Active learning-based image detection method for liver injury type

Abstract

The invention discloses an active learning-based image detection method for a liver injury type. The method comprises the specific steps of 1, locating and extracting a liver target region from an abdominal medical image set by using a liver active shape model; 2, defining and extracting direction mode eigenvectors and grayscale range change mode eigenvectors of pixel points in the liver target region; 3, performing liver pixel density clustering center definition and clustering division based on a two-dimensional space of local density of the pixel points in the liver target region and a minimum distance from one pixel point to another pixel point with the higher density; and 4, performing image detection of the liver injury type based on an active learning method of liver pixel density clustering edge, clustering center and clustering division.

Description

technical field [0001] The invention belongs to the field of computer image processing, in particular to an image detection method of active learning liver damage category. Background technique [0002] With the rapid development of the transportation industry, abdominal injuries caused by traffic accidents are on the rise. In particular, the liver is prone to violent injuries due to its poor elasticity. In addition, there are many pathological injuries in clinical practice, such as liver damage caused by viruses such as hepatitis A and B, or some drugs and chemical substances, or liver congestion caused by liver tumors, cardiac insufficiency, or liver damage caused by certain congenital liver diseases . Violent or pathological injuries will affect the liver, and may involve more organs, causing greater harm to the human body and seriously threatening the life safety and health of patients. The detection of liver injury provides an important basis for the formulation and s...

AI Technical Summary

Problems solved by technology

[0004] (1) Violent liver injury is often accompanied by interference such as hemoperitoneum, effusion, gastrointestinal contents, and gas, and the patient's position cannot cooperate. These interference signals increase the difficulty of liver injury detection

However, in pathological liver injury, the clinical symptoms and signs of patients are usually not obvious enough, which leads to challenges in damage detection

[0005] (2) The artificial detection method has a low information utilization rate (for example, CT has 4096 levels of grayscale, while the grayscale resolution of the human eye is only about 40 levels), and manual reading of a large number of pictures for a long time is prone to fatigue, resulting in a decrease in accuracy And other issues

[0006] (3) Liver damage is diverse, manual detection is limited by personal experience, and there are errors or omissions in detection

[0010] (2) Most methods require the number of damage categories as prior knowledge, and can only classify liver damage into two types: with or without damage, malignant or benign damage, and low-density or high-density damage

In fact, there are multiple possibilities for violent or pathological liver damage, and the number of fixed damage categories limits the scope of automatic liver damage detection.

Images