Intravascular ultrasound image processing method based on deep learning

Abstract

The invention belongs to the technical field of image processing, and discloses an intravascular ultrasonic image processing method based on deep learning, which comprises the following steps of: performing data enhancement processing on an collected ultrasonic image, and inputting the ultrasonic image subjected to the data enhancement processing into an intravascular plaque automatic segmentation model based on deep learning, and inputting the plaque image segmented by the automatic segmentation model into the trained convolutional neural network model to obtain an recognition result corresponding to the plaque image. According to the method, the display definition of the ultrasonic image can be effectively improved by performing data enhancement processing on the collected ultrasonic image, the redundant part of the image is reduced, the size of the image is simplified, the intravascular plaque image can be automatically segmented accurately and quickly through the intravascular plaque automatic segmentation model based on deep learning, and the accuracy of the intravascular plaque segmentation is improved. The method facilitates the subsequent automatic recognition through the convolutional neural network model, and achieves the quick, objective and accurate automatic recognition and evaluation analysis of the vascular lesion.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to an intravascular ultrasound image processing method based on deep learning. Background technique [0002] At present: intravascular ultrasound (IVUS) is the most direct and accurate method for diagnosing coronary heart disease at present. According to the differences in the acoustic characteristics of different vascular tissue components, different ultrasonic echo signals are obtained, so as to obtain a real-time tomographic image of the two-dimensional cross-section of the blood vessel including the blood vessel wall structure and plaque tissue, and realize the detection of blood vessels. Quantitative assessment of mural and intimal atherosclerotic lesions. However, due to the low contrast of ultrasound images, which are greatly affected by noise and various artifacts, it is still difficult to distinguish the tissue characteristics of plaques. In addition,...

AI Technical Summary

Problems solved by technology

In the medical field, especially in pathological and radiological film reading, surgical robots, etc., intelligent diagnosis functions have begun to take shape, but there are many limiting factors in the field of ultrasound: due to the lack of uniform standards for the collection of ultrasound images, image quality Acquisition is restricted by the doctor's operation, and different angles and different acquisition positions have a great influence on the results. At the same time, different tissue structures in ultrasound images vary greatly, and the transmission of dynamic images during acquisition has higher requirements for the recognition and diagnosis of artificial intelligence images. AI development for ultrasound work is still in its infancy

[0003] Through the above analysis, the problems and defects of the existing technology are: the contrast of the existing intravascular ultrasound image is low, it is greatly affected by noise and various artifacts, it is difficult to distinguish the tissue characteristics of the plaque; the resolution is low, Cause different readers to draw different conclusions

Images