Computer generated error model control method

Abstract

The invention relates to a computer-generated error model control method. The method comprises the following steps: positioning a narrow position of a blood vessel part through a target-driven first dynamic delay algorithm; optimizing a first dynamic delay algorithm, and dynamically using the first regional convolutional neural network model and the retina network model; the first region convolutional neural network model firstly generates a candidate region of the target, then performs feature extraction on the candidate region of the target, generates a candidate region K of the target after feature extraction, classifies the candidate region K of the target after feature extraction by using a vector machine model, and judges the position of the candidate region K of the target after feature extraction by means of a linear regression model; and the efficiency is improved.

Description

technical field [0001] The invention relates to a block chain automatic testing method, which is a computer-generated error model control method. Background technique [0002] During implementation, because the object is medical data, its positive and negative samples are prone to imbalance. In most cases, the positive sample is much smaller than the negative sample, that is, it is difficult for the present invention to obtain healthy data samples when collecting data, and most of the people who undergo coronary angiography are coronary heart disease patients, which leads to It is difficult to collect coronary angiographic samples from healthy people. Therefore, the realization background of the present invention is that positive and negative samples are unbalanced, and secondly, the stenosis area shown by coronary artery stenosis angiography usually only occupies a very small part of the developed image, and most of the areas are patient's sternum developing and black back...

AI Technical Summary

Problems solved by technology

[0002] During implementation, because the object is medical data, its positive and negative samples are prone to imbalance

In most cases, the positive sample is much smaller than the negative sample, that is, it is difficult for the present invention to obtain healthy data samples when collecting data, and most of the people who undergo coronary angiography are coronary heart disease patients, which leads to Difficult to collect coronary angiography samples from healthy population