An automatic detection and tracking method of lung tumors based on global optical flow

Abstract

An automatic detection and tracking method for lung tumors based on global optical flow method comprises the following steps: 1) preprocessing X-ray shooting video, processing image including enhancement, etc. to improve pathological discrimination of lung; 2) setting a preparation time for that doctor to manually align the lungs of the patient first, and star global optical flow detection after the alignment; 3) when that detection of the optical flow method is star, the continuous images collect by the X-ray photographing equipment are estimated, and the optical flow values of each moving object are compare; 4) frame selection is carried out for the local flow field with the largest optical flow value, and the tumor frame is selected by a polygon frame with variable size and continuous approximation, and is used as a tracking target in the whole operation process; 5) tracking the selected target, namely tumor, by optical flow estimation; 6) tracking the histogram distribution of thetarget area by using the mean shift as the tracker of the target motion. The invention has strong robustness, low error and high sensitivity.

Description

technical field [0001] The invention relates to the fields of image processing and medical imaging. Specifically, it refers to an automatic detection and tracking method for lung tumors based on the global optical flow method. Background technique [0002] Statistics show that lung cancer is currently the number one cause of cancer death in the world, and with the development of medical technology, tumor ablation, which has the advantages of small wounds, less pain, and less bleeding, has become one of the important ways to cure lung diseases. Minimally invasive is very important in palliative surgery even in the middle and advanced stages of cancer. The traditional minimally invasive needle insertion method requires three-dimensional CT mapping, with the ribs, intercostal spaces, and body surface marker lines as reference points, and the needle is inserted under the guidance of imaging. Because CT cannot generate images in real time, and reference points are selected manu...

AI Technical Summary

Problems solved by technology

The threshold for distinguishing the degree of motion can be automatically adjusted according to different actual situations, but it will always be limited within a certain range