CN109859841A - Diagnosis for liver cancer system, method, equipment and medium neural network based - Google Patents
Diagnosis for liver cancer system, method, equipment and medium neural network based Download PDFInfo
- Publication number
- CN109859841A CN109859841A CN201910156222.5A CN201910156222A CN109859841A CN 109859841 A CN109859841 A CN 109859841A CN 201910156222 A CN201910156222 A CN 201910156222A CN 109859841 A CN109859841 A CN 109859841A
- Authority
- CN
- China
- Prior art keywords
- neural network
- diagnosis
- network based
- liver cancer
- knowledge
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Medical Treatment And Welfare Office Work (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
A kind of diagnosis for liver cancer system, method, equipment and medium neural network based provided by the present application, by storing medical knowledge, and the corresponding explicit rule of the medical knowledge is converted to the implicit rule comprising connection weight by learning algorithm neural network based；Corresponding therapeutic scheme is provided for the different state of an illness according to the implicit rule.The application can simply and efficiently improve diagnosing cancer of liver, and guarantee the reliability of diagnosis, and efficiently solve doctor because the diagnosis and treatment of liver cancer specification and cross discipline mastery of knowledge be not enough in time, and bring limits during diagnosing cancer of liver, system self-learning capability is substantially increased, and continuous self-perfection in the process of running.
Description
Technical field
This application involves nerual network technique fields.More particularly, to a kind of diagnosis for liver cancer system neural network based
System, method, equipment and medium.
Background technique
Currently, the diagnostic and therapeutic method for being medically related to primary carcinoma of liver is varied, their applicable elements also compared with
For complexity, it is more difficult for the condition assessment of liver cancer patient that this has resulted in clinician, used in different medical mechanism
Treatment method is also different.In clinical workflow, on the one hand, how the liver cancer patient complicated and changeable in face of the state of an illness dashes forward
The limitation of broken knowledge level and clinical experience, selects most suitable diagnose and treat scheme for patient, is clinician one
A problem urgently to be resolved.During liver cancer diagnosis and treatment, for the diagnosis and treatment of liver cancer specification and cross discipline mastery of knowledge, often shadow
Ring the combined therapy effect of Patients with Primary；On the other hand, the operating pressure of clinician is big, and seldom having time goes to learn
The medical outcome for understanding and updating is practised, it is not fine for leading to universal effect of the newest medical research in clinic.
In the sorts of systems occurred at present, it there is no and be related to the mycin system of primary carcinoma of liver intelligent decision.Both
It is merely resting on and designs and develops or experimental stage toward many systems, can not really solve the practical problem of clinician, into
One step application is restricted.Expert system is always a branch the most active in artificial intelligence, but due to following problems
Effective solution cannot be obtained, its further development is constrained:
1) " bottleneck " of knowledge acquisition is with inscribing, and knowledge acquisition is a bottleneck problem during expert system is built, this is not only
The speed of expert system development is influenced, and directly influences the quality of knowledge and the function of expert system and performance.
2) " narrow step " problem of knowledge.Expert system can be only applied in the ken being rather narrow at present, solve pre-
Fixed special problem, once the problem of encountering beyond knowledge, can not just solve.
3) complexity and efficiency of expert system.Widely applied knowledge representation has production in expert system at present
Raw formula rule, predicate logic, semantic network, frame and object-oriented method etc., although their structure is different with organization and administration,
But it requires a knowledge working process, is converted into computer storable form deposit knowledge base, according still further to certain when reasoning
Matching algorithm and search strategy find required knowledge into knowledge base.On the one hand this expression and processing mode are needed to knowledge
Carry out reasonable Organization And Management, the increase of another aspect knowledge base scale, the complexity for solving with topic improve and when reasonings
The appearance of " conflict " phenomenon, leads to multiple shot array, asks that interior cannot solve is inscribed together when limited, seriously affects the effect of expert system
Rate.
4) do not have the expert system that function of associate memory is developed at present and do not have self-learning capability and associative memory generally
Function, cannot self-perfection in the process of running, cannot by association and memory etc. modes make inferences even in Given information
With noise, be distorted when, lack effective treatment measures.
Summary of the invention
In view of the foregoing deficiencies of prior art, the application is designed to provide a kind of liver cancer neural network based
Assistant diagnosis system, method, equipment and medium, for solving the problems of the prior art.
In order to achieve the above objects and other related objects, the application provides a kind of diagnosis for liver cancer neural network based
System, the system comprises: expert database, for storing medical knowledge, and learning algorithm neural network based will be described
The corresponding explicit rule of medical knowledge is converted to the implicit rule comprising connection weight；Knowledge base, for storing and according to described
Implicit rule provides corresponding therapeutic scheme for the different state of an illness.
In the embodiment of the application, the medical knowledge includes: expertise, case, clinography, medical consultations
Specification, medical consultations guide are appointed in basis/general medical knowledge, medical journals, medical thesis, Physicians report and therapeutic scheme
It anticipates one or more combinations.
In the embodiment of the application, the learning algorithm neural network based is by the explicit rule of the medical knowledge
The function of being then converted to implicit rule specifically includes: combing out one or more groups of diseases according to the corresponding explicit rule of the medical knowledge
Preliminary incidence relation between the corresponding different symptoms of example, disease and therapeutic scheme；It is defeated by input and expectation by neural network
Training out obtains the connection weight of incidence relation between corresponding different symptoms information, disease information and therapeutic scheme, and according to
To form implicit rule.
It is described to be specifically included by neural network by input and the training of desired output in the embodiment of the application:
It is trained based on one or more groups of diseases by error backpropagation algorithm.
In the embodiment of the application, the knowledge base includes: that subsystem is instructed in detection, for collecting patient assessment's meaning
To and essential information, for determine inspection scheme；Condition assessment subsystem, for the inspection according to the corresponding inspection scheme
As a result to determine illness information, and corresponding disease information is evaluated according to the implicit rule；Scheme Choice subsystem, is used for
There is provided the disease information corresponding therapeutic scheme according to the implicit rule.
In the embodiment of the application, the inspection scheme includes: blood urine just routine, biochemical full item, coagulation function, first
Fetoprotein is quantitative, tumor markers, hepatitis B series, HBV-DNA are quantitative, c-hepatitis antibody, HCV-mRNA are quantitative, colored super
Sound, ultrasonic contrast, Enhanced CT, enhancing nuclear magnetic resonance, chest X-ray, PET-CT, ECT bone scanning, hepatic arteriography
(DSA), electrocardiogram, echocardiogram, pulmonary function test, biopsy biopsy, molecule class detection, indocyanine-green excretion experiment,
Any one or more in Future liver volume and other disease examinations combines.
In the embodiment of the application, the condition assessment subsystem includes: course of disease resolver and staging thrust
Device.
