granularity of the personal information; a medical record data time sequence dividing module 130, configured to divide the medical record data according to the time of the plastic surgery to obtain a set of medical record data of the plastic surgery; a medical record data semantic coding module 140, configured to pass each piece of plastic surgery medical record data in the set of plastic surgery medical record data through a plastic surgery medical record semantic coder to obtain a sequence of plastic surgery medical record data semantic coding feature vectors; the semantic constraint expression module 150 is configured to perform constraint fusion on the sequence of the semantic coding feature vectors of the plastic surgery medical record data based on the context semantic feature vectors of the personal information word granularity, so as to obtain a global semantic coding feature vector of the plastic surgery medical record under the personal information semantic constraint as a global semantic coding feature of the plastic surgery medical record under the personal information semantic constraint; the surgery risk level detection module 160 is configured to determine a risk level tag of the plastic surgery based on global semantic coding features of the medical record of the plastic surgery under the semantic constraint of the personal information.
In the patient data acquisition module 110, personal information and medical record data of the plastic surgery patient are acquired. The accuracy and integrity of data collection including personal basic information, past medical history, allergic history, family history, etc. are ensured. In this way, basic data can be provided for subsequent risk assessment, which is helpful for comprehensively knowing the health condition and preoperative condition of the patient. In the personal information semantic coding module 120, semantic understanding based on word granularity is performed on personal information of the patient with the integer surgery, so as to obtain a personal information word granularity context semantic feature vector. The accuracy and the comprehensiveness of semantic coding are ensured, and context is considered when the personal information is subjected to semantic understanding. By converting the personal information into semantic feature vectors, semantic information is provided for subsequent risk assessment. In the medical record data timing dividing module 130, medical record data is divided according to the time of the plastic surgery, so as to obtain a set of medical record data of the plastic surgery. The accuracy and the integrity of time sequence division are ensured, and the related medical record data of each operation are correctly classified. By time-dividing the medical record data, a set of data over a time sequence is provided for subsequent analysis. In the medical record data semantic coding module 140, the set of plastic surgery medical record data is passed through a plastic surgery medical record semantic coder to obtain a sequence of plastic surgery medical record data semantic coding feature vectors. The semantic coding accuracy and the comprehensiveness of the medical record data are ensured, and the medical terms and the clinical significance are required to be considered when the medical record data are subjected to semantic coding. By converting medical record data into a semantic feature vector sequence, semantic information is provided for subsequent risk assessment. In the semantic constraint expression module 150, constraint fusion is performed on the sequence of semantic coding feature vectors of the plastic surgery medical record data based on the context semantic feature vectors with granularity of personal information words, so as to obtain the global semantic coding feature vectors of the plastic surgery medical record under the semantic constraint of personal information. The rationality and the effectiveness of semantic constraint are ensured, and the weight and the relevance of the information are considered when the personal information and the semantic features of the medical record data are fused. By fusing the semantic features of the personal information and the semantic features of the medical record data, a global semantic coding feature vector is obtained, and the overall condition of a patient is more comprehensively expressed. In the surgical risk level detection module 160, a risk level tag for the plastic surgery is determined based on the global semantic coding features of the medical record of the plastic surgery under the semantic constraint of the personal information. The accuracy and the reliability of the risk level are ensured, and the overall condition and the operation characteristics of the patient need to be considered for the division of the risk level. And through global semantic coding features under semantic constraint, the risk of the overall surgery is more accurately estimated, and auxiliary decision-making information is provided for doctors.
Aiming at the technical problems, in the technical scheme of the application, an intelligent analysis system for shaping data based on machine learning is provided, which can automatically classify and predict risks of the shaping operation by using personal information and medical record data of patients and through a deep learning model, and provides more scientific and effective references for the patients and doctors.
Specifically, in the technical scheme of the application, firstly, personal information and medical record data of a plastic surgery patient are acquired. It should be appreciated that the personal information includes information about the patient's age, sex, physical condition, allergy history, etc., which can help the system understand the patient's overall health and potential risk factors. The medical record data records relevant characteristics of multiple plastic surgeries, including operation types, parts, modes, doctors, hospitals and the like, and the data can provide details about each operation process and operation, so that the risk and possible complications of the operation can be judged. Thus, obtaining personal information and medical record data may provide decision support for a physician. A doctor can formulate a more reasonable and personalized surgical scheme according to personal characteristics of a patient and surgical history and in combination with risk assessment results provided by the system.
Then, considering that semantic features such as personal basic information, overall health condition and potential risk factors of the patient exist in personal information of the plastic surgery patient, in order to capture the semantic features from the personal information so as to facilitate subsequent detection and judgment of the risk of the plastic surgery, in the technical scheme of the application, semantic encoding is needed to be carried out on the personal information of the plastic surgery patient through a personal information context semantic encoder comprising a word embedding layer after word segmentation processing so as to extract context semantic association feature information based on word granularity in the personal information, thereby obtaining context semantic feature vectors of personal information word granularity.
