Puncture guide system for vein and artery recognition based on neural network

Abstract

The invention provides a puncture guide system for vein and artery recognition based on a neural network. The puncture guide system comprises an ultrasonic image input unit, a neural network model, anultrasonic image generation unit, an ultrasonic image display module and a puncture guide unit, wherein the ultrasonic image input unit is used for inputting an ultrasonic image and inputting a to-be-recognized ultrasonic image into the neural network model for processing; the neutral network model is used for acquiring position information of veins and arteries from the to-be-recognized ultrasonic image through the neutral network model; the ultrasonic image generation unit distinctively marks the veins and the arteries according to the acquired position information, and generates an ultrasonic image containing vein markers and artery markers; the ultrasonic image display module is used for displaying a relative position image of a puncture needle with the vein markers and the artery markers as well as puncture parameters in real time; and the puncture guide unit guides the puncture needle to perform puncture according to the relative position image and the puncture parameters. The puncture guide system can automatically recognize the veins and the arteries in the to-be-recognized ultrasonic image so as to assist a doctor in venipuncture.

Description

technical field [0001] The invention relates to the technical field of medical ultrasound, in particular to a neural network-based vein and artery recognition method and system. Background technique [0002] In clinical medicine, many endovascular operations require catheter placement and puncture on veins, such as in the right jugular vein, such as central venous catheter placement, hemodynamic measurements, myocardial biopsy, and cardiac ablation. During jugular venipuncture, doctors rely on vision, touch, and experience to manually judge the location of blood vessels, such as detecting the pressure in the blood vessels and observing the color of blood to identify the nature of the blood vessels. These methods for judging the jugular vein based on blood vessel pressure and blood color have poor reliability. If the puncture needle enters the jugular vein, follow-up operations such as dilation and catheterization will continue; if the puncture needle is located in the caroti...

AI Technical Summary

Problems solved by technology

However, at present, the identification of arteries and veins in ultrasound images is still almost done manually, especially the identification of carotid arteries and jugular veins, which requires the operator to have professional knowledge of ultrasound imaging, which also causes the uncertainty of the identification results, which cannot be minimized. The risk of the surgery; at the same time, it also limits the promotion and popularization of ultrasound-guided jugular vein puncture

Images