In the embodiment of the application, process includes: liver function Child points by stages in the staging thruster
Grade, PS scoring, neoplastic state, whether there is or not vascular invasion, whether there is or not DISTANT METASTASES IN, tumor number, tumor size, bilirubin level, have
Any one or more in no portal hypertension combines.
In the embodiment of the application, the therapeutic scheme includes: operation excision, TACE, radiotherapy, supports to suit the medicine to the illness and control
Any one or more in treatment, local ablation, liver transfer operation, molecular targeted therapy and system chemotherapy combines；The therapeutic scheme is made
It is stored in the expert database for the medical knowledge.
In the embodiment of the application, the system also includes: inference machine, for being directed to the approximate state of an illness for the knowledge
The therapeutic scheme comparative analysis in therapeutic scheme and example case that library provides, to obtain analysis result.
In the embodiment of the application, the analysis result is for updating in the knowledge base or the patent database
The connection weight of the corresponding implicit rule；And/or the analysis result is for explaining the therapeutic scheme.
In order to achieve the above objects and other related objects, the application provides a kind of diagnosis for liver cancer neural network based
Method, which comprises storage medical knowledge, and learning algorithm neural network based is corresponding aobvious by the medical knowledge
Formula rule is converted to the implicit rule comprising connection weight；Corresponding treatment is provided for the different state of an illness according to the implicit rule
Scheme.
In the embodiment of the application, the medical knowledge includes: expertise, case, clinography, medical consultations
Specification, medical consultations guide, any one in basis/general medical knowledge, medical journals, medical thesis and Physicians report or
Multiple combinations.
In the embodiment of the application, the learning algorithm neural network based is by the explicit rule of the medical knowledge
The method for being then converted to implicit rule includes: to comb out one or more groups of cases pair according to the corresponding explicit rule of the medical knowledge
The preliminary incidence relation between different symptoms, disease and therapeutic scheme answered；Input and desired output are passed through by neural network
Training obtains the connection weight of incidence relation between corresponding different symptoms information, disease information and therapeutic scheme, and shape accordingly
At implicit rule.
It is described specifically to be wrapped by neural network by input and the training method of desired output in the embodiment of the application
It includes: being trained based on one or more groups of diseases by error backpropagation algorithm.
It is described to provide corresponding treatment side for the different state of an illness according to the implicit rule in the embodiment of the application
The method of case includes: to collect patient assessment's intention and essential information, for determining inspection scheme；According to the corresponding reviewing party
The inspection result of case evaluates corresponding disease information according to the implicit rule to determine illness information；According to described hidden
Formula rule provides the disease information corresponding therapeutic scheme.
In the embodiment of the application, the inspection scheme includes: blood urine just routine, biochemical full item, coagulation function, first
Fetoprotein is quantitative, tumor markers, hepatitis B series, HBV-DNA are quantitative, c-hepatitis antibody, HCV-mRNA are quantitative, colored super
Sound, ultrasonic contrast, Enhanced CT, enhancing nuclear magnetic resonance, chest X-ray, PET-CT, ECT bone scanning, hepatic arteriography
(DSA), electrocardiogram, echocardiogram, pulmonary function test, biopsy biopsy, molecule class detection, indocyanine-green excretion experiment,
Any one or more in Future liver volume and other disease examinations combines.
In the embodiment of the application, the inspection result according to the corresponding inspection scheme is to determine that illness is believed
Breath, and include: by course of disease resolver and staging according to the method that the implicit rule evaluates corresponding disease information
Thruster is completed.
In the embodiment of the application, process includes: liver function Child points by stages in the staging thruster
Grade, PS scoring, neoplastic state, whether there is or not vascular invasion, whether there is or not DISTANT METASTASES IN, tumor number, tumor size, bilirubin level, have
Any one or more in no portal hypertension combines.
In the embodiment of the application, the therapeutic scheme includes: operation excision, TACE, radiotherapy, supports to suit the medicine to the illness and control
Any one or more in treatment, local ablation, liver transfer operation, molecular targeted therapy and system chemotherapy combines；The therapeutic scheme is made
It is the medical knowledge to be stored.
In the embodiment of the application, the method also includes: it is controlled for the approximate state of an illness by what the knowledge base provided
Therapeutic scheme comparative analysis in treatment scheme and example sample, to obtain analysis result.
In the embodiment of the application, the analysis result is for updating in the knowledge base or the patent database
The connection weight of the corresponding implicit rule；And/or the analysis result is for explaining the therapeutic scheme.
In order to achieve the above objects and other related objects, the application provides a kind of diagnosis for liver cancer neural network based
Equipment, the equipment include: memory and processor；The memory is for storing computer program；The processor is used for
The computer program of the memory storage is executed, so as to realize when the equipment executes as described above neural network based
Diagnosis for liver cancer method.
In order to achieve the above objects and other related objects, the application provides a kind of computer readable storage medium, deposits thereon
Computer program is contained, realizes that liver cancer neural network based as described above is auxiliary when the computer program is executed by processor
Auxiliary diagnosis method.
As described above, a kind of diagnosis for liver cancer system, method, equipment and the medium neural network based of the application,
By storing medical knowledge, and the corresponding explicit rule of the medical knowledge is converted to packet by learning algorithm neural network based
Implicit rule containing connection weight；Corresponding therapeutic scheme is provided for the different state of an illness according to the implicit rule.
It has the advantages that
Diagnosing cancer of liver can simply and efficiently be improved, and guarantee diagnosis reliability, and efficiently solve doctor because
The diagnosis and treatment of liver cancer specification and cross discipline mastery of knowledge be not enough in time, and during diagnosing cancer of liver bring limit,
System self-learning capability is substantially increased, and continuous self-perfection in the process of running.
Detailed description of the invention
Fig. 1 is shown as the module signal of neural network based diagnosis for liver cancer system of the application in an embodiment
Figure.
Fig. 2 is shown as the flow diagram that the inspection and guide of subsystem are instructed in detection of the application in an embodiment.
Fig. 3 is shown as the application process signal of diagnosing cancer of liver by stages in the condition assessment subsystem in an embodiment
Figure.
Fig. 4 is shown as the flow diagram of the application dicision of diagnosis and treatment in the Scheme Choice subsystem in an embodiment.
Fig. 5 is shown as the flow diagram of the application neural network in the knowledge base in an embodiment.
Fig. 6 is shown as the process signal of neural network based diagnosis for liver cancer method of the application in an embodiment
Figure.
Fig. 7 is shown as the structural representation of neural network based diagnosis for liver cancer equipment of the application in an embodiment
Figure.
