Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Construction method and device of benign and malignant tumor identification model

A technology for constructing methods and identifying models, applied in the field of gene detection, can solve the problems of difficulty in taking into account other cancer types, building a benign and malignant tumor discrimination model, and taking into account both broad spectrum and accuracy, and achieves broad-spectrum and specificity. sexual effect

Active Publication Date: 2020-06-12
BEIJING GENEPLUS TECH
View PDF6 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, since the NGS panel needs to design probes according to the characteristics of different cancers to capture the unique mutations or copy number variations of different cancers, it is difficult to balance the broad spectrum and accuracy
As disclosed in Chinese patent document CN109971848A, a grading model for detecting the benign and malignant degree of esophageal tumors and / or gastric tumors and its application, the above technical scheme can only be used for detecting esophageal tumors and / or gastric tumors, and it is difficult to take into account other cancer type
However, there are no related reports on the construction of benign and malignant tumor identification models based on next-generation sequencing technology using T cell immune repertoire information infiltrated into tumor tissue

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Construction method and device of benign and malignant tumor identification model
  • Construction method and device of benign and malignant tumor identification model
  • Construction method and device of benign and malignant tumor identification model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0068] Example 1 Construction of tumor benign and malignant differentiation model

[0069] This embodiment provides a method for constructing a model for distinguishing benign from malignant tumors, the flow chart is as follows figure 1 shown, including the following steps:

[0070] (1) Take 20 known benign tumor samples and 65 known malignant tumor samples in Table 1 as the training set; for the above 85 known benign and malignant tumor tissue samples (peripheral blood of tumor samples can also be used, in this In the embodiment, the genomic DNA of the tumor tissue is selected to be amplified by multiplex PCR technology or 5' RACE technology. In this embodiment, the multiplex PCR technology is selected, and the primers used are from the QIAGEN Multiplex PCR Kit, including 32 pairs of forward V Gene primers and 13 pairs of reverse J gene primers, the amplified products were further amplified with Illumina universal primers for the second round to generate a library with a tar...

Embodiment 2

[0097] Example 2 Device for distinguishing benign and malignant tumors

[0098] This embodiment provides a device for distinguishing benign and malignant tumors, the device structure diagram is as follows Figure 4 shown, including:

[0099] The TCR clone identification unit is used to identify the TCR clone type and CDR3 segment of the training set sample;

[0100] CDR3 statistical unit, used to count and calculate the frequency of occurrence of the CDR3 segment;

[0101] Kmer interrupt unit, for recoding the CDR3 segment by Kmer interrupt;

[0102] A data dimensionality reduction unit, used for data dimensionality reduction of the Kmer frequency data after CDR3 recoding;

[0103] The machine learning model training unit is used to associate the benign and malignant information of the known tumor samples in the training set with the Kmer frequency data after data dimensionality reduction, and use the machine learning algorithm to train the model to obtain a benign and mali...

experiment example 1

[0119] In this experimental example, the classification model in machine learning is used to perform model training and performance prediction on the Kmer frequency data after dimensionality reduction processing. Specifically, the original data is divided into 5 parts by stratified sampling by means of 50-fold cross-validation. While keeping the ratio of the two types of data for each fold and the original data similar, the model is trained with 4 fold data each time, and the remaining 1 fold data is used to verify the predictive performance of the model. Construction of Example 1 model is verified. The specific implementation is as follows:

[0120] Implement according to Example 1, get 20 known benign tumor samples in Table 1, 65 known malignant tumor samples, as a training set, process the samples according to steps (1)-(6) in Example 1, and The Kmer frequency data obtained after dimensionality reduction adopts the method of 5-fold cross-validation to divide the original d...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides a construction method and device of benign and malignant tumor identification model, and the method comprises the steps: taking a plurality of known benign tumor samples and malignant tumor samples as a training set, and obtaining TCR clone types and CDR3 segments of the samples in the training set; counting and calculating the occurrence frequency of the CDR3 section, and then recoding the CDR3 section through Kmer interruption; carrying out data dimension reduction processing on the Kmer frequency data obtained after CDR3 recoding; associating benign and malignant information of a known tumor sample in the training set with the Kmer data after data dimension reduction, and performing model training by using a machine learning algorithm to obtain a benign and malignant tumor identification model. The benign and malignant tumor identification model constructed by the construction method is used for carrying out benign and malignant tumor identification on unknowntumor samples, can identify benign and malignant tumor samples of different types, and meets the requirements of broad spectrum and specificity.

Description

technical field [0001] The invention relates to the field of gene detection, in particular to a method and device for constructing a model for distinguishing benign and malignant tumors. Background technique [0002] Identification of benign and malignant tumors is an effective means to reduce cancer overdiagnosis and overtreatment. Traditional screening methods for benign and malignant tumors, such as imaging examination and biopsy, have certain sensitivity and applicability in clinical practice. However, there are still a considerable number of tumor cases where accurate results cannot be judged. At the same time, due to the influence of various aspects such as the experience of the observer and the performance of the equipment, the final judgment of benign and malignant tumors will be affected. [0003] With the development of molecular biology, molecular diagnostic techniques have gradually attracted people's attention in the identification of benign and malignant tumor...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G16H50/50
CPCG16H50/50
Inventor 黄毅易鑫杨玲张超何欣诺王长希
Owner BEIJING GENEPLUS TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com