Coronary artery plaque data detection method and system, storage medium and terminal

Abstract

The invention belongs to the technical field of medical image processing and computer-aided diagnosis, and discloses a coronary plaque data detection method and system, a storage medium and a terminal. The method comprises steps: on a three-dimensional medical image, extracting a three-dimensional image block along a coronary center line point to serve as a sample, and dividing a sample data set;constructing a three-dimensional stepped convolutional neural network by taking the coronary artery three-dimensional image block sample as input and whether the image contains plaques or not as output; sending the training data set into a network, and calculating a loss function to train a network model; and predicting the trained network model on a test set to complete a coronary plaque detection task. Semi-supervised learning is performed in combination with unlabeled data, plaque detection is completed only depending on a small amount of labeled data, the labeling data difficulty is reduced, and the detection precision is improved. The coronary plaque detection method realizes coronary plaque detection, and has the characteristics of no need of blood vessel pre-segmentation, high accuracy and dependence on a small amount of annotation data.

Description

technical field [0001] The invention belongs to the technical fields of medical image processing and computer-aided diagnosis, and in particular relates to a coronary plaque data detection method, system, storage medium and terminal. Background technique [0002] At present, coronary artery disease is one of the most common cardiovascular diseases, and the generation of coronary atherosclerotic plaque is mainly responsible for coronary artery disease. Early detection and identification of coronary artery plaques is of great significance for the prevention and diagnosis of coronary artery disease. Imaging techniques such as intravascular ultrasound and optical coherence tomography provide detailed visualization of intraluminal as well as plaque morphology and allow reliable quantification of atherosclerotic burden and its components, with good discrimination of plaque properties , but these techniques are invasive and costly, time-consuming and difficult to perform, present ...

AI Technical Summary

Problems solved by technology

[0004] (1) The existing threshold method has the problem of intensity overlap, and the detection is not accurate

[0005] (2) Existing methods relying on coronary artery segmentation are difficult to generate fine coronary artery trees due to the influence of calcification and excessively thin distal vessels

[0006] (3) Manual design of features by traditional machine learning methods is time-consuming and labor-intensive

[0007] (4) Existing deep model training requires a large amount of labeled data

[0008] The difficulty in solving the above problems and defects is as follows: in view of the diversity of data features, it is difficult to manually design distinguishing features in the application of traditional machine learning methods; deep learning methods have certain difficulties in obtaining labeled data, and manual labeling by experts is time-consuming and labor-intensive. prone to inter-observer error

Images