Specific embodiment
Illustrate presently filed embodiment below by way of specific specific example, those skilled in the art can be by this specification
Disclosed content understands other advantages and effect of the application easily.The application can also pass through in addition different specific realities
The mode of applying is embodied or practiced, the various details in this specification can also based on different viewpoints and application, without departing from
Various modifications or alterations are carried out under spirit herein.It should be noted that in the absence of conflict, following embodiment and implementation
Feature in example can be combined with each other.
It should be noted that with reference to attached drawing, attached drawing describes several embodiments of the application in described below.It should
Understand, other embodiments also can be used, and mechanical group can be carried out without departing substantially from spirit and scope
At, structure, electrical and operational change.Following detailed description should not be considered limiting, and the application
The range of embodiment only limited by the claims for the patent announced.Term used herein is merely to description is specific
Embodiment, and it is not intended to limit the application.The term of space correlation, for example, "upper", "lower", "left", "right", " following ", " under
Side ", " lower part ", " top ", " top " etc. can be used in the text in order to elements or features shown in explanatory diagram and another
The relationship of one elements or features.
In this application unless specifically defined or limited otherwise, term " installation ", " connected ", " connection ", " fixation ",
Terms such as " fixings " shall be understood in a broad sense, for example, it may be being fixedly connected, may be a detachable connection, or be integrally connected；
It can be mechanical connection, be also possible to be electrically connected；It can be directly connected, can also indirectly connected through an intermediary, it can be with
It is the connection inside two elements.For the ordinary skill in the art, above-mentioned art can be understood as the case may be
The concrete meaning of language in this application.
Furthermore as used in herein, singular " one ", "one" and "the" are intended to also include plural number shape
Formula, unless there is opposite instruction in context.It will be further understood that term "comprising", " comprising " show that there are the spies
Sign, operation, element, component, project, type, and/or group, but it is not excluded for one or more other features, operation, element, group
Presence, appearance or the addition of part, project, type, and/or group.Term "or" and "and/or" used herein are interpreted as including
Property, or mean any one or any combination.Therefore, " A, B or C " or " A, B and/or C " mean " it is following any one:
A；B；C；A and B；A and C；B and C；A, B and C ".Only when the combination of element, functions or operations is inherently mutual under certain modes
When repulsion, it just will appear the exception of this definition.
The diagnostic and therapeutic method for being medically related to primary carcinoma of liver is varied, their applicable elements are also more multiple
Miscellaneous, this has been resulted in, and clinician is more difficult for the condition assessment of liver cancer patient, and different medical mechanism is used treats
Method is also different.In clinical workflow, on the one hand, how the liver cancer patient complicated and changeable in face of the state of an illness is broken through certainly
The limitation of body know-how and clinical experience selects most suitable diagnose and treat scheme for patient, be clinician one urgently
Problem to be solved.During liver cancer diagnosis and treatment, for the diagnosis and treatment of liver cancer specification and cross discipline mastery of knowledge, often affect
The combined therapy effect of Patients with Primary；On the other hand, the operating pressure of clinician is big, and seldom having time goes study
The medical outcome updated is solved, it is not fine for leading to universal effect of the newest medical research in clinic.
In view of above-mentioned various technical problems, diagnosis for liver cancer system neural network based provided by the present application, method,
Equipment and medium, effectively to solve these problems.The main thought of the application is intended to provide learning algorithm neural network based
The corresponding explicit rule of the medical knowledge is converted into the implicit rule comprising connection weight, and according to the implicit rule needle
Corresponding therapeutic scheme is provided to the different state of an illness.
As shown in Figure 1, being shown as neural network based diagnosis for liver cancer system of the application in an embodiment
Module diagram.As shown, the system 100 includes: expert database 110, knowledge base 120.
Expert database 110
The expert database 110, for storing medical knowledge, and learning algorithm neural network based is by the medicine
The corresponding explicit rule of knowledge is converted to the implicit rule comprising connection weight.
In the embodiment of the application, the medical knowledge includes: expertise, case, clinography, medical consultations
Specification, medical consultations guide are appointed in basis/general medical knowledge, medical journals, medical thesis, Physicians report and therapeutic scheme
It anticipates one or more combinations.
It should be noted that the medical knowledge includes but is not limited to the above-mentioned content being previously mentioned, it is all to diagnoses and treatment
(preferably diagnosing cancer of liver treatment) helpful or referential medical knowledge is covered by thought range set forth in the present invention,
Such as Experts lecture content, medicine academic exchange, academic paper is left a message and the territory of various medical knowledges, rule define
Etc..It obtains and stores above-mentioned medical knowledge data using the basic database as this system analysis guiding.
In this present embodiment, the corresponding explicit rule of the medical knowledge can be understood as written or theoretically about the state of an illness
The foundation or logic that the judgement of inspection, illness analysis and therapeutic scheme is followed.And implicit rule is primarily referred to as, it can be by calculating
The foundation or logic that machine identification or the judgement about work-up, illness analysis and therapeutic scheme read are followed (also refer to
Generation and its language or code).
For example, doctor is according to knowledge and clinical experience, according to the illness description of patient and instrument inspection knot
Fruit can tentatively judge that the disease event of patient, then foundation knowledge and clinical experience provide reliable therapeutic scheme, here
Described knowledge and clinical experience then can be understood as medical knowledge.
It is emphasized that doctor, under increasingly hard work pressure, seldom having time goes study to understand update
Medical outcome, and during liver cancer diagnosis and treatment, the diagnosis and treatment of liver cancer specification and cross discipline mastery of knowledge are also not enough
With it is timely, therefore, how to break through doctor knowledge is horizontal and clinical experience become one of liver cancer diagnosis and treatment it is urgently to be resolved
Problem.
The application needs doctor using numerous and jumbled medical knowledge data as basic data source, by Learning Algorithm
The foundation or logic that the judgement about work-up, illness analysis and therapeutic scheme to be learnt is followed, being converted to machine can
Identification or the foundation or logic read sufficiently and timely grasp the diagnosis and treatment of liver cancer specification instead of doctor and cross discipline are known with this
The medical knowledges such as knowledge.
In the embodiment of the application, the learning algorithm neural network based is by the explicit rule of the medical knowledge
The function of being then converted to implicit rule specifically includes:
A, the corresponding different symptoms of one or more groups of cases, disease are combed out according to the corresponding explicit rule of the medical knowledge
Preliminary incidence relation between disease and therapeutic scheme.
In this present embodiment, continuous by going out thousands of groups of cases based on according to medical knowledge numerous and jumbled, of a great variety
Comb out the preliminary incidence relation between different symptoms, disease and therapeutic scheme.It here can be using a variety of cases as training library
To carry out initial training by neural network, the later period is then subject to according to the real case or diagnostic norms that continually enter perfect.
