Disease diagnosis method on the basis of medical image data augmentation

Abstract

The invention discloses a disease diagnosis method on the basis of medical image data augmentation and belongs to the technical field of disease diagnosis methods. The method includes: 1) mapping original image data to a deep characteristic space and extracting the characteristic of highly-linearized semantic information; 2) according to the distribution of the data, corresponding to different types in the medical image, in the characteristic space, acquiring a characteristic covariance matrix for data augmentation; 3) calculating loss function of the data augmentation and continuously optimizing the loss to acquire a model with a higher characteristic extraction capability. The data augmentation method only works in model training, so that there is no more calculation load and calculationtime introduced during the prediction of the medical image data by using the model. By means of the method, fixed medical image data is effectively augmented, so that the demand on the quantity of labeled patient data during training of a deep neural network. The method effectively relieves the problems that the medical image data is difficult to acquire and is high in labeling cost and further increases disease diagnosis accuracy.

Description

technical field [0001] The invention relates to a disease diagnosis method based on medical image data amplification, and belongs to the technical field of disease diagnosis methods. Background technique [0002] Medical imaging refers to the image data of internal tissues and organs of the human body obtained by X-ray projection and other methods. It can provide a large amount of intuitive medical information and is increasingly widely used in modern clinical diagnosis and medical treatment. With the advent of the era of artificial intelligence, computer vision-oriented deep learning technology, as a fast, accurate, and effective image analysis method, is widely used in the automatic analysis of medical images, which can effectively reduce the lack of subjective experience and awareness of doctors. Misdiagnosis and other medical risks brought about by factors such as knowledge fluctuations. However, the deployment of deep learning algorithms requires sufficient labeled dat...

AI Technical Summary

Problems solved by technology

Existing methods mainly start from two aspects: (1) Using the augmentation method of natural pictures, simple geometric transformations (translation, scaling, etc.) and image property transformations (brightness, contrast, etc.) More training samples, but considering the particularity of medical images, these methods have the risk of losing the lesion information contained in the image, and the actual effect is limited; (2) training deep generation network or autoencoder, such as Wasserstein generation Adversarial network (WGAN), variational autoencoder (VAE), etc., by introducing random variables and repeated reasoning, similar medical images are randomly generated, but the implementation process of this method is very complicated, and there is only a small improvement in effect compared with the former

Images