CN113096828B - Diagnosis, prediction and major health management platform based on cancer genome big data core algorithm - Google Patents

Diagnosis, prediction and major health management platform based on cancer genome big data core algorithm Download PDF

Info

Publication number
CN113096828B
CN113096828B CN202110416120.XA CN202110416120A CN113096828B CN 113096828 B CN113096828 B CN 113096828B CN 202110416120 A CN202110416120 A CN 202110416120A CN 113096828 B CN113096828 B CN 113096828B
Authority
CN
China
Prior art keywords
gene
data
big
management platform
health management
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
Application number
CN202110416120.XA
Other languages
Chinese (zh)
Other versions
CN113096828A (en
Inventor
王奔
余鹏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Xikang Software Co ltd
Original Assignee
Xikang Software Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Xikang Software Co ltd filed Critical Xikang Software Co ltd
Priority to CN202110416120.XA priority Critical patent/CN113096828B/en
Publication of CN113096828A publication Critical patent/CN113096828A/en
Application granted granted Critical
Publication of CN113096828B publication Critical patent/CN113096828B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

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
    • G16H80/00—ICT specially adapted for facilitating communication between medical practitioners or patients, e.g. for collaborative diagnosis, therapy or health monitoring
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/084—Backpropagation, e.g. using gradient descent
    • 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
    • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
    • 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
    • G16H50/50—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for simulation or modelling of medical disorders
    • 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

Abstract

The cancer genome big data core algorithm-based diagnosis, prediction and big health management platform comprises a client, a data terminal and a big health management platform; a user registers through a client, and after the registration is completed, gene data are sent to a big health management platform through the client; the data terminal periodically acquires gene big data and transmits the acquired gene big data to a big health management platform; the big health management platform is used for storing and preprocessing gene big data, establishing a disease diagnosis model according to the preprocessed gene big data, diagnosing according to the gene data of a user by the disease diagnosis model, and sending a diagnosis result of the disease diagnosis model to a corresponding client. The invention has the beneficial effects that: the mode of preprocessing the disease big data has higher clustering accuracy, and then the class obtained by clustering is used for training the BP neural network, so that the structure of the gene big data can be effectively simplified, and the complexity and overfitting of algorithm learning of the BP neural network are avoided.