For example, by clinical case or basic specialist medical knowledge, then blood urine just routine, tumor-marker can be substantially depended on
The inspection results such as object, pulmonary function test, Future liver volume, c-hepatitis antibody reason out the disease that corresponding patient may obtain, such as liver
Kinetic energy damage, cirrhosis, hydrohepatosis, liver cancer etc..It then again can be according to specific state of an illness situation, such as liver cancer early stage or advanced stage journey
Degree derives relatively reliable treatment method or scheme.
B, by neural network by input with the training of desired output obtain corresponding different symptoms information, disease information,
And between therapeutic scheme incidence relation connection weight, and form implicit rule accordingly.
In simple terms, the corresponding explicit rule of the medical knowledge is converted to implicit rule, needs neural network continuous
Practice or training input and output, that is, input one or more symptom informations, it would be desirable to access the Diseases diagnosis that I wants as a result,
When not occurring desired output result, then just modification judges the foundation or the corresponding related threshold of logic that the symptom information is followed
Value or parameter, until obtaining desired output result.
It is described to be specifically included by neural network by input and the training of desired output in the embodiment of the application:
It is trained based on one or more groups of diseases by error backpropagation algorithm.
In this present embodiment, the error backpropagation algorithm is BP algorithm
Specifically, the error backpropagation algorithm, in signal backpropagation, by output error (desired output with
The difference of reality output) it is calculated by former access anti-pass, it is reversed by hidden layer, until input layer, by error distribution during anti-pass
To each unit of each layer, the error signal of each layer each unit is obtained, and as the basis of amendment each unit weight.This
Calculating process is completed using gradient descent method, after the weight and threshold value for ceaselessly adjusting each layer neuron, subtracts error signal
It is small to arrive bottom line.
The process that weight and threshold value constantly adjust, is exactly the study and training process of network, by signal forward-propagating with
The adjustment of error back propagation, weight and threshold value is repeated, and is performed until preset learning training number, or output
Error is reduced to the degree of permission.
In this present embodiment, by the training of step B, available connection weight and respective threshold, and can based on this
Relatively reliable makes what is followed between machine recognition or the corresponding different symptoms information of reading, disease information and therapeutic scheme to patrol
The relationship of collecting.
In this present embodiment, continuing on with system passes through knowledge base and neural network after having the input of new example
Study to new example automatically obtains new connection weight distribution, to have updated knowledge base and new expert database.
Knowledge base 120
The knowledge base 120, for storing and providing corresponding treatment side for the different state of an illness according to the implicit rule
Case.
In simple terms, it may determine that is followed between different symptoms information, disease information and therapeutic scheme patrols in formation
After the connection weight for the relationship of collecting, then it can be inputted to obtain desired diagnosis and treatment project output according to symptom.
In the embodiment of the application, the therapeutic scheme includes: operation excision, TACE, radiotherapy, supports to suit the medicine to the illness and control
Any one or more in treatment, local ablation, liver transfer operation, molecular targeted therapy and system chemotherapy combines；The therapeutic scheme is made
It is stored in the expert database for the medical knowledge.
In the embodiment of the application, the knowledge base 120 includes: that subsystem 121, condition assessment subsystem are instructed in detection
System 122 and Scheme Choice subsystem 123.
In this present embodiment, 3 subsystems of herein described knowledge base 120 are with the newest " Primary Hepatic in China
Cancer diagnosis and treatment specification 2017 editions " it is reference guide.And the application is referring also to such as Japan, the newer liver of the other countries such as U.S.
Cancer practice guidelines.
Subsystem 121 is instructed in detection
Subsystem 121 is instructed in the detection, for collecting patient assessment's intention and essential information, for determining reviewing party
Case.
In this present embodiment, the detection instructs subsystem 121 for tentatively collecting patient assessment's intention, essential information
Afterwards, adjuvant clinical doctor determines corresponding inspection item, further to define diagnosing cancer of liver.Pass through the medical intention and base of patient
This information can tentatively judge where body specifically goes wrong, and it is further also correspondingly to may determine which place needs
It checks.These information can be manually entered by doctor, e.g., thoracic cavity pain, expiratory dyspnea, or can according to previous case rapidly extracting
The position that can be gone wrong.
In the embodiment of the application, the inspection scheme includes: blood urine just routine, biochemical full item, coagulation function, first
Fetoprotein is quantitative, tumor markers, hepatitis B series, HBV-DNA are quantitative, c-hepatitis antibody, HCV-mRNA are quantitative, colored super
Sound, ultrasonic contrast, Enhanced CT, enhancing nuclear magnetic resonance, chest X-ray, PET-CT, ECT bone scanning, hepatic arteriography
(DSA), electrocardiogram, echocardiogram, pulmonary function test, biopsy biopsy, molecule class detection, indocyanine-green excretion experiment,
Any one or more in Future liver volume and other disease examinations combines.
It should be noted that the inspection scheme includes but is not limited to the above-mentioned content being previously mentioned, it is all to belong to determining disease
Examination scope required for disease (preferably diagnosing cancer of liver treatment) is covered by thought range set forth in the present invention.
For example, according to the medical purpose of patient and patient groups' essential information, it is divided into healthy population screening, liver cancer is controlled
Inspection, suspicious liver neoplasm make a definite diagnosis three classes before treating, and are being subdivided according to three classes crowd to obtain different inspection schemes.It is specific next
It says, the corresponding detection instructs the flow diagram of the inspection and guide of subsystem 121 to can be found in Fig. 2.As shown, the detection
Subsystem 121 is instructed to be inputted according to user, by exporting corresponding inspection scheme after Fig. 2 analysing patient's condition.Wherein, C1-C4 is respectively
Four kinds of decision conditions, C1: medical treatment scene；C2: the specific sings and symptoms of patient；C3: tumor size type；C4: other diseases
Situation.G1-G11 is 11 kinds of inspection schemes.Specifically, subdivided according to patient assessment's intention and essential information obtain it is as follows
1 liver cancer inspection and guide table of table.G1-G11 totally 11 kinds of inspection schemes are enumerated in table.
1 liver cancer detection scheme table of table
Condition assessment subsystem 122
The condition assessment subsystem 122, for the inspection result according to the corresponding inspection scheme to determine that illness is believed
Breath, and corresponding disease information is evaluated according to the implicit rule.
In this present embodiment, after the condition assessment subsystem 122 completes clinical examination for patient, adjuvant clinical doctor
Teacher assesses conditions of patients, lays a solid foundation for selection therapeutic scheme.The condition assessment subsystem 122 is broadly divided into the course of disease
Resolver and staging thruster.Its state of an illness will further be assessed according to various clinical evidences are collected, and specify liver cancer point
Phase.
In the embodiment of the application, the condition assessment subsystem includes: course of disease resolver and staging thrust
Device.