In a specific embodiment of the present application, the personal information semantic coding module is configured to: and after the personal information of the plastic surgery patient is subjected to word segmentation, the personal information word granularity context semantic feature vector is obtained through a personal information context semantic encoder comprising a word embedding layer.
Further, in a specific embodiment of the present application, the personal information semantic coding module includes: the word segmentation unit is used for carrying out word segmentation processing on the personal information of the plastic surgery patient so as to convert the personal information of the plastic surgery patient into a word sequence composed of a plurality of words; an embedded encoding unit, configured to map each word in the word sequence to a word vector by using the word embedding layer of the personal information context semantic encoder including the word embedding layer to obtain a sequence of word vectors; and a context coding unit, configured to perform global context semantic coding on the sequence of word vectors using the personal information context semantic encoder including the word embedding layer to obtain the personal information word granularity context semantic feature vector.
It should be appreciated that there may be medical record data information for a plurality of plastic surgeries in the medical record data, and each plastic surgery is an independent event, where the medical record data includes details of the surgery, such as the type, location, mode, doctor, hospital, etc. Therefore, in order to independently process and analyze medical record data of each surgery and pay attention to the change in the time dimension, in the technical scheme of the application, the medical record data is divided according to the time of the plastic surgery to obtain a set of medical record data of the plastic surgery. The medical record data are divided according to the operation time, the data of each operation can be independently extracted according to the time dimension, so that the semantic code of each operation is focused on the related information of one plastic operation, the system is helped to better understand the context semantic information, the operation process and the operation details of each operation, and meanwhile, the correlation of the plastic operation in the time dimension is focused, so that the accuracy and the reliability of the operation risk detection are improved.
And then, carrying out semantic coding on each plastic operation medical record data in the plastic operation medical record data set through a plastic operation medical record semantic coder so as to extract semantic understanding characteristic information in each plastic operation medical record data respectively, thereby obtaining a sequence of semantic coding characteristic vectors of the plastic operation medical record data.
Further, considering that the personal information contains personal characteristics and health conditions of the plastic surgery patient, the plastic surgery medical record data semantic coding feature vector sequence has an important effect on surgery risk assessment, and the plastic surgery medical record data semantic understanding feature information of the patient based on each surgery medical record data semantic understanding feature information of the time dimension can reflect detailed detail features of each surgery in time sequence, and the plastic surgery medical record data semantic coding feature vector sequence has an important effect on surgery risk judgment. And, it is also considered that the risk of surgery is affected by the health condition of the individual and the history of surgery when performing plastic surgery risk detection and judgment. Therefore, in order to combine the semantic features of the personal information of the patient with the semantic feature information of the plastic surgery medical record data to obtain more comprehensive and accurate feature representation, in the technical scheme of the application, constraint fusion is further carried out on the sequence of the semantic coding feature vectors of the plastic surgery medical record data based on the context semantic feature vectors of the personal information word granularity to obtain the global semantic coding feature vectors of the plastic surgery medical record under the semantic constraint of the personal information. In particular, by means of constraint fusion, semantic features of personal information of a patient can be introduced into semantic feature expression of plastic surgery medical record data, so that analysis of the plastic surgery medical record data can be more individuated and accurate. Therefore, the global semantic coding feature vector of the plastic surgery medical record under the semantic constraint of the personal information can comprehensively consider the personal information of a patient and the semantic information of the plastic surgery medical record data, and has richer and global feature representation capability.
In a specific embodiment of the present application, the semantic constraint expression module is configured to: based on the personal information word granularity context semantic feature vector, performing constraint fusion on the sequence of the plastic surgery medical record data semantic coding feature vector by using the following fusion formula to obtain a plastic surgery medical record global semantic coding feature vector under the personal information semantic constraint; wherein, the fusion formula is:
；
；
Wherein, Representing the personal information word granularity context semantic feature vector,/>Represents 1×/>Matrix of/>Equal to the dimension of the personal information word granularity context semantic feature vector, B is 1×/>Matrix of/>The number of the semantic coding feature vectors of the plastic surgery medical record data in the sequence equal to the semantic coding feature vectors of the plastic surgery medical record data,/>Is a Sigmoid function,/>Is a weight coefficient,/>And/>Convolution operation representing a1 x 1 convolution kernel,/>Representing each of the sequence of plastic surgery medical record data semantic coding feature vectors, N representing a scale of each of the sequence of plastic surgery medical record data semantic coding feature vectors,/>And expressing the global semantic coding feature vector of the plastic surgery medical record under the semantic constraint of the personal information.
In one embodiment of the present application, the surgical risk level detection module includes: the medical record global semantic feature optimization expression unit is used for performing feature distribution optimization on the plastic operation medical record global semantic coding feature vector under the personal information semantic constraint to obtain the plastic operation medical record global semantic coding feature vector under the optimized personal information semantic constraint; the plastic surgery risk detection unit is used for enabling the plastic surgery medical record global semantic coding feature vector under the optimized personal information semantic constraint to pass through the classifier to obtain a classification result, and the classification result is used for representing a risk grade label of plastic surgery.
