Tumor prediction method
A prediction method and tumor technology, applied in the field of tumor diagnosis, can solve problems such as lack of good tumor prediction methods, large malignant tumors, etc., achieve accurate tumor prediction, meet the needs of use, and design reasonable effects
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Examples
Embodiment 1
[0020] A tumor prediction method, the specific steps are as follows:
[0021] Step 1, collect historical tumor data from various types of malignant tumor patients, and establish a tumor database;
[0022] Step 2: According to the tumor database, the proportional hazard function used to describe the relationship between the patient and the probability of death is established through Kalman filter, finite element and neural network algorithms, and different impact factors are set for different historical tumor data. Various impact factors The sum is 1;
[0023] Step 3: Obtain the gene expression data, living habits and CT images of the subject, convert the data of the subject, and obtain the converted new test sample data;
[0024] Step 4, filling the new converted test sample with missing values to obtain test sample data without missing values;
[0025] Step 5, substituting the test sample data without missing values into the proportional hazard function to obtain the pr...
Embodiment 2
[0027] A tumor prediction method, the specific steps are as follows:
[0028] Step 1. Collect historical tumor data for various types of malignant tumor patients. Historical tumor data include the patient’s tumor gene expression data, living habits, tumor location, CT images, biopsy pathological examination data, and the patient’s history from onset. Time to death, survival or death data, to establish a tumor database;
[0029] Step 2: According to the tumor database, the proportional hazard function used to describe the relationship between the patient and the probability of death is established through Kalman filter, finite element and neural network algorithms, and different impact factors are set for different historical tumor data. Various impact factors The sum is 1;
[0030] Step 3, obtain the gene expression data, living habits and CT images of the subject, map the data of the subject to a matrix form, and then use Laplace transform to obtain the new test sample data;...
Embodiment 3
[0034] A tumor prediction method, the specific steps are as follows:
[0035] Step 1: Collect historical tumor data for various types of malignant tumor patients, delete obviously special data and establish a tumor database;
[0036] Step 2: According to the tumor database, the proportional hazard function used to describe the relationship between the patient and the probability of death is established through Kalman filter, finite element and neural network algorithms, and different impact factors are set for different historical tumor data. Various impact factors The sum is 1;
[0037] Step 3: Obtain the gene expression data, living habits and CT images of the subject, convert the data of the subject, and obtain the converted new test sample data;
[0038] Step 4, fill the missing value of the converted new detection sample through the pre-defined cosine similarity function, and obtain the detection sample data without missing value;
[0039] Step 5, substituting the test ...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More