In this present embodiment, course of disease parsing implement body can be understood as the implicit rule module comprising connection weight.
In the embodiment of the application, process includes: liver function Child points by stages in the staging thruster
Grade, PS scoring, neoplastic state, whether there is or not vascular invasion, whether there is or not DISTANT METASTASES IN, tumor number, tumor size, bilirubin level, have
Any one or more in no portal hypertension combines.
As shown in figure 3, the flow diagram of the diagnosing cancer of liver being shown as in the present embodiment by stages.As shown, illustrating
Condition assessment result including the clinical stages of patient.The condition assessment subsystem 122 can help doctor to simplify work
Make, more comprehensively understanding conditions of patients, avoid omitting, selects therapeutic scheme to provide support for doctor.
Scheme Choice subsystem 123
The Scheme Choice subsystem 123, for providing the disease information corresponding treatment according to the implicit rule
Scheme.
In the embodiment of the application, the therapeutic scheme includes: operation excision, TACE, radiotherapy, supports to suit the medicine to the illness and control
Any one or more in treatment, local ablation, liver transfer operation, molecular targeted therapy and system chemotherapy combines；The therapeutic scheme is made
It is stored in the expert database for the medical knowledge.
In this present embodiment, the Scheme Choice subsystem 123 is on the basis of clear conditions of patients, assisting clinical doctor
Teacher selects the therapeutic scheme for being most suitable for patient, is the committed step of specification liver cancer treatment.
For example, the Scheme Choice subsystem 123 is according to diagnosis and treatment specification, according to dicision of diagnosis and treatment rule shown in Fig. 4
Carry out decision.
Inference machine 130
It is shown in Figure 1, the system 100 further include: inference machine 130, for being directed to the approximate state of an illness for the knowledge base
Therapeutic scheme comparative analysis in 120 therapeutic schemes provided and example case, to obtain analysis result.
In the embodiment of the application, the analysis result is for updating the knowledge base 120 or the patent data
The connection weight of the implicit rule is corresponded in library 110；And/or the analysis result is for explaining the therapeutic scheme.
In this present embodiment, inference machine is inference mechanism neural network based, is a kind of matrix majorization inference machine MACIE
(abbreviation of Matrix Controlled Inference Engine).
In this present embodiment, after analyzing difference existing for two kinds of therapeutic schemes, the analysis result be used as with reference to
The connection weight of the implicit rule is modified or updated, i.e., further constantly corrects the company for improving neural metwork training in the application
Connect weight.In addition, the analysis result can be also used for explaining the therapeutic scheme that the knowledge base 120 provides.
Diagnosis for liver cancer system 100 neural network based described herein will specifically be carried out by following embodiment
It further illustrates.
As shown in figure 5, being shown as the flow diagram of neural network in knowledge base.As shown in the figure, wherein (x1,x2,...,
x6) it is that symptom inputs；x7,x8For disease intermediate node；x9,x10,x11Node layer is exported for therapeutic scheme；x12,x13,x14It is additional
Node layer is increased because of the needs of learning algorithm；ωijConnection weight matrix is constituted, in this matrix, as i >=j,
ωij=0；When between i < j and node i and j there is no arc is connected, ωij=0；Remaining ωijFor the data marked on arc；It is special
Property function be a discrete type threshold function table, calculation are as follows:
Wherein, xjFor node j weighted input and；x'jFor the output of node j.
In addition, calculating for convenience, w is increased in formula0jx0, x0Take constant 1, x0=1, w0jValue be marked on node j
Circle in, the opposite number of its practical upper threshold value, connection weight wijIt is that one group of case passes through error backpropagation algorithm (BP study
Algorithm) training obtains.By all wijValue and the set that constitutes of various symptoms, disease and therapeutic scheme form medical diagnosis on disease
The knowledge base of expert system.
The reasoning of this expert system is completed by network query function.First by initial symptoms X=provided by user
(x1, x2..., x)6Network is inputted, by the disease Y=(x that can be calculated patient7, x8) and for disease selection treatment side
Case Y=(x9, x10, x11).For example, certain patient has symptom x2And x3, for symptom x4, x5, x6No record, then it is defeated to obtain symptom
Incoming vector X=(1, -1, -1,0,0) carries out grid computing:
X7The > 0 of=0 × 1+2 × 1+ (- 2) × (- 1)+3 × (- 1)=1
X'7=1
X7=(- 1) × 1+0 × 3+0 × 3+3 × (- 1)=4 < 0
X'8=-1
It thus may know that patient with disease x7, and do not suffer from disease x8.Further calculate:
X9=(- 2) × 1+2 × 1+2 × (- 1)+3 × (- 4) × 0=-2 < 0
X'9=-1
X10=(- 1) × 1+1 × 1+ (- 4) × (- 1)+3 × (- 1)=1 > 0
X'10=1
X12=2 × 1+ (- 4) × (- 1)+5 × 1=11 > 0
X'12=1
X13=3 × 1+ (- 2) × (- 1)+2 × 1=7 > 0
X'13=1
X14=0 × 1+ (- 1) × (- 1)+(- 3) × 1=-2 < 0
X'12=-1
X11The > 0 of=3 × 1+ (- 3) × 1+ (- 3) × (- 1)+(- 3) × 1+1 × 1+ (- 1) × (- 1)=2
X'11=1
To know that patient needs to select therapeutic scheme X10And X11It is treated, for doctor's reference.
It should be understood that the division of each database (module) of system above is only a kind of division of logic function, it is practical real
It can completely or partially be integrated on a physical entity now, it can also be physically separate.And these modules can all with
Software is realized by way of processing element calls；It can also all realize in the form of hardware；It can be passed through with part of module
Processing element calls the form of software to realize that part of module passes through formal implementation of hardware.For example, sorter model training module
It can be the processing element individually set up, also can integrate and realized in some chip of above system, in addition it is also possible to
The form of program code is stored in the memory of above system, called by some processing element of above system and executed with
The function of upper sorter model training module.The realization of other modules is similar therewith.Furthermore these modules completely or partially can be with
It integrates, can also independently realize.Processing element described here can be a kind of integrated circuit, the processing with signal
Ability.During realization, each step of the above method or the above modules can pass through the hardware in processor elements
The instruction of integrated logic circuit or software form is completed.
For example, the above module can be arranged to implement one or more integrated circuits of above method, such as:
One or more specific integrated circuits (Application Specific Integrated Circuit, abbreviation ASIC), or,
One or more microprocessors (digital signal processor, abbreviation DSP), or, one or more scene can compile
Journey gate array (Field Programmable Gate Array, abbreviation FPGA) etc..For another example, when some above module passes through place
When managing the form realization of element scheduler program code, which can be general processor, such as central processing unit
(Central Processing Unit, abbreviation CPU) or it is other can be with the processor of caller code.For another example, these modules
It can integrate together, realized in the form of system on chip (system-on-a-chip, abbreviation SOC).