Further, in a specific embodiment of the present application, the medical record global semantic feature optimization expression unit includes: the feature optimization fusion subunit is used for carrying out optimization fusion on the sequence of the personal information word granularity context semantic feature vector and the plastic surgery medical record data semantic coding feature vector by taking a feature value as a granularity to obtain an optimization fusion feature vector; and the cascade fusion subunit is used for cascading the optimized fusion feature vector with the global semantic coding feature vector of the plastic surgery medical record under the semantic constraint of the personal information to obtain the global semantic coding feature vector of the plastic surgery medical record under the semantic constraint of the optimized personal information.
In the above technical solution, the personal information word granularity context semantic feature vector expresses word granularity context text semantic features of personal information of the plastic surgery patient based on word segmentation semantic embedding, and the sequence of plastic surgery medical record data semantic coding feature vectors expresses coding semantic features of each plastic surgery medical record data in the plastic surgery medical record data set, so that variability in coding semantic feature calculation dimension due to inconsistency of source data semantic distribution exists between the personal information word granularity context semantic feature vector and the sequence of plastic surgery medical record data semantic coding feature vectors.
In this way, in order to promote the fusion effect of the personal information word granularity context semantic feature vector and the sequence of the plastic surgery medical record data semantic coding feature vector under the classification judgment based on the classifier when the sequence of the plastic surgery medical record data semantic coding feature vector is subjected to constraint fusion based on the personal information word granularity context semantic feature vector, the applicant of the present application preferably performs optimization fusion on the personal information word granularity context semantic feature vector and the sequence of the plastic surgery medical record data semantic coding feature vector with feature values as granularity to obtain an optimized fusion feature vector, which is specifically expressed as: optimizing and fusing the sequence of the personal information word granularity context semantic feature vector and the plastic surgery medical record data semantic coding feature vector by taking a feature value as granularity by using the following optimization formula to obtain an optimized and fused feature vector; wherein, the optimization formula is:
；
Wherein, And/>Personal information word granularity context semantic feature vector/>, obtained through linear interpolationAnd the cascade feature vector/>, which is obtained by cascading the sequences of the semantic coding feature vectors of the plastic surgery medical record dataCharacteristic value of/>And/>Representing the square of the one norm of the feature vector and the square root of the two norms of the feature vector, respectively, the personal information word granularity context semantic feature vector/>And cascading eigenvectors/>Having the same eigenvector length/>And/>Is a weight superparameter,/>Is the eigenvalue of the optimized fusion eigenvector,/>Representing the calculation of a value of a natural exponential function raised to a power by a value,/>Representing per-position addition,/>Representing per-position subtraction.
Here, the optimizing fusion performs division based on a foreground manifold and a background manifold of a vector scale on a serialization fusion representation of the context semantic feature vector of the personal information word granularity and the sequence of the semantic encoding feature vector of the plastic operation medical record data based on correspondence under a feature value granularity, so as to stack dynamic feature value channeling association of the context semantic feature vector of the personal information word granularity and the sequence of the semantic encoding feature vector of the plastic operation medical record data under a feature correspondence channel hyper-manifold aggregation mechanism, thereby marking feature semantic information of variation between the context semantic feature vector of the personal information word granularity and the sequence of the semantic encoding feature vector of the plastic operation medical record data, and realizing full-connection-like stacking fusion according to variability of semantic content between the context semantic feature vector of the personal information word granularity and the sequence of the semantic encoding feature vector of the plastic operation medical record data under different calculation dimensions. In this way, the optimized fusion feature vector and the global semantic coding feature vector of the plastic operation medical record under the semantic constraint of the personal information are cascaded, so that the fusion effect of the context semantic feature vector of the personal information word granularity and the sequence of the semantic coding feature vector of the plastic operation medical record data is improved, and the accuracy of the classification result obtained by the global semantic coding feature vector of the plastic operation medical record under the semantic constraint of the personal information through the classifier is improved. In this way, the risk of the overall surgery can be automatically classified and predicted by using personal information and medical record data of the patient and through a deep learning model, and more scientific and effective references are provided for the patient and doctors.
And then, the global semantic coding feature vector of the plastic surgery medical record under the semantic constraint of the optimized personal information passes through a classifier to obtain a classification result, wherein the classification result is used for representing a risk level label of plastic surgery. That is, the global semantic expression characteristic information of the medical record of the plastic surgery under the semantic constraint of the personal information of the patient of the plastic surgery is utilized to carry out classification processing, so that the risk level of the plastic surgery is judged. In particular, in the technical scheme of the application, the label of the classifier is a risk grade label of plastic surgery, so that after the classification result is obtained, the risk of plastic surgery can be automatically classified and predicted based on the classification result, and more scientific and effective reference is provided for patients and doctors.
