Method for building optimal hyperplane, dynamic optimizing system and building device

Abstract

The invention relates to the field of cancer detection, in particular to a method for building an optimal hyperplane. The method comprises the following steps of selecting a plurality of normal person and cancer patients as samples, and obtaining protein information of all the samples; utilizing the protein information to establish a protein information feature data space of all the samples; according to Tanimoto distance, building the optimal hyperplane, wherein the optimal hyperplane is used as a classifier for identifying the normal persons and the cancer patients. The invention also provides a dynamic correction system for early cancer detection. The method has the advantages that the optimal hyperplane is built according to the protein information of the samples, and is used as the classifier to identify the normal persons and the cancer patients; the protein information and the protein information feature data space are classified, and the Minkowski distance metric and slack variable are introduced into the protein information feature data space, so as to improve the building of the optimal hyperplane; finally, the combined component classifier is dynamically designed, so as to realize the intelligence and accuracy optimizing of the system.

Description

technical field [0001] The invention relates to the field of cancer detection, in particular to a method for constructing an optimal hyperplane, a dynamic optimization system and a construction device. Background technique [0002] Tumor marker detection has become one of the routine tumor detection methods, and plays an important role in the screening, diagnosis and prognosis of tumors and the realization of individualized treatment. At present, the commonly used tumor marker detection methods include enzyme-linked immunosorbent assay (ELISA), gene chip, protein chip, mRNA transcription detection, and tumor cell metabolite (marker) detection. [0003] Enzyme-linked immunosorbent assay (ELISA) is one of the most commonly used methods among the different immunoassays. It refers to the adsorption of soluble antigens or antibodies to polystyrene and other solid phase carriers. The basic principles of the ELISA method are: ① Make the antigen or antibody bind to the surface of ...

AI Technical Summary

Problems solved by technology

[0017] 1) High cost of testing: In order to ensure the accuracy of tumor diagnosis, it is often necessary to increase the types of tumor markers. However, conventional testing can only detect one tumor marker, and the cost of multiple testing materials and labor costs also varies. to improve

[0018] 2) Long detection time: routine detection requires professional inspectors to go through a series of sampling, testing, and analysis to conclude that the detection technology requires high professionalism and cannot be quickly implemented in a large number of hospitals (clinics) with limited conditions. Laboratory testing is difficult to guarantee the accuracy of tumor diagnosis

[0019] 3) The amount of samples required for detection is large: conventional tumor marker detection is mostly used to assist doctors in evaluating the results of cancer patients' treatment, and taking a large number of blood samples from patients is not conducive to postoperative treatment and affects their physical health. timely recovery

[0020] 4) The accuracy of the diagnostic results is low: the current commonly used tumor marker detection methods have certain false positives, and there are errors in the analysis results. The lack of a unified and sensitive analysis system makes it difficult to guarantee the accuracy of tumor diagnosis

Images