As shown in fig. 6, showing that the process of diagnosis for liver cancer method neural network based in one embodiment of the application is shown
It is intended to.
In some embodiments, the method can be applied to controller, such as: ARM controller, FPGA controller, SoC
Controller, dsp controller or MCU controller etc..In some embodiments, the method can also be applied to include depositing
Reservoir, storage control, one or more processing units (CPU), Peripheral Interface, RF circuit, voicefrequency circuit, loudspeaker, Mike
Wind, input/output (I/O) subsystem, display screen, other outputs or the computer for controlling the components such as equipment and outside port；
The computer includes but is not limited to such as desktop computer, laptop, tablet computer, smart phone, smart television, a number
The PCs such as word assistant (Personal Digital Assistant, abbreviation PDA).In other embodiments, described
Method applies also for server, and the server can be arranged in one or more real according to many factors such as function, loads
On body server, it can also be made of server cluster be distributed or concentration.
In this present embodiment, which comprises
Step S601: storage medical knowledge, and learning algorithm neural network based is corresponding aobvious by the medical knowledge
Formula rule is converted to the implicit rule comprising connection weight.
Step S602: corresponding therapeutic scheme is provided for the different state of an illness according to the implicit rule.
It is and upper it should be noted that the embodiment of the diagnosis for liver cancer method neural network based of the present embodiment
The embodiment of diagnosis for liver cancer system neural network based is similar in text, and so it will not be repeated.
As shown in fig. 7, showing that the structure of diagnosis for liver cancer equipment neural network based in one embodiment of the application is shown
It is intended to.The equipment 700 includes: memory 701 and processor 702, and the memory 701 is stored with computer program, described
Processor 702 is used to execute the computer program that the memory 701 stores, so as to realize such as Fig. 6 when the equipment 700 executes
The diagnosis for liver cancer method neural network based.
The memory 701 may include random access memory (Random Access Memory, abbreviation RAM),
It may further include nonvolatile memory (non-volatile memory), for example, at least a magnetic disk storage.
The processor 702 can be general processor, including central processing unit (Central Processing Unit,
Abbreviation CPU), network processing unit (Network Processor, abbreviation NP) etc.；It can also be digital signal processor
(Digital Signal Processing, abbreviation DSP), specific integrated circuit (Application Specific
Integrated Circuit, abbreviation ASIC), field programmable gate array (Field-Programmable Gate Array,
Abbreviation FPGA) either other programmable logic device, discrete gate or transistor logic, discrete hardware components.
In order to achieve the above objects and other related objects, the application provides a kind of computer readable storage medium, deposits thereon
Computer program is contained, diagnosis for liver cancer neural network based as described in Figure 6 is realized when which is executed by processor
Method.
The computer readable storage medium, those of ordinary skill in the art will appreciate that: realize that above-mentioned each method is implemented
The all or part of the steps of example can be completed by the relevant hardware of computer program.Computer program above-mentioned can store
In a computer readable storage medium.When being executed, execution includes the steps that above-mentioned each method embodiment to the program；And it is aforementioned
Storage medium include: the various media that can store program code such as ROM, RAM, magnetic or disk.
In conclusion a kind of diagnosis for liver cancer system, method, equipment and the medium neural network based of the application,
By storing medical knowledge, and the corresponding explicit rule of the medical knowledge is converted to packet by learning algorithm neural network based
Implicit rule containing connection weight；Corresponding therapeutic scheme is provided for the different state of an illness according to the implicit rule.
The application effectively overcomes various shortcoming in the prior art and has high industrial utilization value.
The principles and effects of the application are only illustrated in above-described embodiment, not for limitation the application.It is any ripe
Know the personage of this technology all can without prejudice to spirit herein and under the scope of, carry out modifications and changes to above-described embodiment.Cause
This, those of ordinary skill in the art is complete without departing from spirit disclosed herein and institute under technical idea such as
At all equivalent modifications or change, should be covered by claims hereof.
Claims (24)
1. a kind of diagnosis for liver cancer system neural network based, which is characterized in that the system comprises:
Expert database, for storing medical knowledge, and learning algorithm neural network based is corresponding by the medical knowledge
Explicit rule is converted to the implicit rule comprising connection weight；
Knowledge base, for storing and providing corresponding therapeutic scheme for the different state of an illness according to the implicit rule.
2. diagnosis for liver cancer system neural network based according to claim 1, which is characterized in that the medicine is known
Knowledge includes: expertise, case, clinography, medical consultations specification, medical consultations guide, basis/general medical knowledge, doctor
Term periodical, medical thesis, any one or more combination in Physicians report and therapeutic scheme.
3. diagnosis for liver cancer expert system neural network based according to claim 1, which is characterized in that the base
The function that the explicit rule of the medical knowledge is converted to implicit rule is specifically included in the learning algorithm of neural network:
The corresponding different symptoms of one or more groups of cases, disease are combed out according to the corresponding explicit rule of the medical knowledge and are controlled
Preliminary incidence relation between treatment scheme；
Corresponding different symptoms information, disease information and treatment are obtained with the training of desired output by input by neural network
The connection weight of incidence relation between scheme, and implicit rule is formed accordingly.
4. diagnosis for liver cancer system neural network based according to claim 3, which is characterized in that described by nerve
Network is specifically included by input and the training of desired output: being instructed based on one or more groups of diseases by error backpropagation algorithm
Practice.
5. diagnosis for liver cancer system neural network based according to claim 3, which is characterized in that the knowledge base
Include:
Subsystem is instructed in detection, for collecting patient assessment's intention and essential information, for determining inspection scheme；
Condition assessment subsystem, for the inspection result according to the corresponding inspection scheme to determine illness information, and according to institute
It states implicit rule and evaluates corresponding disease information；
Scheme Choice subsystem, for providing the disease information corresponding therapeutic scheme according to the implicit rule.
6. diagnosis for liver cancer system neural network based according to claim 5, which is characterized in that the reviewing party
Case includes: that blood urine is just conventional, biochemical item, coagulation function, alpha-fetoprotein entirely are quantitative, tumor markers, hepatitis B series, HBV-
DNA is quantitative, c-hepatitis antibody, HCV-mRNA is quantitative, colorful ultrasonic, ultrasonic contrast, Enhanced CT, enhances nuclear magnetic resonance, chest
X-ray, PET-CT, ECT bone scanning, hepatic arteriography (DSA), electrocardiogram, echocardiogram, pulmonary function test, biopsy
Any one or more in biopsy, the detection of molecule class, indocyanine-green excretion experiment, Future liver volume and other disease examinations
Combination.