In a specific embodiment of the present application, the plastic surgery risk detection unit includes: the full-connection coding subunit is used for carrying out full-connection coding on the global semantic coding feature vectors of the plastic surgery medical record under the semantic constraint of the optimized personal information by using a plurality of full-connection layers of the classifier so as to obtain coding classification feature vectors; and the classification subunit is used for passing the coding classification feature vector through a Softmax classification function of the classifier to obtain the classification result.
In summary, the intelligent analysis system 100 for shaping data based on machine learning according to the embodiment of the present application is illustrated, which can automatically classify and predict the risk of the shaping surgery by using personal information and medical record data of the patient and through a deep learning model, and provides more scientific and effective references for the patient and doctor.
As described above, the intelligent analysis system 100 for shaping data based on machine learning according to the embodiment of the present application can be implemented in various terminal devices, such as a server or the like for intelligent analysis of shaping data based on machine learning. In one example, the machine learning based shaping data intelligent analysis system 100 according to embodiments of the present application may be integrated into a terminal device as a software module and/or hardware module. For example, the machine learning based shaping data intelligent analysis system 100 may be a software module in the operating system of the terminal device, or may be an application developed for the terminal device; of course, the intelligent analysis system 100 for shaping data based on machine learning can also be one of a plurality of hardware modules of the terminal device.
Alternatively, in another example, the machine-learning-based shaped data intelligent analysis system 100 and the terminal device may be separate devices, and the machine-learning-based shaped data intelligent analysis system 100 may be connected to the terminal device through a wired and/or wireless network and transmit interactive information in a agreed data format.
Fig. 2 is a flowchart of a shaping data intelligent analysis method based on machine learning according to an embodiment of the present application. Fig. 3 is a schematic diagram of a system architecture of a shaping data intelligent analysis method based on machine learning according to an embodiment of the present application. As shown in fig. 2 and 3, an intelligent analysis method for shaping data based on machine learning includes: 210, acquiring personal information and medical record data of a plastic surgery patient; 220, carrying out semantic understanding based on word granularity on the personal information of the plastic surgery patient to obtain a personal information word granularity context semantic feature vector; 230, dividing the medical record data according to the time of the plastic surgery to obtain a set of medical record data of the plastic surgery; 240, passing each plastic surgery medical record data in the plastic surgery medical record data set through a plastic surgery medical record semantic encoder to obtain a sequence of plastic surgery medical record data semantic encoding feature vectors; 250, based on the personal information word granularity context semantic feature vector, performing constraint fusion on the sequence of the plastic surgery medical record data semantic coding feature vector to obtain a plastic surgery medical record global semantic coding feature vector under personal information semantic constraint as a plastic surgery medical record global semantic coding feature under personal information semantic constraint; 260, determining the risk level label of the plastic surgery based on the global semantic coding features of the plastic surgery medical record under the semantic constraint of the personal information.
In the intelligent analysis method of plastic data based on machine learning, semantic understanding based on word granularity is carried out on personal information of the plastic surgery patient to obtain a context semantic feature vector of the personal information word granularity, and the method comprises the following steps: and after the personal information of the plastic surgery patient is subjected to word segmentation, the personal information word granularity context semantic feature vector is obtained through a personal information context semantic encoder comprising a word embedding layer.
In the intelligent analysis method of plastic data based on machine learning, after the personal information of the plastic surgery patient is subjected to word segmentation, the personal information word granularity context semantic feature vector is obtained through a personal information context semantic encoder comprising a word embedding layer, and the method comprises the following steps: word segmentation processing is carried out on personal information of the plastic surgery patient so as to convert the personal information of the plastic surgery patient into a word sequence composed of a plurality of words; mapping each word in the word sequence to a word vector by using a word embedding layer of the personal information context semantic encoder comprising the word embedding layer to obtain a sequence of word vectors; and performing global-based context semantic coding on the sequence of word vectors using the personal information context semantic coder comprising a word embedding layer to obtain the personal information word granularity context semantic feature vector.
It will be appreciated by those skilled in the art that the specific operation of each step in the above-described intelligent analysis method for shaping data based on machine learning has been described in detail in the above description of the intelligent analysis system for shaping data based on machine learning with reference to fig. 1, and thus, repetitive description thereof will be omitted.
Fig. 4 is an application scenario diagram of an intelligent analysis system for shaping data based on machine learning according to an embodiment of the present application. As shown in fig. 4, in this application scenario, first, personal information (C2 as illustrated in fig. 4) and medical record data (C1 as illustrated in fig. 4) of a plastic surgery patient are acquired; the acquired personal information and medical record data are then input into a server (S as illustrated in fig. 4) deployed with a machine-learning-based intelligent analysis algorithm for orthopedic data, wherein the server is capable of processing the personal information and medical record data based on the machine-learning-based intelligent analysis algorithm for orthopedic surgery to determine a risk level tag for orthopedic surgery.