Description

Diagnosis, prediction and major health management platform based on cancer genome big data core algorithm
Technical Field
The invention relates to the field of health management, in particular to a diagnosis, prediction and major health management platform based on a cancer genome major data core algorithm.
Background
Scientific and technical progress promotes rapid innovation of various industries, particularly the success of whole gene sequencing in the aspect of biology, so that the acquisition cost of cancer gene expression data is reduced sharply, and a wide platform is provided for systematic research of cancer genomes. Based on the gene data, the computer aided diagnosis is carried out on the gene data by machine learning, and the precision of pathological change diagnosis is higher. However, the characteristics of the gene data are high dimension, small sample size, large signal-to-noise ratio, high feature dimension and strong correlation, and when the machine learning classification algorithm is applied to the gene data, the trailed fitting and dimension disaster are easily caused. Therefore, how to dig out valuable information in such gene data sets is a hot issue of research. Clustering is an important method for preprocessing big data, and is widely applied to the fields of medical data analysis, image processing, text processing and the like, because many data in the fields have the characteristics of few samples, high noise and high dimension, the cluster analysis is carried out on the gene big data, the class structure implicit in the gene data can be found, and the data objects are divided into different clusters or classes, so that the similarity between the objects in the same class is larger, and the similarity between the objects in different classes is smaller, the structure of the big data can be effectively simplified, and the noise data in the big data can be removed simultaneously in the clustering process, thereby avoiding the influence of the noise data on the accuracy of a machine classification algorithm.
Disclosure of Invention
In view of the above problems, the present invention aims to provide a diagnosis, prognosis and major health management platform based on cancer genome big data core algorithm.
The purpose of the invention is realized by the following technical scheme:
the cancer genome big data core algorithm-based diagnosis, prediction and big health management platform comprises a client, a data terminal and a big health management platform;
a user registers through a client, and after the registration is completed, gene data are sent to a big health management platform through the client;
the data terminal is used for periodically acquiring gene big data according to a preset acquisition cycle, transmitting the acquired gene big data to the big health management platform after the current acquisition cycle is finished, and updating the gene big data stored in the big health management platform;
the big health management platform comprises a gene database, a gene data preprocessing module, a disease diagnosis model and a disease diagnosis report generating module, wherein the gene database stores gene big data by adopting block link points, when new gene big data are stored in the gene database, the gene data preprocessing module is used for preprocessing the gene big data stored in the gene database, and the preprocessed gene big data are used for retraining a BP (back propagation) neural network, so that the disease diagnosis model based on the BP neural network is generated, when the big health management platform receives the gene data sent by a client, the disease diagnosis report generating module inputs the received gene data into the disease diagnosis model, and sends a diagnosis result of the disease diagnosis model to the corresponding client.
Preferably, when the preprocessed gene big data are used for training the BP neural network, the gene data in each class obtained by clustering the gene data preprocessing module are used for respectively training the BP neural network, so that a disease diagnosis model based on the BP neural network is obtained.
Preferably, the gene data preprocessing module clusters the gene big data by adopting a density peak value clustering algorithm, and improves a calculation mode of local density of the data in the density peak value clustering algorithm, specifically:
let Y denote the Gene big data set, YiRepresents the ith gene data, N (Y) in the gene big data set Yi) Representing gene data yiAnd N (y) isi)＝{yi，j||yi，j-yi|≤dc，j＝1，2，...，M(yi) In which d iscDenotes the truncation distance, yi，jA set of representations N (y)i) J (th) gene data of (1), M (y)i) A set of representations N (y)i) The amount of gene data in (a); definition m (y)i) Representing gene data yiLocal density value of m (y)i) The values of (A) are:
Figure BDA0003026870220000021
wherein, ω (y)i) Number of genesAccording to yiAnd ω (y) is the local density compensation coefficient ofi) The values of (A) are:
Figure BDA0003026870220000022
wherein, let yi，lRepresents the set N (y)i) The first gene data in (1), and yi，l≠yi，j，N(yi，j) Representing gene data yi，jEta (y) of the neighborhood gene data seti，l，N(yi，j) ) represents gene data yi，lAnd the set N (y)i，j) A judgment function in between, and
Figure BDA0003026870220000023
f(yi，j，yi) Representing gene data yi，jAnd gene data yiA judgment function in between, and
Figure BDA0003026870220000024
J(yi，j) Representing gene data yi，jIn the set N (y)i) A density coefficient of medium, and
Figure BDA0003026870220000025
J(yi) Representing gene data yiIn the set N (y)i) A density coefficient of medium, and
Figure BDA0003026870220000031
theta is a given positive integer to ensure that the denominator is not 0.
Preferably, the gene data preprocessing module clusters the gene big data by using a density peak clustering algorithm, corrects a local density value of each gene data in the gene big data set Y, and sets m' (Y ″)i) Representation of the Gene data yiThe corrected local density value is m' (y)i) The values of (A) are:
m′(yi)＝m(yi)ρ(yi)
Figure BDA0003026870220000032
in the formula, yi，kRepresents the set N (y)i) Of (a) and yi，k≠yi，ρ(yi) Representing gene data yiCorresponding noise penalty factor, m (y)i，k) Representing gene data yi，kLocal density value of s (y)i) Representing gene data yiThe area of (a) detects the coefficient, and
