Classification recognition method for feature points in blood vessel image on the basis of transfer learning

Abstract

The invention discloses a classification recognition method for feature points in blood vessel image on the basis of transfer learning. The method comprises two parts, i.e., blood vessel image simulated training and blood vessel image type recognition. In the blood vessel image simulated training, a feature point dataset is manufactured, and a feature point dataset is used for training a deep learning model to obtain a classification model of the feature points in the blood vessel image on the basis of transfer learning. In the blood vessel image type recognition, the feature points of the blood vessel image are extracted, the extracted feature points are input into the blood vessel image simulated training to obtain the classification model of the feature points in the blood vessel imageon the basis of the transfer learning, and the type of the feature point in the blood vessel image is obtained. By use of the method, the classification model can be used for quickly and accurately identifying whether the feature point in the blood vessel image is a branching point or a cross point, so that the problems of complex steps and low accuracy and efficiency in an existing method are solved, and the method has a great assistance function for the clinical medicine.

Description

technical field [0001] The invention relates to the technical field of transfer learning and image processing, in particular to a method for classifying and identifying feature points in blood vessel images based on transfer learning. Background technique [0002] Medical image processing is a new discipline and technology that has developed rapidly with the development and maturity of computer technology and the advancement of clinical diagnostic technology. Nowadays, medical image processing technology is more and more widely used in clinical practice. At the same time, due to increasing global wealth and aging, all systemic diseases affecting the vascular network are becoming more common such as age-related macular degeneration, diabetic retinopathy, glaucoma, hypertension, arteriosclerosis and multiple sclerosis Wait. While fundus images are often used to diagnose these pathologies, the geometric properties of the vascular system at bifurcations and intersections, such ...

AI Technical Summary

Problems solved by technology

These methods are not only sensitive to noise and slight fluctuations in blood vessel width, but also often detect too many pseudo-nodes at the same node position, and when the intersection angle is obtuse, that is, when the intersection of two blood vessels overlaps a lot, it is easy to make a cross Points are judged as two bifurcation points. Therefore, the past methods for identifying and classifying feature points have the problems of complicated steps, low accuracy and low efficiency.

Images