The foregoing description of the embodiments has been provided for the purpose of illustrating the general principles of the application, and is not meant to limit the scope of the application, but to limit the application to the particular embodiments, and any modifications, equivalents, improvements, etc. that fall within the spirit and principles of the application are intended to be included within the scope of the application.
Claims (7)
1. An intelligent analysis system for shaping data based on machine learning, comprising:
The patient data acquisition module is used for acquiring personal information and medical record data of the plastic surgery patient;
The personal information semantic coding module is used for carrying out semantic understanding based on word granularity on the personal information of the plastic surgery patient so as to obtain a personal information word granularity context semantic feature vector;
the medical record data time sequence dividing module is used for dividing the medical record data according to the time of the plastic surgery to obtain a set of medical record data of the plastic surgery;
the medical record data semantic coding module is used for enabling each piece of plastic operation medical record data in the plastic operation medical record data set to pass through the plastic operation medical record semantic coder to obtain a sequence of plastic operation medical record data semantic coding feature vectors;
The semantic constraint expression module is used for carrying out constraint fusion on the sequence of the semantic coding feature vectors of the plastic surgery medical record data based on the personal information word granularity context semantic feature vectors so as to obtain the global semantic coding feature vectors of the plastic surgery medical record under the personal information semantic constraint as the global semantic coding features of the plastic surgery medical record under the personal information semantic constraint;
The surgery risk level detection module is used for determining a risk level label of plastic surgery based on global semantic coding features of the plastic surgery medical record under the semantic constraint of the personal information;
Wherein, the semantic constraint expression module is used for:
based on the personal information word granularity context semantic feature vector, performing constraint fusion on the sequence of the plastic surgery medical record data semantic coding feature vector by using the following fusion formula to obtain a plastic surgery medical record global semantic coding feature vector under the personal information semantic constraint;
wherein, the fusion formula is:
；
；
Wherein, Representing the personal information word granularity context semantic feature vector,/>Represents 1×/>Matrix of/>Equal to the dimension of the personal information word granularity context semantic feature vector, B is 1×/>Matrix of/>The number of the semantic coding feature vectors of the plastic surgery medical record data in the sequence equal to the semantic coding feature vectors of the plastic surgery medical record data,/>Is a Sigmoid function,/>Is a weight coefficient,/>And/>Convolution operation representing a1 x 1 convolution kernel,/>Representing each of the sequence of plastic surgery medical record data semantic coding feature vectors, N representing a scale of each of the sequence of plastic surgery medical record data semantic coding feature vectors,/>Representing the global semantic coding feature vector of the plastic surgery medical record under the semantic constraint of the personal information;
wherein, the operation risk level detection module includes:
The medical record global semantic feature optimization expression unit is used for performing feature distribution optimization on the plastic operation medical record global semantic coding feature vector under the personal information semantic constraint to obtain the plastic operation medical record global semantic coding feature vector under the optimized personal information semantic constraint;
The plastic surgery risk detection unit is used for enabling the global semantic coding feature vector of the plastic surgery medical record under the semantic constraint of the optimized personal information to pass through the classifier to obtain a classification result, and the classification result is used for representing a risk grade label of plastic surgery;
the medical record global semantic feature optimization expression unit comprises:
The feature optimization fusion subunit is used for carrying out optimization fusion on the sequence of the personal information word granularity context semantic feature vector and the plastic surgery medical record data semantic coding feature vector by taking a feature value as a granularity to obtain an optimization fusion feature vector;
The cascade fusion subunit is used for cascading the optimized fusion feature vector with the global semantic coding feature vector of the plastic surgery medical record under the semantic constraint of the personal information to obtain the global semantic coding feature vector of the plastic surgery medical record under the semantic constraint of the optimized personal information;
Wherein the feature optimization fusion subunit comprises: optimizing and fusing the sequence of the personal information word granularity context semantic feature vector and the plastic surgery medical record data semantic coding feature vector by taking a feature value as granularity by using the following optimization formula to obtain an optimized and fused feature vector; wherein, the optimization formula is:
；
Wherein, And/>Personal information word granularity context semantic feature vector/>, obtained through linear interpolationAnd the cascade feature vector/>, which is obtained by cascading the sequences of the semantic coding feature vectors of the plastic surgery medical record dataCharacteristic value of/>And/>Representing the square of the one norm of the feature vector and the square root of the two norms of the feature vector, respectively, the personal information word granularity context semantic feature vector/>And cascading eigenvectors/>Having the same eigenvector length/>And (2) andIs a weight superparameter,/>Is the eigenvalue of the optimized fusion eigenvector,/>Representing the calculation of a value of a natural exponential function raised to a power by a value,/>Representing per-position addition,/>Representing per-position subtraction.