Figure BDA0003026870220000037
s (Y) is a region detection coefficient corresponding to the whole gene data in the gene big data set Y, and
Figure BDA0003026870220000033
wherein, the first and the second end of the pipe are connected with each other,
Figure BDA0003026870220000034
means of region detection coefficients representing gene data in the gene big data set Y, and
Figure BDA0003026870220000035
σ (Y) represents the mean square error of the region detection coefficients of the gene data in the gene big data set Y, and
Figure BDA0003026870220000036
α represents a regulatory parameter, and M (Y) represents the amount of gene data in the gene big data set Y.
The beneficial effects created by the invention are as follows: based on the gene big data, the BP neural network algorithm is used for learning the gene big data, so that a disease diagnosis model is established for diagnosing diseases, and the precision of disease diagnosis is high; aiming at the characteristics of large gene data, such as high dimensionality, large signal-to-noise ratio and strong correlation, when the BP neural network algorithm is applied to the large gene data, training overfitting and high complexity are easily caused, and aiming at the defects, the invention carries out clustering pretreatment on the large gene data, and then trains the BP neural network algorithm respectively by using the gene data in the cluster obtained, thereby effectively simplifying the structure of the large gene data and avoiding the learning complexity and overfitting of the BP neural network algorithm; in addition, noise data in the gene big data set can be effectively removed in the clustering process, so that the noise data are prevented from influencing the diagnosis precision of the disease diagnosis model obtained by training.
Drawings
The invention is further described by means of the attached drawings, but the embodiments in the attached drawings do not constitute any limitation to the invention, and for a person skilled in the art, without inventive effort, other drawings can be obtained according to the following figures.
FIG. 1 is a schematic diagram of the present invention.
Detailed Description
The invention is further described with reference to the following examples.
Referring to fig. 1, the diagnosis, prediction and major health management platform based on the cancer genome major data core algorithm of the embodiment includes a client, a data end and a major health management platform;
a user registers through a client, and after the registration is completed, gene data are sent to a big health management platform through the client;
the data terminal is used for periodically acquiring gene big data according to a preset acquisition cycle, transmitting the acquired gene big data to the big health management platform after the current acquisition cycle is finished, and updating the gene big data stored in the big health management platform;
the big health management platform comprises a gene database, a gene data preprocessing module, a disease diagnosis model and a disease diagnosis report generating module, wherein the gene database stores gene big data by adopting block link points, when new gene big data are stored in the gene database, the gene data preprocessing module is used for preprocessing the gene big data stored in the gene database, and a BP neural network is retrained by using the preprocessed gene big data, so that the disease diagnosis model based on the BP neural network is generated, when the big health management platform receives the gene data sent by a client, the disease diagnosis report generating module inputs the received gene data into the disease diagnosis model and sends the diagnosis result of the disease diagnosis model to the corresponding client.
The preferred embodiment learns the gene big data by using the BP neural network algorithm based on the gene big data, thereby establishing a disease diagnosis model to diagnose diseases and having higher precision of pathological change diagnosis.
Preferably, the gene data preprocessing module is used for clustering gene big data stored in a gene database and removing noise data in the gene big data in the clustering process.
Preferably, when the preprocessed gene big data are used for training the BP neural network, the gene data in each class obtained by clustering the gene data preprocessing module are used for respectively training the BP neural network, so that a disease diagnosis model based on the BP neural network is obtained.
Aiming at the characteristics of high dimensionality, large signal-to-noise ratio and strong correlation of gene big data, when the BP neural network algorithm is applied to the gene big data, training overfitting and high complexity are easily caused, and aiming at the defects, the method carries out clustering preprocessing on the gene big data, and then trains the BP neural network by utilizing the gene data in the cluster obtained by clustering, so that the structure of the gene big data can be effectively simplified, and the learning complexity and overfitting of the BP neural network algorithm are avoided; in addition, noise data in the gene big data set can be effectively removed in the clustering process, so that the noise data are prevented from influencing the diagnosis precision of the disease diagnosis model obtained by training.
Preferably, the gene data preprocessing module clusters the gene big data by adopting a density peak value clustering algorithm, and improves a calculation mode of local density of the data in the density peak value clustering algorithm, specifically:
let Y denote the Gene big data set, YiRepresents the ith gene data, N (Y), in the gene big data set Yi) Representing genetic datayiAnd N (y) isi)＝{yi，j||yi，j-yi|≤dc，j＝1，2，...，M(yi) In which d iscDenotes the truncation distance, yi，jA set of representations N (y)i) J (th) gene data of (1), M (y)i) A set of representations N (y)i) The amount of gene data in (a); definition m (y)i) Local density value representing gene data yi, then m (y)i) The values of (A) are:
Figure BDA0003026870220000051
wherein, ω (y)i) Representing gene data yiAnd ω (y) is the local density compensation coefficient ofi) The values of (A) are:
Figure BDA0003026870220000052