7. diagnosis for liver cancer system neural network based according to claim 5, which is characterized in that the state of an illness is commented
Estimating subsystem includes: course of disease resolver and staging thruster.
8. diagnosis for liver cancer system neural network based according to claim 7, which is characterized in that the disease point
In phase thruster by stages process include: liver function Child classification, PS scoring, neoplastic state, whether there is or not vascular invasion, whether there is or not at a distance
Transfer, tumor number, tumor size, bilirubin level, whether there is or not any one or more in portal hypertension to combine.
9. diagnosis for liver cancer system neural network based according to claim 1 or 5, which is characterized in that described to control
Treatment scheme includes: operation excision, TACE, radiotherapy, supports symptomatic treatment, local ablation, liver transfer operation, molecular targeted therapy and be
Any one or more combination in chemotherapy of uniting；The therapeutic scheme is stored in the expert database as the medical knowledge
In.
10. diagnosis for liver cancer system neural network based according to claim 1, which is characterized in that the system
Further include: inference machine, the treatment side in therapeutic scheme and example case for providing the knowledge base for the approximate state of an illness
Case comparative analysis, to obtain analysis result.
11. diagnosis for liver cancer system neural network based according to claim 10, which is characterized in that the analysis
As a result for updating the connection weight for corresponding to the implicit rule in the knowledge base or the patent database；And/or it is described
Analysis result is for explaining the therapeutic scheme.
12. a kind of diagnosis for liver cancer method neural network based, which is characterized in that the described method includes:
Medical knowledge is stored, and the corresponding explicit rule of the medical knowledge is converted to packet by learning algorithm neural network based
Implicit rule containing connection weight；
Corresponding therapeutic scheme is provided for the different state of an illness according to the implicit rule.
13. diagnosis for liver cancer method neural network based according to claim 12, which is characterized in that the medicine
Knowledge include: expertise, case, clinography, medical consultations specification, medical consultations guide, basis/general medical knowledge,
Any one or more in medical journals, medical thesis and Physicians report combines.
14. diagnosis for liver cancer method neural network based according to claim 12, which is characterized in that described to be based on
The method that the explicit rule of the medical knowledge is converted to implicit rule includes: by the learning algorithm of neural network
The corresponding different symptoms of one or more groups of cases, disease are combed out according to the corresponding explicit rule of the medical knowledge and are controlled
Preliminary incidence relation between treatment scheme；
Corresponding different symptoms information, disease information and treatment are obtained with the training of desired output by input by neural network
The connection weight of incidence relation between scheme, and implicit rule is formed accordingly.
15. diagnosis for liver cancer method neural network based according to claim 14, which is characterized in that described by mind
It is specifically included through network by input and the training method of desired output: passing through error backpropagation algorithm based on one or more groups of diseases
It is trained.
16. diagnosis for liver cancer method neural network based according to claim 14, which is characterized in that the foundation
The implicit rule is directed to the method that the different state of an illness provide corresponding therapeutic scheme
Patient assessment's intention and essential information are collected, for determining inspection scheme；
Inspection result according to the corresponding inspection scheme evaluates correspondence according to the implicit rule to determine illness information
Disease information；
There is provided the disease information corresponding therapeutic scheme according to the implicit rule.
17. diagnosis for liver cancer method neural network based according to claim 16, which is characterized in that the inspection
Scheme include: quantitative just conventional blood urine, biochemical full item, coagulation function, alpha-fetoprotein, tumor markers, hepatitis B series,
HBV-DNA is quantitative, c-hepatitis antibody, HCV-mRNA are quantitative, colorful ultrasonic, ultrasonic contrast, Enhanced CT, enhancing nuclear magnetic resonance,
Chest X-ray, PET-CT, ECT bone scanning, hepatic arteriography (DSA), electrocardiogram, echocardiogram, pulmonary function test, puncture
Pathological biopsy, the detection of molecule class, indocyanine-green excretion experiment, any one in Future liver volume and other disease examinations or
Multiple combinations.
18. diagnosis for liver cancer method neural network based according to claim 16, which is characterized in that the foundation
The inspection result of the corresponding inspection scheme evaluates corresponding disease letter according to the implicit rule to determine illness information
The method of breath includes: to be completed by course of disease resolver and staging thruster.
19. diagnosis for liver cancer method neural network based according to claim 18, which is characterized in that the disease
By stages in thruster by stages process include: liver function Child classification, PS scoring, neoplastic state, whether there is or not vascular invasion, whether there is or not remote
Place's transfer, tumor number, tumor size, bilirubin level, whether there is or not any one or more in portal hypertension to combine.
20. diagnosis for liver cancer method neural network based described in 2 or 16 according to claim 1, which is characterized in that described
Therapeutic scheme include: operation excision, TACE, radiotherapy, support symptomatic treatment, local ablation, liver transfer operation, molecular targeted therapy and
Any one or more in system chemotherapy combines；The therapeutic scheme is as the medical knowledge to be stored.
21. diagnosis for liver cancer method neural network based according to claim 12, which is characterized in that the method
Further include: the therapeutic scheme comparative analysis in therapeutic scheme and example sample that the knowledge base is provided for the approximate state of an illness,
To obtain analysis result.
22. diagnosis for liver cancer method neural network based according to claim 21, which is characterized in that the analysis
As a result for updating the connection weight for corresponding to the implicit rule in the knowledge base or the patent database；And/or it is described
Analysis result is for explaining the therapeutic scheme.
23. a kind of diagnosis for liver cancer equipment neural network based, which is characterized in that the equipment includes: memory and place
Manage device；
The memory is for storing computer program；The processor is used to execute the computer journey of the memory storage
Sequence, so as to realize that the liver cancer neural network based as described in any one of claim 12 to 21 is auxiliary when the equipment executes
Auxiliary diagnosis method.
24. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer program
The diagnosis for liver cancer neural network based as described in any one of claim 12 to 21 is realized when being executed by processor
Method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910156222.5A CN109859841A (en) | 2019-03-01 | 2019-03-01 | Diagnosis for liver cancer system, method, equipment and medium neural network based |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910156222.5A CN109859841A (en) | 2019-03-01 | 2019-03-01 | Diagnosis for liver cancer system, method, equipment and medium neural network based |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109859841A true CN109859841A (en) | 2019-06-07 |
Family
ID=66899628
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910156222.5A Pending CN109859841A (en) | 2019-03-01 | 2019-03-01 | Diagnosis for liver cancer system, method, equipment and medium neural network based |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109859841A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110500709A (en) * | 2019-07-29 | 2019-11-26 | 七彩安科智慧科技有限公司 | The online method of discrimination of conditioners running conditions |
| CN111028934A (en) * | 2019-12-23 | 2020-04-17 | 科大讯飞股份有限公司 | Diagnostic quality inspection method, diagnostic quality inspection device, electronic equipment and storage medium |
| CN111354459A (en) * | 2020-02-25 | 2020-06-30 | 成都联客信息技术有限公司 | Auxiliary diagnosis expert system for traditional Chinese medicine massage |
| CN111701150A (en) * | 2020-07-02 | 2020-09-25 | 中国科学院苏州生物医学工程技术研究所 | Intelligent optical diagnosis and treatment equipment |
| CN111768866A (en) * | 2020-06-30 | 2020-10-13 | 四川大学华西医院 | System in virtual ward of painful acute surgery |
| CN113298752A (en) * | 2020-02-06 | 2021-08-24 | 西门子医疗有限公司 | Method, system, program and medium for automatically characterizing liver tissue of a patient |
| CN116434960A (en) * | 2023-06-13 | 2023-07-14 | 中国人民解放军总医院 | Cardiogenic shock clinical decision support system, equipment and storable medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108154928A (en) * | 2017-12-27 | 2018-06-12 | 北京嘉和美康信息技术有限公司 | A kind of methods for the diagnosis of diseases and device |
| CN109192299A (en) * | 2018-08-13 | 2019-01-11 | 中国科学院计算技术研究所 | A kind of medical analysis auxiliary system based on convolutional neural networks |
-
2019
- 2019-03-01 CN CN201910156222.5A patent/CN109859841A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108154928A (en) * | 2017-12-27 | 2018-06-12 | 北京嘉和美康信息技术有限公司 | A kind of methods for the diagnosis of diseases and device |
| CN109192299A (en) * | 2018-08-13 | 2019-01-11 | 中国科学院计算技术研究所 | A kind of medical analysis auxiliary system based on convolutional neural networks |
Non-Patent Citations (2)
| Title |
|---|
| 罗保平等: "人工神经网络在晚期肝癌治疗计划系统中的应用", 《华中科技大学学报(医学版)》 * |
| 陈斌才等: "基于BP神经网络的医疗诊断专家系统", 《医学信息》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110500709A (en) * | 2019-07-29 | 2019-11-26 | 七彩安科智慧科技有限公司 | The online method of discrimination of conditioners running conditions |
| CN111028934A (en) * | 2019-12-23 | 2020-04-17 | 科大讯飞股份有限公司 | Diagnostic quality inspection method, diagnostic quality inspection device, electronic equipment and storage medium |
| CN111028934B (en) * | 2019-12-23 | 2022-02-18 | 安徽科大讯飞医疗信息技术有限公司 | Diagnostic quality inspection method, diagnostic quality inspection device, electronic equipment and storage medium |
| CN113298752A (en) * | 2020-02-06 | 2021-08-24 | 西门子医疗有限公司 | Method, system, program and medium for automatically characterizing liver tissue of a patient |
| US11861827B2 (en) | 2020-02-06 | 2024-01-02 | Siemens Healthcare Gmbh | Techniques for automatically characterizing liver tissue of a patient |
| CN111354459A (en) * | 2020-02-25 | 2020-06-30 | 成都联客信息技术有限公司 | Auxiliary diagnosis expert system for traditional Chinese medicine massage |
| CN111768866A (en) * | 2020-06-30 | 2020-10-13 | 四川大学华西医院 | System in virtual ward of painful acute surgery |
| CN111701150A (en) * | 2020-07-02 | 2020-09-25 | 中国科学院苏州生物医学工程技术研究所 | Intelligent optical diagnosis and treatment equipment |
| CN116434960A (en) * | 2023-06-13 | 2023-07-14 | 中国人民解放军总医院 | Cardiogenic shock clinical decision support system, equipment and storable medium |
| CN116434960B (en) * | 2023-06-13 | 2023-10-13 | 中国人民解放军总医院 | Cardiogenic shock clinical decision support system, equipment and storable medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109859841A (en) | Diagnosis for liver cancer system, method, equipment and medium neural network based | |
| Kao et al. | Context-aware symptom checking for disease diagnosis using hierarchical reinforcement learning | |
| US10902588B2 (en) | Anatomical segmentation identifying modes and viewpoints with deep learning across modalities | |
| Crimi et al. | Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries: 6th International Workshop, BrainLes 2020, Held in Conjunction with MICCAI 2020, Lima, Peru, October 4, 2020, Revised Selected Papers, Part I | |
| Pei et al. | A review of the application of multi-modal deep learning in medicine: bibliometrics and future directions | |
| Su et al. | Attention2majority: Weak multiple instance learning for regenerative kidney grading on whole slide images | |
| CN117316466B (en) | Clinical decision method, system and equipment based on knowledge graph and natural language processing technology | |
| Stoean et al. | Ensemble of classifiers for length of stay prediction in colorectal cancer | |
| Huang et al. | Medical knowledge-based network for patient-oriented visual question answering | |
| Venkataramani et al. | Towards continuous domain adaptation for medical imaging | |
| Hou et al. | Multi-label learning with visual-semantic embedded knowledge graph for diagnosis of radiology imaging | |
| Kadi et al. | Systematic mapping study of data mining–based empirical studies in cardiology | |
| Sarp et al. | A highly transparent and explainable artificial intelligence tool for chronic wound classification: XAI-CWC | |
| Tian et al. | Fully-automated functional region annotation of liver via a 2.5 D class-aware deep neural network with spatial adaptation | |
| Shaheen et al. | Hi-Le and HiTCLe: Ensemble Learning Approaches for Early Diabetes Detection using Deep Learning and eXplainable Artificial Intelligence | |
| Sun et al. | ExpertNet: Defeat noisy labels by deep expert consultation paradigm for pneumoconiosis staging on chest radiographs | |
| Ali et al. | Automated segmentation of brain tumour images using deep learning-based model VGG19 and ResNet 101 | |
| Mehrankia et al. | Prediction of heart attacks using biological signals based on recurrent GMDH neural network | |
| Zhang et al. | Bert with enhanced layer for assistant diagnosis based on Chinese obstetric EMRs | |
| Cen et al. | Contextual embedding bootstrapped neural network for medical information extraction of coronary artery disease records | |
| Li et al. | Kinet: a non-invasive method for predicting Ki67 index of glioma | |
| Raj | Enhancing Thyroid Cancer Diagnostics Through Hybrid Machine Learning and Metabolomics Approaches. | |
| Gupta et al. | Unification of machine learning and blockchain technology in healthcare industry | |
| Lu et al. | Prediction of fractional flow reserve with enhanced ant lion optimized support vector machine | |
| Hu et al. | Slimmable transformer with hybrid axial-attention for medical image segmentation |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190607 |