2. The intelligent analysis system for shaping data based on machine learning of claim 1, wherein the personal information semantic coding module is configured to: and after the personal information of the plastic surgery patient is subjected to word segmentation, the personal information word granularity context semantic feature vector is obtained through a personal information context semantic encoder comprising a word embedding layer.
3. The intelligent analysis system for shaping data based on machine learning of claim 2, wherein the personal information semantic coding module comprises:
the word segmentation unit is used for carrying out word segmentation processing on the personal information of the plastic surgery patient so as to convert the personal information of the plastic surgery patient into a word sequence composed of a plurality of words;
an embedded encoding unit, configured to map each word in the word sequence to a word vector by using the word embedding layer of the personal information context semantic encoder including the word embedding layer to obtain a sequence of word vectors;
And the context coding unit is used for carrying out global-based context semantic coding on the sequence of the word vectors by using the personal information context semantic coder containing the word embedding layer so as to obtain the personal information word granularity context semantic feature vector.
4. The intelligent analysis system for plastic data based on machine learning according to claim 3, wherein the plastic surgery risk detection unit comprises:
The full-connection coding subunit is used for carrying out full-connection coding on the global semantic coding feature vectors of the plastic surgery medical record under the semantic constraint of the optimized personal information by using a plurality of full-connection layers of the classifier so as to obtain coding classification feature vectors;
and the classification subunit is used for passing the coding classification feature vector through a Softmax classification function of the classifier to obtain the classification result.
5. An intelligent analysis method for shaping data based on machine learning is characterized by comprising the following steps:
acquiring personal information and medical record data of a plastic surgery patient;
semantic understanding based on word granularity is carried out on personal information of the plastic surgery patient so as to obtain a personal information word granularity context semantic feature vector;
dividing the medical record data according to the time of plastic surgery to obtain a set of medical record data of plastic surgery;
Passing each plastic surgery medical record data in the set of plastic surgery medical record data through a plastic surgery medical record semantic encoder to obtain a sequence of plastic surgery medical record data semantic encoding feature vectors;
Based on the personal information word granularity context semantic feature vector, performing constraint fusion on the sequence of the plastic surgery medical record data semantic coding feature vector to obtain a plastic surgery medical record global semantic coding feature vector under personal information semantic constraint as a plastic surgery medical record global semantic coding feature under personal information semantic constraint;
determining a risk level label of plastic surgery based on global semantic coding features of the plastic surgery medical record under the semantic constraint of the personal information;
Based on the personal information word granularity context semantic feature vector, performing constraint fusion on the sequence of the plastic surgery medical record data semantic coding feature vector to obtain a plastic surgery medical record global semantic coding feature vector under personal information semantic constraint, wherein the plastic surgery medical record global semantic coding feature vector is used as a plastic surgery medical record global semantic coding feature under personal information semantic constraint and is used for:
based on the personal information word granularity context semantic feature vector, performing constraint fusion on the sequence of the plastic surgery medical record data semantic coding feature vector by using the following fusion formula to obtain a plastic surgery medical record global semantic coding feature vector under the personal information semantic constraint;
wherein, the fusion formula is:
；
；
Wherein, Representing the personal information word granularity context semantic feature vector,/>Represents 1×/>Matrix of/>Equal to the dimension of the personal information word granularity context semantic feature vector, B is 1×/>Matrix of/>The number of the semantic coding feature vectors of the plastic surgery medical record data in the sequence equal to the semantic coding feature vectors of the plastic surgery medical record data,/>Is a Sigmoid function,/>Is a weight coefficient,/>And/>Convolution operation representing a1 x 1 convolution kernel,/>Representing each of the sequence of plastic surgery medical record data semantic coding feature vectors, N representing a scale of each of the sequence of plastic surgery medical record data semantic coding feature vectors,/>Representing the global semantic coding feature vector of the plastic surgery medical record under the semantic constraint of the personal information;
Wherein, based on the global semantic coding feature of the plastic surgery medical record under the semantic constraint of the personal information, determining the risk level label of the plastic surgery comprises the following steps:
performing feature distribution optimization on the global semantic coding feature vector of the plastic surgery medical record under the personal information semantic constraint to obtain the global semantic coding feature vector of the plastic surgery medical record under the optimized personal information semantic constraint;
the global semantic coding feature vector of the plastic surgery medical record under the semantic constraint of the optimized personal information is passed through a classifier to obtain a classification result, and the classification result is used for representing a risk grade label of plastic surgery;
The feature distribution optimization is performed on the global semantic coding feature vector of the plastic surgery medical record under the personal information semantic constraint to obtain the global semantic coding feature vector of the plastic surgery medical record under the optimized personal information semantic constraint, and the method comprises the following steps:
Optimizing and fusing the personal information word granularity context semantic feature vector and the sequence of the plastic surgery medical record data semantic coding feature vector by taking a feature value as granularity to obtain an optimized fusion feature vector;
Cascading the optimized fusion feature vector with the global semantic coding feature vector of the plastic surgery medical record under the semantic constraint of the personal information to obtain the global semantic coding feature vector of the plastic surgery medical record under the semantic constraint of the optimized personal information;
the optimizing and fusing the personal information word granularity context semantic feature vector and the sequence of the plastic surgery medical record data semantic coding feature vector with feature values as granularity to obtain an optimizing and fusing feature vector comprises the following steps: optimizing and fusing the sequence of the personal information word granularity context semantic feature vector and the plastic surgery medical record data semantic coding feature vector by taking a feature value as granularity by using the following optimization formula to obtain an optimized and fused feature vector; wherein, the optimization formula is:
；
Wherein, And/>Personal information word granularity context semantic feature vector/>, obtained through linear interpolationAnd the cascade feature vector/>, which is obtained by cascading the sequences of the semantic coding feature vectors of the plastic surgery medical record dataCharacteristic value of/>And/>Representing the square of the one norm of the feature vector and the square root of the two norms of the feature vector, respectively, the personal information word granularity context semantic feature vector/>And cascading eigenvectors/>Having the same eigenvector length/>And (2) andIs a weight superparameter,/>Is the eigenvalue of the optimized fusion eigenvector,/>Representing the calculation of a value of a natural exponential function raised to a power by a value,/>Representing per-position addition,/>Representing per-position subtraction.