wherein, let yi，lA set of representations N (y)i) The first gene data in (1), and yi，l≠yi，j，N(yi，j) Representing gene data yi，jEta (y) of the neighborhood gene data seti，l，N(yi，j) ) represents gene data yi，lAnd set N (y)i，j) A judgment function therebetween, and
Figure BDA0003026870220000053
f(yi，j，yi) Representing gene data yi，jAnd gene data yiA judgment function in between, and
Figure BDA0003026870220000054
J(yi，j) Representing gene data yi，jIn the set N (y)i) A density coefficient of medium, and
Figure BDA0003026870220000055
J(yi) Representing gene data yiIn the set N (y)i) A density coefficient of medium, and
Figure BDA0003026870220000056
theta is a given positive integer to ensure that the denominator is not 0, and theta may take on a value of 0.001.
The preferred embodiment improves the way of calculating the local density of data in the conventional density peak clustering algorithm, which assigns a truncation distance d to the data set when calculating the local density of datacTo determine neighborhood data information of data, but in reality, when the sizes of classes in a data set are different, the same truncation distance d is usedcThe neighborhood data information of each data in the data set cannot be accurately described, so that the local density of the calculated data cannot accurately reflect the data density of the area where the data is located, and the accuracy of selecting the clustering center is affectedcThe local density value of each data in the data set cannot be accurately obtained, and in the local density compensation coefficient provided in the preferred embodiment, when the size of the class in which the data is located is larger, the truncation distance d is largercWhen the data in the same class are all data, the local density compensation coefficient of the data is close to 1, i.e. the density coefficient of the data is not changed, and when the class in which the data is located is smaller in size, the truncation distance d is smallercWhen larger, the value of the density compensation factor of the data will be larger, i.e. the local density value of the data is increased, thereby avoiding truncation distance dcThe influence of (a); to sum up, the calculation method of the local density of the data in the density peak clustering algorithm provided by the preferred embodiment can well solve the problem that the same truncation distance d is adopted when multiple size classes exist in the data setcThe neighborhood data information of the data can not be well reflected, thereby influencing the accuracy of selecting the clustering center.
Preferably, the gene data preprocessing module clusters the gene big data by adopting a density peak clustering algorithm, and corrects the local density value of each gene data in the gene big data set YLet m' (y)i) Representation of the Gene data yiThe corrected local density value is m' (y)i) The values of (A) are:
m′(yi)＝m(yi)ρ(yi)
Figure BDA0003026870220000061
in the formula, yi，kA set of representations N (y)i) The kth gene data in (1), and yi，k≠yi，ρ(yi) Representing gene data yiCorresponding noise penalty factor, m (y)i，k) Representing gene data yi，kLocal density value of s (y)i) Representing gene data yiThe area of (a) detects the coefficient, and
Figure BDA0003026870220000066
s (Y) is a region detection coefficient corresponding to the whole gene data in the gene big data set Y, and
Figure BDA0003026870220000062
wherein the content of the first and second substances,
Figure BDA0003026870220000063
means of region detection coefficients representing gene data in the gene big data set Y, and
Figure BDA0003026870220000064
σ (Y) represents the mean square error of the region detection coefficients of the gene data in the gene big data set Y, and
Figure BDA0003026870220000065
alpha represents a regulating parameter, the value of alpha can be 2.5, and M (Y) represents the gene data amount in the gene big data set Y;
and clustering the gene data in the gene big data set Y according to the local density of the corrected gene data by using a density peak value clustering algorithm.
In the preferred embodiment, in consideration of the fact that the local density of data at the edge of the class in the data set is smaller than the local density of data at the inside of the class, and when the noise data is closer to the edge of the class, the local density of the noise data calculated by using the truncation distance and the local density of the edge data closer thereto are closer, that is, in the process of clustering according to the local density of the data, the case where the edge data is mistaken for the noise data or the noise data is mistaken for the edge data is likely to occur, in order to further improve the accuracy of clustering and avoid the case where the noise data affects the accuracy of the clustering result, the preferred embodiment further corrects the local density value of the gene data in the gene big data set, and corrects the local density value of the gene data according to the region detection coefficient of the gene data and the region detection coefficient corresponding to the whole data in the gene big data set Y, when the gene data is edge data, the value of the region detection coefficient of the gene data is smaller, namely the value of the noise penalty coefficient is close to 1, namely the local density value of the gene data is not changed, and when the gene data is noise data, the value of the region detection coefficient of the gene data is larger, namely the value of the noise penalty coefficient is smaller, namely the local density value of the noise data is further reduced, so that the difference of the local density values between the intra-class data and the noise data in the gene big data set is enhanced, especially the difference of the local density values between the class edge data and the noise data is enhanced, and therefore when the density peak clustering algorithm carries out clustering according to the corrected local density value, the class data and the noise data can be better distinguished, the noise data is prevented from influencing the accuracy of the clustering result, namely the accuracy of the clustering result is improved.
Finally, it should be noted that the above embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the protection scope of the present invention, although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions can be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims (3)