6. The intelligent analysis method for plastic data based on machine learning according to claim 5, wherein the semantic understanding of personal information of the plastic surgery patient based on word granularity to obtain a personal information word granularity context semantic feature vector comprises: and after the personal information of the plastic surgery patient is subjected to word segmentation, the personal information word granularity context semantic feature vector is obtained through a personal information context semantic encoder comprising a word embedding layer.
7. The intelligent analysis method for plastic data based on machine learning according to claim 6, wherein the step of obtaining the personal information word granularity context semantic feature vector through a personal information context semantic encoder comprising a word embedding layer after performing word segmentation processing on the personal information of the plastic surgery patient comprises the steps of:
Word segmentation processing is carried out on personal information of the plastic surgery patient so as to convert the personal information of the plastic surgery patient into a word sequence composed of a plurality of words;
mapping each word in the word sequence to a word vector by using a word embedding layer of the personal information context semantic encoder comprising the word embedding layer to obtain a sequence of word vectors;
And performing global-based context semantic coding on the sequence of word vectors by using the personal information context semantic coder containing the word embedding layer to obtain the personal information word granularity context semantic feature vector.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410122213.5A CN117649943B (en) | 2024-01-30 | 2024-01-30 | Shaping data intelligent analysis system and method based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410122213.5A CN117649943B (en) | 2024-01-30 | 2024-01-30 | Shaping data intelligent analysis system and method based on machine learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117649943A CN117649943A (en) | 2024-03-05 |
| CN117649943B true CN117649943B (en) | 2024-04-30 |
Family
ID=90043739
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410122213.5A Active CN117649943B (en) | 2024-01-30 | 2024-01-30 | Shaping data intelligent analysis system and method based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117649943B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118098482B (en) * | 2024-04-22 | 2024-06-28 | 吉林大学 | Intelligent medical management system and method based on 5G technology |
| CN118267845B (en) * | 2024-06-03 | 2024-09-10 | 克拉玛依市富城油气销售有限公司 | Automatic monitoring system and method for concentration change of deacidification process amine liquid |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2932052A1 (en) * | 2013-11-29 | 2015-06-04 | Plexina Inc. | System for converting native patient data from disparate systems into unified semantic patient record repository supporting clinical analytics |
| CN106815601A (en) * | 2017-01-10 | 2017-06-09 | 西安电子科技大学 | Hyperspectral image classification method based on recurrent neural network |
| CN113591757A (en) * | 2021-08-07 | 2021-11-02 | 王新 | Automatic operation device and equipment for eye reshaping |
| WO2022266395A1 (en) * | 2021-06-17 | 2022-12-22 | Northwestern University | Adaptive discovery and mixed-variable optimization of next generation synthesizable microelectronic materials |
| CN116091414A (en) * | 2022-12-14 | 2023-05-09 | 皖西学院 | Cardiovascular image recognition method and system based on deep learning |
| CN116415581A (en) * | 2023-04-21 | 2023-07-11 | 浙江晓扬科技有限公司 | Teaching data analysis system based on intelligent education |
| CN116629275A (en) * | 2023-07-21 | 2023-08-22 | 北京无极慧通科技有限公司 | Intelligent decision support system and method based on big data |
| CN117012364A (en) * | 2023-10-08 | 2023-11-07 | 吉林大学 | Medical health service cloud platform based on industrial Internet technology |
| CN117052970A (en) * | 2023-09-07 | 2023-11-14 | 滨特尔集团有限公司 | Intelligent control system and method for pneumatic ball valve assembly |
| CN117235630A (en) * | 2023-11-15 | 2023-12-15 | 吉林大学 | Intelligent disease area visual management system and method thereof |
| CN117251699A (en) * | 2023-11-17 | 2023-12-19 | 北京无极慧通科技有限公司 | Medical big data analysis method and system based on artificial intelligence |
| CN117271777A (en) * | 2023-11-03 | 2023-12-22 | 江西汉氏医学发展有限公司 | Medical beauty data classification processing method and system |
| CN117457229A (en) * | 2023-12-26 | 2024-01-26 | 吉林大学 | Anesthesia depth monitoring system and method based on artificial intelligence |