1. The cancer genome big data core algorithm-based diagnosis, prediction and big health management platform is characterized by comprising a client, a data terminal and a big health management platform;
a user registers through a client, and after the registration is completed, gene data are sent to a big health management platform through the client;
the data end periodically acquires gene big data according to a preset acquisition period, transmits the acquired gene big data to the big health management platform after the current acquisition period is finished, and updates the gene big data stored in the big health management platform;
the big health management platform comprises a gene database, a gene data preprocessing module, a disease diagnosis model and a disease diagnosis report generating module, wherein the gene database stores gene big data by adopting block link points, when new gene big data are stored in the gene database, the gene data preprocessing module is used for preprocessing the gene big data stored in the gene database, and a BP neural network is retrained by using the preprocessed gene big data, so that the disease diagnosis model based on the BP neural network is generated, when the big health management platform receives the gene data sent by a client, the disease diagnosis report generating module inputs the received gene data into the disease diagnosis model and sends the diagnosis result of the disease diagnosis model to the corresponding client; the gene data preprocessing module is used for clustering gene big data stored in a gene database and removing noise data in the gene big data in the clustering process;
the gene data preprocessing module clusters gene big data by adopting a density peak value clustering algorithm, and improves the calculation mode of the local density of the data in the density peak value clustering algorithm, and specifically comprises the following steps:
let Y denote the Gene big data set, YiRepresents the ith gene data, N (Y), in the gene big data set Yi) Representing gene data yiAnd N (y) isi)＝{yi，j||yi，j-yi|≤dc，j＝1，2，...，M(yi) In which d iscThe distance of truncation is indicated by the distance of the truncation,yi，ja set of representations N (y)i) J (th) gene data of (1), M (y)i) A set of representations N (y)i) The amount of gene data in (a); definition m (y)i) Representing gene data yiLocal density value of m (y)i) The values of (A) are:
Figure FDA0003541516840000011
wherein, ω (y)i) Representing gene data yiAnd ω (y) is the local density compensation coefficient ofi) The values of (A) are:
Figure FDA0003541516840000021
wherein, let yi，lA set of representations N (y)i) The first gene data in (1), and yi，l≠yi，j，N(yi，j) Representing gene data yi，jEta (y) of the neighborhood gene data seti，l，N(yi，j) ) represents gene data yi，lAnd set N (y)i，j) A judgment function in between, and
Figure FDA0003541516840000022
f(yi，j，yi) Representing gene data yi，jAnd gene data yiA judgment function in between, and
Figure FDA0003541516840000023
J(yi，j) Representing gene data yi，jIn the set N (y)i) A density coefficient of medium, and
Figure FDA0003541516840000024
J(yi) Representing gene data yiIn the set N (y)i) A density coefficient of medium, and
Figure FDA0003541516840000025
theta is a given positive integer for ensuringThe denominator is not 0.
2. The cancer genome big data core algorithm-based diagnosis, prediction and big health management platform according to claim 1, wherein when the BP neural network is trained by using the preprocessed gene big data, the BP neural network is trained by using the gene data in each class obtained by clustering by the gene data preprocessing module, so as to obtain a disease diagnosis model based on the BP neural network.
3. The cancer genome big data core algorithm-based diagnosis, prognosis and big health management platform according to claim 2, wherein the genetic data preprocessing module clusters genetic big data using a density peak clustering algorithm.
CN202110416120.XA 2021-04-19 2021-04-19 Diagnosis, prediction and major health management platform based on cancer genome big data core algorithm Active CN113096828B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110416120.XA CN113096828B (en) 2021-04-19 2021-04-19 Diagnosis, prediction and major health management platform based on cancer genome big data core algorithm

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110416120.XA CN113096828B (en) 2021-04-19 2021-04-19 Diagnosis, prediction and major health management platform based on cancer genome big data core algorithm

Publications (2)

Publication Number Publication Date
CN113096828A CN113096828A (en) 2021-07-09
CN113096828B true CN113096828B (en) 2022-06-10

Family

ID=76678325

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110416120.XA Active CN113096828B (en) 2021-04-19 2021-04-19 Diagnosis, prediction and major health management platform based on cancer genome big data core algorithm

Country Status (1)

Country Link
CN (1) CN113096828B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107133496A (en) * 2017-05-19 2017-09-05 浙江工业大学 Gene expression characteristicses extracting method based on manifold learning Yu closed loop depth convolution dual network model
CN107992720A (en) * 2017-12-14 2018-05-04 浙江工业大学 Cancer based on coexpression network targets marker mapping method
CN108932402A (en) * 2018-06-27 2018-12-04 华中师范大学 A kind of protein complex recognizing method
CN110164508A (en) * 2018-02-13 2019-08-23 宏观基因有限公司 Biological data providing method, biological data storage method and the biological data Transmission system based on multi-tiling chain
CN112532630A (en) * 2020-11-30 2021-03-19 罗鑫龙 Gene big data disease prediction system based on 5G and block chain

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112597597A (en) * 2020-12-29 2021-04-02 浙江天行健智能科技有限公司 Road feel simulation method based on K-Medoids and BP neural network

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107133496A (en) * 2017-05-19 2017-09-05 浙江工业大学 Gene expression characteristicses extracting method based on manifold learning Yu closed loop depth convolution dual network model
CN107992720A (en) * 2017-12-14 2018-05-04 浙江工业大学 Cancer based on coexpression network targets marker mapping method
CN110164508A (en) * 2018-02-13 2019-08-23 宏观基因有限公司 Biological data providing method, biological data storage method and the biological data Transmission system based on multi-tiling chain
CN108932402A (en) * 2018-06-27 2018-12-04 华中师范大学 A kind of protein complex recognizing method
CN112532630A (en) * 2020-11-30 2021-03-19 罗鑫龙 Gene big data disease prediction system based on 5G and block chain

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
Robust Unsupervised Feature Selection;Mingjie Qian;《Proceeding of the Twenty-Third International Joint Conference on Artificial Intelligence》;20130803;第1621-1627页 *
结肠癌患者诊断的基因标志物识别算法;谢娟英;《模式识别与人工智能》;20171115;第30卷(第11期);第1019-1029页 *

Also Published As

Publication number Publication date
CN113096828A (en) 2021-07-09

Similar Documents

Publication Publication Date Title
CN112365464B (en) GAN-based medical image lesion area weak supervision positioning method
CN111161254A (en) Bone age prediction method
CN113693563B (en) Brain function network classification method based on hypergraph attention network
WO2015173435A1 (en) Method for predicting a phenotype from a genotype
CN111009321A (en) Application method of machine learning classification model in juvenile autism auxiliary diagnosis
CN113160974B (en) Mental disease biological type mining method based on hypergraph clustering
CN115841607A (en) Brain network structure and similarity joint learning method based on graph attention network
CN110348516B (en) Data processing method, data processing device, storage medium and electronic equipment
CN113610118A (en) Fundus image classification method, device, equipment and medium based on multitask course learning
Chen et al. Classification of breast tumors in ultrasound using biclustering mining and neural network
Wang et al. Extended ResNet and label feature vector based chromosome classification
CN109740669B (en) Breast cancer pathological image classification method based on depth feature aggregation
Wu et al. Evolutionary feature construction for ultrasound image processing and its application to automatic liver disease diagnosis
Chen et al. An approach based on biclustering and neural network for classification of lesions in breast ultrasound
CN111583194B (en) High-dimensional feature selection algorithm based on Bayesian rough set and cuckoo algorithm
CN117195027A (en) Cluster weighted clustering integration method based on member selection
CN113096828B (en) Diagnosis, prediction and major health management platform based on cancer genome big data core algorithm
CN116956138A (en) Image gene fusion classification method based on multi-mode learning
CN115036034B (en) Similar patient identification method and system based on patient characterization map
CN109886320B (en) Human femoral X-ray intelligent recognition method and system
CN117036288A (en) Tumor subtype diagnosis method for full-slice pathological image
CN116228759A (en) Computer-aided diagnosis system and apparatus for renal cell carcinoma type
CN116884067A (en) Micro-expression recognition method based on improved implicit semantic data enhancement
CN116664932A (en) Colorectal cancer pathological tissue image classification method based on active learning
CN116797817A (en) Autism disease prediction technology based on self-supervision graph convolution model

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
TA01 Transfer of patent application right

Effective date of registration: 20220519

Address after: 050035 18 / F, block a, interstellar center, No. 155, Changjiang Avenue, Shijiazhuang high tech Zone, Shijiazhuang, Hebei Province

Applicant after: Xikang Software Co.,Ltd.

Address before: 578000 administration center building, No.8 Yantian Road, xinyingwan District, Yangpu Economic Development Zone, Hainan Province

Applicant before: Meili medical technology (Yangpu) Co.,Ltd.

TA01 Transfer of patent application right
GR01 Patent grant
GR01 Patent grant