-
2024
- 2024-01-30 CN CN202410122213.5A patent/CN117649943B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2932052A1 (en) * | 2013-11-29 | 2015-06-04 | Plexina Inc. | System for converting native patient data from disparate systems into unified semantic patient record repository supporting clinical analytics |
| CN106815601A (en) * | 2017-01-10 | 2017-06-09 | 西安电子科技大学 | Hyperspectral image classification method based on recurrent neural network |
| WO2022266395A1 (en) * | 2021-06-17 | 2022-12-22 | Northwestern University | Adaptive discovery and mixed-variable optimization of next generation synthesizable microelectronic materials |
| CN113591757A (en) * | 2021-08-07 | 2021-11-02 | 王新 | Automatic operation device and equipment for eye reshaping |
| CN116091414A (en) * | 2022-12-14 | 2023-05-09 | 皖西学院 | Cardiovascular image recognition method and system based on deep learning |
| CN116415581A (en) * | 2023-04-21 | 2023-07-11 | 浙江晓扬科技有限公司 | Teaching data analysis system based on intelligent education |
| CN116629275A (en) * | 2023-07-21 | 2023-08-22 | 北京无极慧通科技有限公司 | Intelligent decision support system and method based on big data |
| CN117052970A (en) * | 2023-09-07 | 2023-11-14 | 滨特尔集团有限公司 | Intelligent control system and method for pneumatic ball valve assembly |
| CN117012364A (en) * | 2023-10-08 | 2023-11-07 | 吉林大学 | Medical health service cloud platform based on industrial Internet technology |
| CN117271777A (en) * | 2023-11-03 | 2023-12-22 | 江西汉氏医学发展有限公司 | Medical beauty data classification processing method and system |
| CN117235630A (en) * | 2023-11-15 | 2023-12-15 | 吉林大学 | Intelligent disease area visual management system and method thereof |
| CN117251699A (en) * | 2023-11-17 | 2023-12-19 | 北京无极慧通科技有限公司 | Medical big data analysis method and system based on artificial intelligence |
| CN117457229A (en) * | 2023-12-26 | 2024-01-26 | 吉林大学 | Anesthesia depth monitoring system and method based on artificial intelligence |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117649943A (en) | 2024-03-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117649943B (en) | Shaping data intelligent analysis system and method based on machine learning | |
| CN107193919A (en) | The search method and system of a kind of electronic health record | |
| CN109805898B (en) | Critical death prediction method based on attention mechanism time sequence convolution network algorithm | |
| Frize et al. | Clinical decision support systems for intensive care units: using artificial neural networks | |
| US11031121B2 (en) | System and method for indicating the quality of information to support decision making | |
| CN103690240A (en) | Medical system | |
| US10847261B1 (en) | Methods and systems for prioritizing comprehensive diagnoses | |
| CN114283947B (en) | Health management method and system suitable for surgical patients | |
| CN113838577B (en) | Convenient layered old people MODS early death risk assessment model, device and establishment method | |
| CN113077893B (en) | Intelligent auxiliary tool adaptation decision-making system and method | |
| CN107292103A (en) | A kind of prognostic chart picture generation method and device | |
| Geetha et al. | Evaluation based approaches for liver disease prediction using machine learning algorithms | |
| Prabha et al. | A big wave of deep learning in medical imaging-analysis of theory and applications | |
| CN116737945B (en) | Mapping method for EMR knowledge map of patient | |
| CN117457162A (en) | Emergency call sub-diagnosis method and system based on multi-encoder and multi-mode information fusion | |
| Wagh et al. | Early detection of PCOD using machine learning techniques | |
| Karpagavalli et al. | Machine learning approach for preoperative anaesthetic risk prediction | |
| Wang et al. | Deep learning for automating the organization of institutional dermatology image stores | |
| CN111968747A (en) | VTE intelligent prevention and control management system | |
| CN116741350B (en) | File management system for hospital X-ray images | |
| CN118297360B (en) | Intelligent matching method for chest pain center construction | |
| CN113782132B (en) | Dynamic updating method, device and medium of label information | |
| Saleh | Artificial intelligence and statistical techniques to predict probability of injury survival | |
| Misbah et al. | Automatic distal radius fracture detection and classification using deep convolutional neural network with radiological images | |
| Misbah et al. | Original Research Article Automatic distal radius fracture detection and classification using deep convolutional neural network with radiological images |
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 |